The Big Search Handbook

iabuk.net

Search, Display and Social for Ambitious Brands

To discuss a project, get in touchjames.cragg@tugagency.com

0207 033 6933

www.tugagency.com

We combine hardcore analytics and creative thinking

to find sharper audience insights and new angles

on your market. Our hands on approach to campaign

optimisation drives more from our technology stack,

and leads to demonstrably better performance across

markets, languages and countries.

Digital Media AgencyLondon | Toronto | Sydney

Contents

Foreword 02 1 Why search is never JUST search: The 5 big questions of the CMO/CEO search agenda 06

2 Search at the heart of modern marketing 12 3 Brand Vs DR 18 4 Mobile: The migration from tablet to mobile and dissecting the idea that search on desktop is dying 24 5 Five reasons why search advertisers need to consider responsive web design for their site 30 6 Audience based search marketing 34 7 Analytics 38 8 Drive more relevant and cost effective traffic with ad extensions 42 9 The evolution of attribution 48 10 Showrooming 54 11 Data protection and privacy 60 12 International search engines 64 13 Audience planning in search: enhance keywords with audience signals 70

14 Jargon Buster 76

Dan Thwaites, CSO, Tug

Searching for new angles to grow your business? Then this book is for you.

Foreword

5

One of the biggest risks in business is that we become removed from the people that we want to love our brand and buy our products.

It’s all too easy to do, with the many trials, tribulations and distractions of our daily working lives. But if you ever fear that this might be the case, then search is the perfect antidote.

Search is one of the very few channels where you can - and must - interact, at scale, with your market every single day. I wish I’d discovered it sooner in my career.

Tug is an award-winning, full service digital media agency. Working with ambitious brands, targeting digital audiences around the world through search, display and social.

To discuss a project, get in touch

james.cragg@tugagency.com

0207 033 6933

www.tugagency.com

6

If you love finding out new things and new angles on your customers.

If you love ingenuity and creativity.

If you love fast feedback or simply have a low patience threshold.

If you love to prove a point. And are brave enough to admit and learn from your mistakes.

Then you’re probably already working with search in some way, or you should be.

It’s the perfect mixer, adding texture and scale to other data.

Clients at the top of their game are using it to inform which markets they enter, when and how.

It can make other channels work better, help prove which ones are working and point out those that might not be working as well as you thought.

It’s fast as hell and responds beautifully to creativity. In fact it outright demands daily ingenuity if you want to compete with it.

You’ll know when you’re right. But you’ll very quickly know when you’re wrong, too.

If you can treat those two imposters just the same, then you’re reading the right book.

Whether you’re new to search, or an old hand at it, there’s always something new to discover.

Keep searching.

Dan Thwaites CSO, Tug

7

Why search is never JUST search: The 5 big questions of the CMO/CEO search agenda

1

By Ollie Vaughan, Head of Biddable Media, Tug

Approximately 2.5 million searches are made on Google every minute, 24/7, 365 days a year. Global search spend is also currently growing at approximately 13% per year, so it’s no surprise that vast quantities of articles have been written on best practice paid search techniques.

But we think there’s a problem.

And it’s this: paid search is an extremely sophisticated specialism. As a result of this complexity, it’s often set up, managed and optimised in a very autonomous fashion.

But if brands are to reap the full benefits from this dynamic discipline, they need to ensure search advertising is connected to the business as a whole, rather than as a separate (albeit efficient) unit.

So the CMO/ CEO challenge is how to fit search into the bigger picture, without necessarily having to master the ever changing operational details.

What follows is our guide for brand leaders to ensure that their ENTIRE business is making the most out of their investment in search.

1

10

5 big questions for the CEO/CMO to ask about search

1. What insight is it bringing back to the wider business?Paid search generates some of the best quality, real time data available to an advertiser. It is also, in many ways, the closest a brand can get to listening and talking to their customers every minute of every single day.

For example, analysing keyword data can tell an advertiser what products they are most competitive for or can even highlight problems their users may be having with their products.

Likewise, analysing ad copy test results can point out which aspects of their brand and products are (or indeed are not) most attractive to their audience.

But if you’re not actively looking for these insights, you could miss what’s right under your nose.

And on a higher level, paid search data can also be merged with broader data sets to give insights for the business as a whole.

As an example, when asked to investigate potential new markets for a global SaaS (Software as a Service) business, we used keyword search volumes data as a starting point for examining demand in potential markets. We combined this with programmatic display data and key commercial data to successfully identify and prioritise countries and select the best products for each.

2. How is search using other channels’ insights & data sources?There are a wealth of opportunities to connect search up with other channels and data sources to make it work even better.

For example, plugging in social data can highlight signals of how your audience are talking about your product in real time. This can rapidly provide new angles for keyword creation as well as inspiring new areas for ad copy testing.

But brands can move faster still.

Marketers can now plug into other third party services to automate changes to their search activity - plugging into API’s using Google Scripts. For example, if demand for a given product is dependent on the weather, Google can now connect to the OpenWeather API and automatically change bids based on the weather conditions.

Also, many search advertisers are still not always using their most basic 1st party data to its full potential. Brands should adapt their search activity based on a specific visitor’s digital interactions with their business. By setting up retargeting pixels on their website, they can use RLSA (retargeting lists for search ads) to adapt their search activity - perhaps increasing /decreasing bids according to how they have been using the site.

Why search is never JUST search: The 5 big questions of the CMO/CEO search agenda

11

1

3. How are other channels using search data and insights?Brands should be adapting their behaviour in other channels according to what they learn in search.

For example, when a specific visitor has engaged with them through a paid search ad, brands should be looking at how they can continue the interactions across other channels. Technology such as DoubleClick’s ‘display retargeting from search’, enables marketers to create segments of users who have clicked on their paid search ads and retarget them via programmatic display.

Ad copy testing can be one of the quickest and cheapest ways to test which aspects of the brand, product or ‘calls to action’ work best for the business. Results should be shared to ensure other channels (incl. Display, Radio & TV) get a head start in their creative testing.

And of course, one of the easiest ways to measure whether branding activity from other channels is having an impact, is to monitor and compare the volume of Brand searches.

4. How to keep it all connected?It’s one thing knowing that your search activity and data should be working with your other channels, but it is another thing making it happen. As with many aspects of management, this comes down to people, technology and culture.

Encouraging active cross-channel collaboration is critical and can be a valuable competitive edge. Too often teams focus solely on their specific channels metrics (e.g. increasing click through rates) and can easily, although unintentionally, lose sight of the OVERALL business intention (e.g. building their brand, increasing sales) and how their channel can support these - even if they are not directly driving them.

But culture can only create the intention and desire. For the most effective marketing, brands need to have the right tools at their disposal.

The best tools are taking a broader perspective. From analytics packages, tracking multiple channels through an ad server or creating and maintaining a multi-faceted view of our audience through a Data Management Platform (DMP), the picture is much wider.

Brand leaders will no doubt be presented with the outputs of these tools. Additionally, they must ask how the business is developing its ability to analyse and share the bigger picture.

12

1

5. Never be satisfied. What are we learning next?Constant industry and market change means that there is no one correct answer.

Teams need to keep testing different ideas, processes and technologies within each of their individual advertising channels, whilst also finding new ways to benefit from their combined effect.

So after they have been shown the success of their search team’s work, brand leaders should always ask what’s next.

Big questions, big pictureIf you’re the type of CMO/CEO who wants to learn all the details, that’s fantastic. Plenty of agencies, ours included, would love to help you.

But if your time doesn’t permit delving at that level of detail, then asking these 5 questions will help make sure that your teams make the most of search and that search is connected up to the rest of your business.

13

Why search is never JUST search: The 5 big questions of the CMO/CEO search agenda 1

by James Murray, Search Advertising Lead UK, Microsoft

Search at the Heart of Modern Marketing

2

16

A convergence of marketing disciplinesThe stars are aligning for the rise to prominence of Search within the wider marketing mix. For too long Search has been the last kid to be picked for the football team. However, attitudes are changing and Search is quickly becoming the dependable captain of the team, providing the solid foundation which enables other players to flourish.

But with new roles comes new responsibility. As we move towards a world of greater collaboration, the modern search professional is expected to diversify their skills and expertise to encompass display, social, email, and increasingly offline disciplines like TV and billboard advertising. As skills and disciplines converge, this presents a new opportunity to harness the power of these channels together.

Ultimately this is a great thing. Why? Because it’s what our customers expect and deserve. We’ve been banging the drum of integrated experiences for years. But with the proliferation of devices on the market it is now truer than ever that customers don’t see channels, they just see the brand. Not only that, but integration has its own intrinsic value, bringing increased brand awareness and engagement.

The halo effect of multiscreen advertising Intuitively, we have known for a while that there is a halo effect in Search when consumers are exposed to advertising across multiple screens and platforms.

Historically though, this has been very difficult to measure. Using the Microsoft ID as an identifier across multiple platforms and properties it is now possible to show the correlation between ad exposure and the impact that has on brand engagement.

The research shows that exposure to ads across multiple screens leads to increased online activity either through searches for the brand or direct visits to the brand’s website or social media outlet. Put simply, the more channels you use in your advertising, the more people will engage with your brand.

2

17

In this case, compared to the control group, users were almost twice as likely to search for brand-related terms or visit a relevant brand website when exposed to an ad on PC, but were almost 2.5 times as likely when exposed to ads on PC, mobile and Xbox.

The message here is clear, advertising across multiple screens increases the likelihood of people searching for your brand or visiting your website. That halo effect of Search has huge implications for cross-device conversions.

Users who searched for brand-related terms or visited brand websites/social pages after being exposed to ads

3.4%

Control

6.7%

PC

7.1%

PC + Mobile 8.3%

PC + Mobile + Xbox

Search at the Heart of Modern Marketing 2

Shifting the needle on brand perceptionsWhat’s more, further research from Microsoft shows that by combining different digital media you can achieve greater impact with brand awareness.

For this study Microsoft worked with a top finance brand and exposed participants in a controlled research study to multiple ad formats across both search and display. The idea was to see what the benefit was of using different types of advertising as opposed to multiple impressions of the same type of ad.

What we can see here is that compared to the control group, people who saw either two search ads or two display ads had an incremental uplift in aided brand awareness of 3-4%. However, the group that was exposed to both search and display ads were 8% more likely than the control group to recall the ad and therefore had increased brand awareness.

The ramifications here are that search and display need to work in conjunction as the combined effect is greater than the sum of its parts. Search and display both have significant parts to play in getting your brand message across, but people exposed to both types of advertising are the ones that are most aware of the brand.

1 Search & 1 Display Ad2 Display Ads2 Search AdsControl - No Ads

100

103

104

108

2

What does this mean for search marketers?Multi-channel, cross-channel, omni-channel – whatever the phrase du jour we shouldn’t be surprised that marketing is becoming more integrated. Brands need to be capitalising on this trend now, not dismissing it as something for the future.

Search has a vital role to play in the new marketing mix, not only as a mechanism to improve the halo effect of other marketing disciplines but as a cornerstone of integrated brand awareness and engagement.

As the glue that holds together digital, and by extension marketing, search has an opportunity to take the lead in this new world of collaboration. The net result of seizing such an opportunity is that everybody wins. Not only will your campaigns perform better across platforms and devices, with increased engagement and brand awareness, more importantly, your customers will love you. Because you’re delivering what they want – joined up messaging, wherever they are and however they chose to interact with your brand.

And that my friends, is marketing nirvana.

Search at the Heart of Modern Marketing 2

19

by Ben Irons, Head of Paid Search, PHD

Brand Vs DR

3

The value of Paid Search marketing has traditionally always been seen through a DR lens and has sat at the lower end of the consumer purchase funnel, however advertisers are now able to utilise Paid Search as a marketing channel to raise awareness of their brand through traditional PPC, Display and also Video advertising to fuel demand at the higher end of the consumer purchase funnel whilst also gathering rich audience data for remarketing or future targeting.

How is search behaviour evolving and how can brands position themselves through Paid Search? Consumers are evolving the way they search and this is leading to opportunities for brands to raise awareness and associate themselves with certain search behaviours and trends. From the latest growth trends Google are sharing, searches beginning with ‘why’ are growing 1.5x as fast as searches beginning with ‘what’, and searches for ‘how-to’ have tripled in the last three years.

Brands can look at these trends and position themselves accordingly by focusing on three main areas.1. Purpose searches

2. Category searches

3. Product searches

By focusing more on purpose searches and harnessing the increasing number of searches in these areas brands can associate themselves with a purpose rather than focusing purely on a product.

22

3

As an example, if we compare search volume for a branded product, a generic product and a purpose keyword we can see that the purpose keyword (in this case ‘running’) generated more searches in 2014 than the product and branded search queries:

The opportunity here is pursuing the purpose rather than focusing on the product keyword and raise awareness of your brand in this way rather than relying on generic keyword bidding.

Volume of searches January 2014 to December 2014 from Google Trends.

Running Shoes

Nike RunningRunning

Apr 2014 July 2014 Oct 2014

23

Brand Vs DR 3

3

So is ‘branding’ in PPC just about bidding on branded Keywords?PPC strategies are constantly evolving and the use of branded keywords is much more than a straight ‘on or off’ decision. In certain verticals where competition is high and where brands are bidding on each other’s keywords then it is necessary to protect your brand’s position against competitors with an ‘always on’ strategy.

The search market is seeing a trend for increasing CPCs on branded keywords to achieve P1 in the SERP. Some advertisers are experiencing between a 25% and 30% increase year on year in Brand CPCs which is putting pressure on the channel from a CPA perspective.

Brand’s should consider their overall search marketing position when reviewing their brand bidding strategy and consider the following points.

• Do they need to protect the brand position in a hostile market?

• How strong is the brand’s SEO position for high cost brand keywords?

• What is the main business objective and how are we using Search to achieve that objective?

• Should we only use Paid Search to target new customers through RLSA?

What role does remarketing play in branding and DR?Audiences are becoming more and more evident in PPC strategies across all verticals. Knowing who your consumers are, what they are searching on, and how to reach them is a critical part of any PPC strategy.

BrandingWith the ability to share remarketing lists across YouTube, the Google Display Network and Google, search avenues are opening to increase branding activity by targeting audiences similar to the ones you know who have converted or visited your website and targeting these audiences across their internet journeys.

Remarketing can also work to mean you’re only targeting new customers rather than remarketing to existing converters through negative remarketing. This tactic can work well across all verticals meaning marketing budgets for Search can be made to work harder to achieve new customer growth and meet business objectives.

DRRemarketing is a strong tactic for driving conversions from potential consumers who have visited a brand’s website within a recent time period. The introduction of RLSA has increased the opportunity to capture a previous visitor who is still in the ‘consideration’ phase of the purchase journey and has shown intent towards your brand by visiting the site. Ensure that remarketing lists are set up in the AdWords account and that RLSA campaigns are active to capture this audience when they are searching again in the generic keyword space.

24

What about the role of Video in branding and DR?Visual can be a key element to any sort of branding effort, and videos do a great job of providing sight, sound, and movement that users will associate with your brand. Utilising Brand Lift Surveys can offer insight into brand recall following Ads being viewed and can contribute to the measurement of your branding campaign.

Alternatively, with AdWords Video campaigns you can measure traditional direct response metrics such as clicks and conversions, but also how many users viewed your entire video. Google also tracks any shares and likes you get from your video ad.

Brand Vs DR 3

25

by Chris Soo, Head of Biddable Media, Unique Digital

Mobile – the migration from tablet to mobile and dissecting the idea that search on desktop is dying

4

28

4

“The year of the mobile” has been a topic of discussion for the past 5 years. 2014 may finally have been when “the year of mobile” actually came to fruition. In 2012, Facebook were only just launching their mobile ad offering, fast forward to Q3 2014 and mobile now accounts for 66% ($3.2B) of their overall ad revenue. It’s also estimated that mobile search will account for around 40% of Google’s overall search revenue in 2014. A recent study by eMarketer says that whilst they expect mobile search spend to grow 47.7% in 2015, desktop search is set to decrease by 9.4%. Does this mean the death of desktop search?

Mobile search ad spend is expected to see double digital growth for at least the next 4 years. There’s no hiding the fact that the growth of search ad spend is heavily dependent on the growth of mobile. However, it’s important to understand the role of mobile and add a layer of context before we start reallocating budget to mobile from desktop. In order to dissect this argument we need to look at 3 key areas.

1. The definition of mobile search

2. Understanding the mobile user

3. Understanding the role of desktop

Us Digital Ad Spending Growth, by Channel, 2012-2018 (% change)

2012 2013 2014 2015 2016 2017 2018

Mobile* 178.3% 122.0% 83.0% 50.0% 41.0% 26.5% 23.0%

Display 238.0% 134.6% 87.2% 52.3% 41.9% 27.0% 22.7%

Video 257.5% 169.0% 119.0% 70.4% 50.2% 33.0% 24.2%

Banners, rich media, sponsorships and other**

235.2% 129.4% 81.6% 48.4% 39.9% 25.5% 22.3%

Search 201.7% 120.8% 82.3% 47.7% 39.3% 25.2% 22.2%

SMS.MMS/P2P messaging

-11.0% -0.3% -0.5% 3.2% 4.9% 0.8% 0.8%

Other (classified and directories, email and lead generation)

86.6% 253.5% 104.8% 85.6% 70.1% 44.3% 40.3%

The Definition of Mobile SearchThe eMarketer report defines mobile ad spend as all ad spend on mobiles as well as tablets. Google on the other hand consider tablet as an extension of desktop, as such desktop and tablet targeting was rolled up into one as part of enhanced campaigns. That said, the growth of tablet sales are starting to slow down and only 2% of people only use a tablet to go online. This means that even if we exclude tablets from the mix, the market is still shifting towards mobile. Knowing that mobile is still the bulk of the growth we need to understand who are the people searching on mobile. Are they brand new users or existing users spending more time online on multiple devices.

Understanding the Mobile UserGoogle’s Consumer Barometer research shows that the average UK person owns 2.7 connected devices. However, the users rarely use mobile exclusively to access the internet. By only focusing on mobile you’ll miss out on at least 28% of the overall audience. In fact, only 6% of people exclusively access the internet via smartphone vs 20% who are desktop exclusive.

This means that we shouldn’t be treating mobile users as a new audience, rather an extension of your current audience. Therefore increasing the importance of being able to measure the user journey across multiple devices. Google have taken a step towards this with their “estimated total conversions” available through Adwords. This uses Google signed-in data to estimate whether a keyword was clicked on via one device but converted through a different device. This isn’t limited to Google. Other vendors are entering the space to give a different or more robust single customer view. In the ad server space, Atlas’ use of Facebook IDs will allow the tracking of activity across devices as well as solving the cookie problem. This will allow us to understand the true value and ROI of ad spend regardless of channel or device.

29

Mobile – the migration from tablet to mobile and dissecting the idea that search on desktop is dying 4

Desktop*** 6.6% 3.0% -1.2% -3.0% 9.4% -8.0% -10.4%

Display 9.8% 5.2% 4.7% 3.2% -1.2% 1.4% 2.4%

Video 37.0% 33.9% 27.7% 17.7% 8.5% 7.9% 6.6%

Banners, rich media, sponsorships and other**

4.4% -2.2% -3.4% -3.6% -6.8% -2.8% -0.7%

Search 5.1% -0.1% -8.2% -10.8% -21.6% -23.8% -37.0%

Other (classified and directories, email and lead generation)

2.9% 7.1% 3.8% 1.5% -3.4% -4.4% -8.8%

Note: *ad spending on tablet is included; **includes ads such as Facebook’s Sponsored Stories and Twitter’s Promoted Tweets; ***includes spending primarily on desktop and laptop-based adsSource: eMarketer, June 2014

With what combinations of devices do people go online?

Computer- only users

Smartphone-only users

Tablet-only users

Computer & smartphone-only users

Computer and tablet-only users

Smartphone and tablet-only users

Smartphone, tablet and computer users

20%

6%

2%

30%

6%

8%

29%

Question asked: “How often do you use the Internet - through your web browser or apps - for personal reasons (i.e. all non business or work related purposes) on [computer/smartphone/tablet]?” Total Responses: 755 Base: Internet users (accessing via computer, tablet or smartphone) Source: The Connected Consumer Survey 2014

30

Understanding the Role of DesktopUp to now mobile investment has been dependent on the readiness of clients. For clients with no mobile sites or mobile app presence it made less sense to push budget in an area which was known to have a lower conversion. Technology has caught up but conversion rates still differs significantly. Data from Monetate shows that despite the development of mobile sites, the conversion rate across mobile is still 71% lower than desktop. Part of this will be down to the context of the search being made. More people research on mobile than purchase on mobile (21% vs 6%), this is partly down to activities such as showrooming - explained in another chapter of this handbook. However, this is reversed when it comes to desktop (83% vs 87%). So, users are almost 4 times more likely to research a product on a desktop, and 14 times more likely to purchase an item on desktop. Until user experiences or buying confidence on mobile changes, the desktop still has a key role when it comes to purchase.

With this in mind, we need to also start thinking about how we can alter the messaging we use across devices, whether we need to look at different KPIs for mobile such as app downloads and how we’re engaging with post app download tracking to measure the true ROI of an app download via search.

In summary, the growth in search is definitely reliant on mobile and we do expect to see a decline in desktop searches. However, desktop still has a very important part to play. Desktop continues to be part of the user journey and as users diverge across multiple devices (desktop, tablet and mobile) we need to be able to measure correctly, show the right messaging and still be visible when your customers are actively searching. Mobile isn’t a new channel, it’s an extension to our current customer base.

1. www.bloomberg.com/news/2014-10-28/facebook-posts- third-quarter-revenue-exceeding-analyst-estimates.html 2. www.dotretailer.com/blog/ecommerce-conversion-rates- for-mobile-are-over-42-less-than-site-average-why/ 3. www.monetate.com/research/#ufh-i-34269668 ecommerce-quarterly-eq3-2014

4

Smartphone

Computer

Tablet

Other internet-enabled device

Don’t know

Which devices did people use to make their purchase?

6%

87%

7%

1%

0%

Smartphone

Computer

Tablet

Other internet-enabled device

Didn’t recall device type

Which devices did people use for product research?

21%

83%

16%

2%

1%

Question asked: “Which of your devices did you use at any stage of your research online?” Total Responses: 1999 Base: Internet users | Answering based on a recent purchase (in select categories) researched online. Source: The Consumer Barometer Survey 2014

Question asked: “Which device did you use to make your purchase?” Total Responses: 1526 Base: Internet users | Answering based on a recent purchase (in select categories) purchased online. Source: The Consumer Barometer Survey 2014 Note: Answers may not sum to 100% due to rounding, no answers, don’t knows.

Mobile – the migration from tablet to mobile and dissecting the idea that search on desktop is dying 4

31

by Joe Roman, Head of Paid Search, agenda21

5 Reasons Why Search Advertisers Need to Consider Responsive Web Design for their Site

5

34

Figure 1: Screen shots of the agenda21 homepage on desktop, tablet & mobile

Figure 2: Search volume by device 2013-2014

With smartphone and tablet use in the UK continuing to rise, it is increasingly important for advertisers to consider the options available to display their websites on a wider range of devices.

Responsive design is one such option which “uses flexible layouts, adaptable images and cascading style sheet media queries to build web pages that detect the visitor’s screen size and change the layout accordingly” (source: techtarget.com). The sites that make content both accessible and consistent will be the ones that succeed. Below are 5 reasons why Responsive Design must be a priority for advertisers in the near future.

1. Growth in the Use of “Mobile” DevicesThe use of mobile devices has grown year on year and shows no sign of slowing down. Coupled with this, the sales and use of PC and desktop computers is on a steady decline.

68% smartphone penetration in the UK 40% tablet penetration in the UK 55% of smartphone users search on their phone everyday(source: Google Consumer Barometer)

With such high penetration rates of mobile devices, we are seeing a considerable shift in search behaviour. Search volume statistics show desktop searches to be decreasing and tablet and mobile searches increasing. Whilst search statistics will vary depending on the vertical, generally they will look as they do in figure 2.

It is crucial for advertisers to acknowledge the change in search behaviour. Especially since the different categories of device are starting to homogenise, meaning it is crucially important to optimise your site for mobile.

Search Volume by Device2013

TabletTablet

Mobile

DesktopDesktopMobile

2014

5

35

2. Improved Site EngagementMobile devices are often the first “proactive” point of interaction for users, particularly in cases where they have been subject to other media, such as a TV advert. Indeed, 14% of users surveyed used their smartphone during their last purchase. It could therefore be argued that mobile content needs to be the most engaging. At the very least, mobile sites must be consistent with what can be consumed on more “traditional” devices for a more congruous user experience. Some of our clients at agenda21 have experienced conversion rates doubling after introducing a responsive design site.

Mobile devices do, however, ultimately yield lower conversion rates. There is no universally accepted answer for why this is the case, but we can speculate that trust issues and early adoption of mobile searching are at play. Furthermore, mobile sites can offer limited functionality, often focusing too heavily on a conversion point meaning additional information can be tricky to find. Despite this, conversion rates for mobile devices should increase in the near future as users become more used to the layout and functionality.

3. Mobile Search Space is Still CheaperBy using responsive design to ensure optimal experience for users, there is an opportunity to capitalise on cheaper CPCs.

The controversial migration to enhanced campaigns last year, forcing search advertisers to advertise in the tablet and mobile space, without any control over tablet investment and only a bid modifier for mobile, caused issues for advertisers, specifically with mobile conversion rates. Despite Google’s best efforts, advertisers are still reluctant to invest in mobile search activity with the same vigour as desktop and tablet, meaning a cheaper environment across many verticals.

The cheaper traffic can actually offset poorer conversion rates to bring about a more efficient return.

4. Data ConsolidationResponsive design means using only one site. Therefore only one site needs to be tagged. This means data no longer needs to be stitched together in order to get an overall picture. Data can then be segmented to view site stats for the different devices, making analytics and optimisation more straightforward.

5. Responsive Design vs AppAs a paid search practitioner, despite the inroads that are being made in to in-app tracking, I would always suggest that having a consistent site experience across all devices is more important. 66% of smartphone users surveyed expect sites to work as well on their mobile as on their desktop.

A good, optimised site provides the initial engagement, whereas an app’s purpose is largely focused on user retention through offering a richer experience. An app is very rarely going to provide you with brand new customer engagement.

Rather than a choice of the two, ideally an organisation should prioritise which is the right move for them to maintain and increase resonance across different devices.

In conclusion, advertisers must make planning around responsive design a priority, not only to reflect changes in search behaviour, but to deepen engagement for their customers, and tap into the relatively cheap search environment that currently exists. Consumers are using more and more devices which means sites must adapt to provide consistent content to their audiences no matter what device they are using.

5 Reasons Why Search Advertisers Need to Consider Responsive Web Design for their Site 5

by Johanna Deschamps, Product Marketing Manager UK & France, Marin Software

Audience Based Search Marketing

6

Paid search remains so successful for one primary reason. That is because it creates a direct link between users and their intent. For marketers, the ability to deliver an ad promoting golf clubs to users searching for [best golf clubs] represents a revenue opportunity that commanded more than half of online ad spend in 2014 — that’s more than display, social or mobile. To date, these channels have failed to match the performance characteristics of search. However, audience-oriented advertising combined with the power of data is starting to change that.

Integration of Audiences with IntentAcross the online world, advertisers increasingly have the ability to understand their customers not just by their online declarations through search (e.g. I need “new golf clubs”), but also by their digital trail — the footprints of intent-based behaviour.

A customer’s previous visits to a product page, mobile check-ins and social shares can be combined with offline, CRM and third-party data to create a powerful combination of attributes. When aggregated across a large population of individuals, this provides advertisers with a clear understanding of their customers across all channels, not just search.

Applying Audience DataToday, the integration of audience data in digital advertising exists most commonly in display. Across Google and Facebook marketers can target users by interest categories. However, just beyond these networks, there are four near-term applications of audience-oriented data that have far-reaching implications for performance marketers. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.

6

38

1. Audience AnalyticsDirect response advertisers today can make some pretty good guesses about their customers based on search keywords and conversion tendencies. In structuring search campaigns, many advertisers build a “taxonomy of intent”, sometimes without explicitly knowing so. If a travel company creates a “budget travel” campaign and a “luxury travel” campaign, it’s easy to make assumptions about the differences in the users who gravitate toward one versus the other.

“Colouring” those insights with additional data — household income, marital status, age, will validate or potentially challenge those assumptions. For example, it may be that the average household income for “luxury travel” consumers is much lower than expected because they’re more aspirational buyers. This revelation might cause an advertiser to rethink creative strategy.

2. Search Bidding Based on AudiencesArmed with powerful insights on what types of customers they are acquiring, many advertisers are already re-thinking their bidding strategy on media. For example, Marin Software and MoneySupermarket’s DADI Award winning entry in the search category applied audience data to search bidding. In MoneySupermarket’s car insurance vertical they identified that consumers over 25 were more valuable than consumers under 25. As a result, they could bid up or down on search ads depending which side of that age group the searcher fell. Consequently, the concept of “buying a demographic” in the world of display has been adapted to search by simply bidding up keywords that have a higher propensity to drive engagement from a more valuable demographic.

3. Retargeting Using Audience DataRetargeting users on search has been a reality for a while now through Google Remarketing Lists for Search Ads (RLSA). In the simplest approach to search retargeting, an advertiser can implement distinct strategies for dealing with just two audiences: people who have been to my site before and people who have not.

An advertiser may decide that users who have already visited are closer to purchase and might bid more aggressively when that audience searches for a relevant keyword. The potential targeting options are far greater when one starts to layer additional first or third-party data into the mix. For example, knowing the age and gender of a potential customer might alter an advertiser’s creative or bidding strategy when retargeting that individual.

4. Lookalike TargetingAs advertisers get smarter about understanding their existing customers, it becomes easier to identify users that have a high probability of converting into new customers. Algorithmic tools have made it possible to not only analyse a segment of existing customers (e.g. “budget travellers”), but also to identify similar segments of the same population (e.g. “lads holiday planners”) who have the potential to convert at a similar rate. This level of insight enables marketers to quickly generate display or social ads targeted at those similar segments and acquire more like-minded customers.

What’s Your Audience?Tapping into user intent is no longer adequate for remaining competitive in the new digital landscape. Advertisers must increase their level of sophistication and continue investing in new, innovative audience technology to win the battle for revenue online.

Audience Based Search Marketing 6

39

by Joe Doveton, Head of Conversion, Oban Digital

Analytics

7

Coming into 2015, Web Analytics is at something of a crossroads.

For a decade or more, the market has been dominated by two conflicting business models and approaches. On the one side, there is Google Analytics (GA), perhaps the greatest free tool ever in the history of digital marketing. GA has built a monolithic market share based on simplicity and ubiquity, and as a plug-and-play system that works with their dominant advertising solution, AdWords, one would expect them to be hard to beat. On the other side, led by the mighty Adobe Marketing Cloud and SiteCatalyst are some extraordinarily powerful paid tools. SiteCatalyst in particular has become a byword in reliability and pure processing power. Also worthy of a mention on the paid side are Webtrends, and of course IBM Coremetrics- both of these long lived solutions are still considered industry powerhouses.

However, data requirements are changing fast. Users are increasingly mobile, skipping the web and accessing digital properties via branded apps. Channels are more complex and interdependent. A customer may start their engagement with your brand via a query in a search engine, refine their product selection on their iPad or Android tablet via a shopping app, before purchasing in store. The importance of analysing these touchpoints requires cross channel identification of customers (no mean feat in the days of cookie consent and privacy concerns) plus really granular data aimed at better defining user intent.

All of this means more (or if you really must, “big”) data.

7

EN SaleThis Goal was completed in 678 sessions0.36% funnel conversion rate

Prices Page187,688

Customer Details14,008

EN Sale678

0.36% funnel conversion rate

14,008 (7.46%)proceeded to

Customer Details

(entrance)/index.html/lond-transport/index.html?aid=19/how-it-works-what-you-get-with-the-london-pas.../london-attraction/index.html

(exit)/london-attraction/index.html/lond-transport/index.html/how-it-works/index.html/london-pass-prices.php

/checkout.php/london-pass-prices.php(exit)/london-attractions/index.html/how-it-works/index.html

20,91719,85510,61410,36710,050

62,76526,62816,27415,61614,289

13,005143622515

187,688

0

0

173,680

13,330

678 (4.84%)proceeded to EN Sale

42

43

Analytics 7

So as we come up to 2015, here are the major challenges:

SkillsThe web analyst, presumably previously kept in a darkened room, is emerging as the most important person in a digital business. The ability to anticipate, interpret, act on and optimise based on data has never been so important. This army of analysts requires leaders- so Gartner predicts that by 2015, 25% of large global corporates will have appointed a Chief Data Officer. So, hiring the right staff- and enough of them- is going to become vital to the success of online business.

GranularityMost old style analytics tools give you the view at page level, i.e. they can tell you stats like journeys from URL to another and data like average page dwell time. In the modern web, analysts are looking for much more granular data at an element level. For some time, tools like ClickTale and Crazy Egg have provided sophisticated visual analytics like Heatmaps or Scrollmaps. But as technology moves on we may need data on interaction with dynamic elements like Javascript drop-downs or AJAX generated dynamic pricing. Expect future data analysts to require much more information at this “microconversion” level - especially more modern interactions like mobile swipes and taps.

Data Visualisation

As the data gets more complex, the ability to mine the most relevant statistics and present them in a format that the board will understand becomes paramount. Third party data platforms like, Plateau, offer a neutral canvas for web analysts to present their own spin on data. Analysts who can code in the statistical programme R will also become increasingly valuable.

Cross-PlatformThe customer as an individual is being put at the heart of everything a business does on the modern web, so tracking individuals through complex journeys on multiple devices is a primary objective for analysts. Ethical methods of persistent tracking are being employed to track visitors via journeys across desktop to tablet or mobile, and incentives offered by merchants for users to log-devices. Complex technology is being used to test customers redeeming offers or vouchers in store; and the building of personas based on clusters of consumer behaviour is allowing businesses to personalise landing pages and experiences based on data. Integration with data held in legacy systems like eCRM (once again, see IBM and Adobe for this) allows web publishers to also leverage reams of historic data about the behaviour of their customers.

Predictive AnalyticsProcessing web analytics in real time is a real problem for many businesses. Even Google Analytics uses a smaller data sample as soon as the users adds in filters. However, the ultimate aim of the web analyst is to be proactive - not reactive. So, corporations need to learn how to anticipate trends and forecast changes in user behaviour based on certain conditions. Seasonality is relatively easy to predict- most retail businesses expect their busiest months to be November and December. The key trends are predicting the impact of product scarcity of surplus, of effective or less effective marketing campaigns (e.g. TV) plus assessing the impact of good and bad PR for your company in order to optimise and personalise in real-time.

Written by Kevin Bharakhda, Biddable Media Account Manager, Tug

Drive More Relevant and Cost Effective Traffic with Ad Extensions

8

Google’s Ad Rank algorithm changed to include Ad Extensions, making this feature even more important to advertisers using Pay per Click (PPC) advertising. Ad Extensions visually improve PPC ads and allow advertisers to communicate more information to the customer. More importantly, they now affect how much the advertiser will pay each time a customer clicks on their ad. Using Ad Extensions efficiently therefore gives the advertiser the opportunity to receive higher volumes of clicks and conversions for their budget.

Despite this there are a high proportion of advertisers who are not making the most of the Ad Extensions available to them. In this chapter we outline a list of the types of Ad Extensions which we believe to be the most important, outlining which verticals they are best used for and when they should be used.

Improve Customer Journeys with Enhanced SitelinksOne of the most commonly used ad extension is Enhanced Sitelinks. This sitelink is used to direct the customer to useful and relevant content beyond the main PPC ad’s landing page by deep linking the customer further into the site. Compared to the 35 character limit for traditional sitelinks, Enhanced Sitelinks allow for 95 characters giving the advertiser the opportunity to provide a fuller description on where the link will take the customer. Using this Ad Extension, Citrix were able to boost CTR by 15% leading to a 20% increase in conversion rate, which ultimately brought CPA down by 32%.

Example ad:

Key Verticals: All

When should they be used? When the advertiser wants to give the customer multiple options to access specific webpages on their website.

8

46

Target Customers Through Proximity with Location ExtensionsThe Location Extension allows the advertiser to attach a business address to their PPC ad. This helps to attract customers who are either in close proximity to the business or are searching for it in a particular location. LEGOLAND Discovery Centre used this feature and were able to increase CTR by 67%, thus improving conversion rate by 9.73% and decreasing CPC by 30%.

Example ad:

Key Verticals: Retail & Hospitality

When should they be used? When an advertiser has a physical location that customers can visit.

Promote Your Brand’s Online Footprint with Social ExtensionsTo use Social Extensions, an advertiser must have a Google+ page with high quality posts and over 100 followers. This allows users to see how successful and reliable a business is amongst other customers; hence building up their trust in the advertiser. Guinness World Records used social extensions and were able to increase their CTR by 11.5%.

Example ad:

Key Verticals: Retail & Travel

When should they be used? To help with the pre-qualification of a brand’s identity using their social media following.

.

Drive More Relevant and Cost Effective Traffic with Ad Extensions 8

47

Advertise Unique Selling Points with Callout ExtensionsCallout Extensions appear beneath the main PPC ad and are used to highlight valuable information to customers about their product or service, drawing attention to important details or offers they may have. Citrix used this feature in e-Commerce accounts and found that it increased CTR by 24% and conversion rate by 10%.

Example ad:

Key Verticals: SaaS

When should they be used? To highlight key features of a service or product

Promote Awards and Accreditations with Review ExtensionsThis type of Ad Extension allows an advertiser to show a recommendation from a reputable third party within their search ad; hence building trust with their potential customers. The positive effect of Review Extensions has been proven by Citrix, who used this feature and increased their CTR by 7%.

Example ad:

Key Verticals: SaaS, Hospitality, Leisure & Travel

When should they be used? When an advertiser has a positive review from a trustworthy third party

48

8

Increase Digital Downloads with App ExtensionsWith this extension customers can download digital goods such as apps, books, music and movies quickly and easily. With the click-to-download option the customer can be directed to the App Store and given the option to install the App on their mobile device. Tug has seen App Extensions increase CTR by over 40% and improve conversion rates by over 30%.

Key Verticals: Any advertisers with apps registered at an App Store

When should they be used? When an advertiser’s objective is to promote downloads of their mobile app in addition to gaining traffic to the site.

Increase Customer Touchpoints with Call Extensions This extension allows advertisers to drive additional leads to the business by adding a call button to a mobile ad or a phone number to a desktop ad. Advertisers can track all calls and conversions from customers who have dialled a forwarding Unique Telephone Number applied by Google. The positive effects of Call Extensions saw Ocean Florida increase their click through rate by 11%.

Example ad:

Key Verticals: Hospitality, Travel, Utility and Services, Local Businesses

When should they be used? When a brand has the ability to generate sales over the phone

To conclude, since Google updated their Ad Rank algorithm to include the advertisers’ usage of Ad Extensions, it has never been more important to ensure that advertisers are making the most of these easy to implement features. Not only do they improve the customer journey, conversion rate and CTR but their positive effect on Ad Rank can also help to reduce CPC; ultimately driving cheaper, more relevant traffic to an advertiser’s website.

.

49

Drive More Relevant and Cost Effective Traffic with Ad Extensions 8

by John Tegner, Product Marketing Manager, Sizmek and Graham Ratcliffe, VP Product and Marketing, Abakus

The evolution of attribution

9

Last event or last click attribution was an important innovation in the emergence of Search Engine marketing and its measurement. Since the mid-1990s this model has dominated the digital marketing industry, and has been adopted across 3rd party adservers, and programmatic ad delivery platforms alike.

Current trends see marketers investing in an increasingly diverse mix of advertising formats to reach consumers. To succeed in today’s world marketers need to be able understand how these channels work together holistically to drive success.

Despite recognition that last event attribution is a flawed measurement, its convenience and simplicity mean that it is still widely used by brands and agencies. Furthermore, the volume of data collected along throughout the consumer journey, and the resources required to digest and analyse it, lead many marketers to believe there is no viable alternative.

It is recognised that the last event attribution results in:

• Upper funnel activity is undervalued

• Lack of insight into cross-channel impact

• Siloed internal marketing teams compete rather than collaborate

What are the alternatives to last event attribution and what should marketers consider when implementing an attribution solution?

52

9

53

The evolution of attribution 9

Approach How It Works Pros and Cons

Simplistic

100% of credit is awarded to a single point, most commonly the last event, bu last click, and first click models are also used

• Ignore the impact of upper funnel activity

• Excludes cross-channel impact

• Encourage siloed marketing

Rules based

Fractional credit awarded to touch points along the user journey.. Rules are set by the advertiser or agency analytics team, such as even-weighted, position-based and time decay

All hueristic models are based on perceived impact of events on a user journey path vs. objective analysis

Factors based

Additional custom factors are incorporated, such as:

• Ad recency: more credit given closer to conversion

• Order: more credit given earlier in the funnel

• Ad size: more credit for larger ads

All hueristic models are based on perceived impact of events on a user journey path vs. objective analysis

Advanced Attribution

Leverages data mining techniques and statistical models to accurately calculate the assignment of credit.

Mathematical approaches used:

• Clustering

• Logistic Regression

• Markov-Chain

• Game Theory

• Based on valid statistical analysis, these models offer a less arbitrary methodology for awarding attribution credit

• Increasingly gaining favour

• May involve additional investment or complexity

Alternative 1: Heuristic modelsOver recent years, heuristic multi-touch attribution models have been adopted, enabling credit to be assigned to multiple touch points in the customer journey dependent on their perceived contribution to the conversion. These multi-touch, rules-based models appear superior to last click as different events in the user journey can be evaluated to see how attribution would change if looked at from another perspective. However, due to the speculative nature of these insights, heuristic models often result in credit being as arbitrarily attributed to touch points as with last event attribution.

Alternative 2: Advanced attributionAs data processing has evolved, marketers now have greater access to sophisticated techniques that use algorithms and statistical models to understand relationships and patterns within the consumer path to conversion. This approach has been found to be a more effective way to understand the impact of each touch point along the consumer journey compared to rules-based attribution. These techniques require time and financial investment, as well as specific skills and expertise. Investment is unlikely to be forthcoming without proof of success, yet no proof is possible without investment. The ‘black box’ solutions offered by many vendors compound this issue, and the market may not be comfortable with the application of an advanced statistical methodology to attribution.

9

54

Exposures along the path to conversion

Exposures along the path to conversion

Individual user paths

+site data +non-converters

Conversion Event

Conversion Event

Cre

dit

Rules based (ex: recency)

Algorithmic Attribution

10% 15% 20% 25% 30%

Considerations when implementing attributionWhen reviewing the features of different attribution models, marketers should focus on results. As implementing an advanced attribution solution can be time-consuming and resource intensive, marketers should look for software-based solutions that can process data easily and efficiently.

Marketers should look for an intuitive interface that provides both a holistic summary of performance and granular actionable insights. Accurate measurement is the foundation of attribution and the approach must be proven to work mathematically and be transparent enough to explain to a non-technical audience. Finally an attribution solution should offer media neutrality, as a neutral media source can prevent a conflict of interest arising in reporting performance results.

Attribution is vital to search success. With any attribution strategy, be sure to:

• Look at your data holistically across all channels

• Isolate the key conversions that you want to optimize

• Incorporate cost data

• Incorporate viewable impression data

Checklist for selecting an attribution vendor• Can the approach be explained?

• What level of resources is required for implementation?

• How is data presented?

• Is the output actionable?

The evolution of attribution has begun. Experiments with heuristic attribution have resulted in a rejection of rules-based models in favour of advanced attribution. By adopting an approach that provides actionable insights, marketers can develop deeper understanding of the touch points on the consumer path to purchase.

The evolution of attribution 9

55

Exposures along the path to conversion

Conversion Event

Cre

dit

Last Click

0% 0% 0% 0%

100%

by Matt Brown, Head of Earned Media, Unique Digital

Showrooming

10

“The staff at Jessops would like to thank you for shopping with Amazon”

Read the sign in a shop window shortly after the camera chain went into administration. It was a dry reaction to a growing problem for bricks and mortar focused retailers.In 2010, retail analysts and media outlets in the US began to talk about the electronics chain Best Buy becoming a “showroom for Amazon.” From this, the term “showrooming” was born i.e. consumers using a retailer as a showroom to view products in person before buying them from an online retailer.

A Columbia Business School study examined 3,000 consumers in three markets (US, UK, Canada) in order to shed light on how consumers are actually using their mobile devices in-store. Who are these shoppers? What are they doing with their mobile devices in-store and why?

10

58

Showrooming 10

Less than £25K

£25–50K

More than £50K

40%

38%

22%

0% 10% 40%30%20%

65+ years old

50-64 years old

40-49 years old

30-39 years old

<30 years old

6%

20%

22%26%

26%

Demographics of a ShowroomerTheir age distribution remains relatively constant through age 64, before dropping off at 65-plus, as might be expected. This seems to indicate that mobile-assisted shopping is not early adopter behaviour, however rather behaviour linked to the mainstream adoption of smartphones and their integration into consumers’ lives.

When looking at income, we see a relatively high proportion of showroomers earn below £50,000 per annum, this almost halves for those earning over that salary. This would indicate that, as expected, people on a higher salary are less price sensitive.

59

10

Reasons for ShowroomingAccording to the Columbia Business School study, the biggest drivers for showroomers are lower prices (69%) and free shipping from the online retailer (47%) and showroomers are nearly as likely to use the store’s own website (70%) as another website (75%) to gather product information. This suggests that price is obviously a key concern however people are also looking for the convenience of having the product delivered to their home or work. Furthermore, showroomers do not show allegiance to a particular brand or site, they are just looking for the best possible deal.

Looking at price point triggers, we see that anything above £60 encourages showroomers to pull out their smartphones and compare against competitors. This is especially true in the Electronics vertical where 83% of the study respondents’ price checked when in-store.

Strategies & Opportunities for RetailersOver the past few years, retailers at all levels have experimented with strategies to combat this trend.

Be Found OnlineThis is an obvious one and easier said than done in many cases. However online retailers should ensure that all bases are covered cost effectively:

• Paid search coverage on particular product names

• Product listing ads are fully utilised

• Ad extensions (especially Seller Ratings and Review extensions) are being used where possible

• Google Places has the up-to-date store details

• Make sure that Google can crawl the site effectively and index all pages, especially at a product level. This will help these pages appear higher in organic listings

• Ensure the retailer ‘owns’ the first page of Google for their brand terms through PPC, SEO, Google Places, Wikipedia and Google Plus posts

These are just a few of a myriad of search and social strategies that can be used to ensure online visibility in the fight against showrooming.

60

36%

70%

53%

35%

18%

0%

Under £30 £300+£30-£60 £60-£150 £150-£300

48%62%62%58%

Showrooming 10

Make It Easy to BuyIf you can’t beat them, join them. Giving consumers the option to have free click & collect, free home delivery, gift wrapping, a good returns policy, interest free credit and obviously a massive call to action all help in capitalising on showrooming.

Reviews & AdviceAnother key trend from showrooming is product and brand review checking so ensure that each product has a review section. Furthermore, mark these up with schema tags to enable Google to show the stars in the search results.

Convenience & Service“I’m a passionate cyclist and I go to a shop that

is much more expensive than the internet.

But by the time you’ve been there for an hour, their enthusiasm is so overwhelming that you really don’t want to go elsewhere.”

Rather than reducing prices to compete, look at how customer service can be improved pre and post-sale both online and in-store.

Social Check-Ins & OffersEncourage sharing by providing special offers to social savvy showroomers when they check-in when they are in-store. This will also give the opportunity to communicate with them after purchase and beyond.

Loyalty ProgrammesNearly half of all showroomers are more likely to purchase products in-store despite equal or cheaper prices online when they are a loyalty program member, make sure that the following services are offered:

• Discounts

• Price matching

• Free delivery

• Extended warranties

• Donations to charities

These are just a few strategies to help combat and take advantage of the showrooming trend that has materialised in recent years. Many can be difficult to achieve however, if done well, the rewards will be reaped for years to come.

Make it easy to buy by utilising flexible purchasing and reviews Source: goldsmiths.co.uk, amazon.co.uk

Encourage sharing by providing special offers to social savvy showroomers

61

by Yves Schwarzbart, Regulatory Affairs Manager, IAB UK

Data Protection and Privacy

11

Data-driven advertising models are at the forefront of the UK’s thriving digital economy.

With this success come growing responsibilities and as interest in the sector rises up the agenda of politicians and regulators alike, so does one of its most important challenges: privacy.

Privacy is by its very nature – a flexible concept that means different things to different people, let alone different generations. However, the concerns over online privacy are real and should be taken very seriously by everyone involved in digital advertising.

That being the case, equipping individuals with tools that allow them to make informed choices and set safeguards for their privacy is essential. Businesses taking leadership is increasing trust in technology, and with it raising levels of confidence in taking advantage of all its possibilities, creating benefits for everyone. Providing transparency about the information that is used and giving users control over this data stand out among the many instruments in the toolkit with which to achieve this goal.

The legal frameworkSearch advertising is no exception to this task, being embedded in a legal framework in the UK in which the Data Protection Act 1998 (DPA) builds its cornerstone. The DPA is regulated by the Information Commissioner’s Office (ICO) and governs all organisations that collect and process personal data. According to the law, personal data means information which can (1) by itself or (2) in combination with other information identify a living individual.

Compliance with the DPA is primarily based on eight good practice principles which any business in search advertising that processes personal data must follow. These stipulate that data must be:

1. Fairly and lawfully processed

2. Processed for the intended purpose

3. Adequate, relevant and not excessive

4. Accurate and up-to-date

5. Not kept longer than necessary

6. Processed in line with individual’s rights

7. Kept secure

8. Not transferred to non-EEA countries that do not possess adequate data protection rules (determined by the EU)

64

11

Organisations also have to be aware that the processing of sensitive data, e.g. health information or political orientation, is subject to stricter legal obligations (i.e. explicit consent).

Designed to complement the Data Protection Act, The Privacy and Electronic Communications (EC Directive) Regulations 2003 (PECR) lay out more comprehensive rules specifically for electronic communications.

Recent changes to PECR in the form of amendments – more commonly known as the ‘cookie law’ – have posed key challenges for the digital advertising industry. The most prominent of these is the requirement to obtain consent for using all technologies – including, but not limited to cookies that store information or gain access to information on a user’s device.

A look into the futureIt is common for laws to be adjusted to new realities. The most recent development in the field of privacy regulations kicked off in January 2012 when the European Commission (EC) published proposals to update European data protection rules. The EC reforms will override existing legislation such as the Data Protection Act 1998 and would also impact PECR.

The proposals, as currently drafted, pose significant challenges to the industry, including search. By broadening the scope of personal data, the reforms seek to adopt a ‘one size fits all’ approach to regulating information. The result would put non-directly identifiable data – such as most of the information collected and used in digital advertising – in the same boat as, for example, a person’s name. Combined with the proposal to require ‘explicit’ consent for all data, this would effectively undermine the implied consent regime that is currently in force under PECR. The reforms also include a proposal to introduce a general ‘right to be forgotten’ – a right that has received considerable attention in 2014 as a result of a ruling by the Court of Justice of the European Union (CJEU) (see box-out).

The reforms have yet to be agreed and continue to be on the EU’s agenda in 2015 and beyond. The IAB – assisted by all its Councils – will carry on advocating a pragmatic and workable approach that ensures the new rules are fit for purpose.

For more information on data protection and privacy including the IAB’s concise fact sheets on these issues, please visit our website www.iabuk.net/policy.

Right to be forgottenOn 13 May 2014, the CJEU handed down a landmark privacy ruling. The court essentially judged on three issues and decided that:

• Search engines are data controllers and therefore regulated as such under existing European data protection law, including the Data Protection Act 1998

• Search engines that operate from outside the EU are subject to EU data protection rules if they have an EU sales subsidiary

• An individual has the right – under certain conditions – to have search results on the basis of his or her name removed

The ‘right to be forgotten’ has caused considerable debate among both policymakers and industry. However, its full implications as well as its future role in the EC data protection reforms remain subject to intense scrutiny.

65

Data Protection and Privacy 11

by Oscar Romero, Head of Biddable Media, Starcom MediaVest Group

International Search Engines

12

Search connects users across the globe, and search engines are the catalyst for making that connection. With the majority of industries now digitised and online turnover accounting for increasing proportions of revenue for global brands, it is vital to optimise to different search engines and cultures to increase awareness, visibility, traffic and conversions.

Domestic markets in Europe have been stagnating or only driving single digit growth due to the slowing down of their respective economies. This is clearly illustrated when looking at overall search volumes across Western European markets where we can see very little growth, and in some verticals even a decline in numbers.

Many brands face an increasing number of competitors and aggregators playing in the same space with very little differentiation other than price. Ultimately, this can have two major impacts on a brand. Firstly, an increase in competition in the search landscape can inflate average CPCs, which can result in digital media budgets not going as far as they did in previous years. In addition, CPC inflations can negatively impact a brand’s return on investment which will likely cause alarm if operating on tight margins.

Secondly, as new brands enter a market it is not uncommon for more established brands to push their prices down in order to stay in competition. Lower prices will directly impact the average order value and the only real way to compensate for this is by driving a higher volume of transactions at a lower cost.

The biggest opportunity for global brands is to implement a market diversification strategy by targeting emerging markets where competition is less intense and average order values are potentially higher due to a growing middle class sector with higher disposable incomes.

12

68

When the preceding point is coupled with the fact that search is still in its infancy in some emerging markets, it would appear international expansion is the biggest area of opportunity for global brands. As such, it should be imperative for international advertisers and global brands to target audiences in multiple markets which means that they need to work with local search engines.

Without a doubt, Google is the biggest and by far the most important search engine worldwide and currently holds almost 69% of the global market share. However, there is a high concentration of advertisers on Google and the scope for new opportunities is very limited. Whilst Google has attempted to diversify its search offering through various tools, features and betas, the fact remains it is saturated and if brands want to reach new and larger audiences they need to look further afield.

Before a brand expands its search activity to a new market, there are some key points to consider about the country they wish to operate in and the search engine they will actually use. A few examples of these are cultural differences, language requirements, account set up process/functionalities and search engine compatibility with bid management tools.

At present, perhaps the two biggest challengers to Google in other markets are China’s search engine Baidu and Yandex in Russia. Both engines are operating in the world’s biggest internet markets and leveraging these platforms would allow brands to seize opportunities that would not be available with just Google or Bing.

International Search Engines 12

69

With the extensive rise of China’s middle class and rapidly growing internet access penetration, China is a huge market to tap into. China’s search engine market has maintained stable growth in 2014 with Baidu accounting for 82% of the market share.

But operating search in China is completely different from anywhere else in the world. One of the biggest reasons for this is the government’s strict censorship regulations which affects anyone accessing the internet. Unlike with Google or Yandex, global brands require a physical business location within China to set up an account on Baidu, and sometimes might also require a prepaid campaign budget. The overall registration process could therefore take up to two months. Furthermore, brands should have a site with a Chinese domain name (com.cn or cn) which is hosted on a server that is physically located within the country.

With a share of over 60% Yandex is strongly dominating the Russian search market and is the most-visited website in Russia. It also has strongholds in Poland, Ukraine, Kazakhstan, Belarus, and Turkey. Additionally, Yandex extends its advertising to a partner network including many highly frequented Russian sites such as the portal Rambler and Mail.ru. Yandex’s clear advantage is the establishment of a robust search algorithm which is optimised based on the specificities of the Cyrillic alphabet. This is also reflected in Yandex’s unique approach to keyword match types, with four options to capture the Russian language’s intricacies; this is completely different to Google’s approach. Another big difference to Google is that Russian ads must link to Russian landing pages. In order to run non-Russian sites through PPC, the target language must be incorporated inside the ad text.

12

70

Source: WebCertain Group Ltd – International Digital Hub (14 January 2014)

Brands will still need to continue to work with the big “G” but if local search engines dominate in these emerging target markets, it makes sense to focus attention there, as competition is not going to be as fierce and return on investment likely to be higher.

In addition, experienced marketers can leverage their learnings from working in mature search markets to develop search strategies and optimisations which will likely put them ahead of their competitors in emerging markets.

Ultimately, the impact of using localised search engines can be small but also crucial in achieving a brands goals and objectives and this is highlighted by the fact bid management platform such as DS3 and Marin have integrated with localised search engines like Baidu and Yandex.

The “Big Three” in comparison and key features

Google Yandex Baidu

LanguagesAll Russian, English,

Ukrainian, TurkishChinese, English

Advertising Options

Search and Display Network, YouTube

Search, Content and Display Network, BaYan (keyword targeted banner on homepage)

Search and

Display Network

Targeting Options

Language Age & Gender Time/Day Geo-Targeting

Age & Gender Time/Day Geo-Targeting

Time/Day Geo-Targeting

Ad Specs - Title 25 characters 33 characters 50 bytes

Ad Specs - BodyTwo lines with 35 characters each

75 characters Two lines with

80 bytes each

Ad Extensions

Phone and location extensions with offer, product, social and app extensions

Phone and location extensions for select places, called ‘vCards’

Phone extensions for some qualified clients, but no location extensions

Sitelinks

Up to 4 sitelinks with 30 characters each

Up to 3 sitelinks, which altogether can’t exceed 66 characters

Up to 5 sitelinks in ads depending on title length

Bidding Keyword level Keyword level Keyword level

International Search Engines 12

71

by Philip Miles, Director, DoubleClick UK, Google

Audience Planning in Search – Enhance Keywords with Audience Signals

13

Sometimes the same keyword can mean different things for different users. For example, a user searching for “laptops” close to your store is probably looking for a different answer than a user searching for the same keyword in her office.

Learnings from programmatic displayWhile programmatic display has learnt a lot from search marketing with regards to auctions and bidding, it has also successfully discovered how to use audience signals. In fact, a recent study from the Boston Consulting Group showed that using advanced audience targeting techniques can reduce CPA by 32% on average for display campaigns.1 What if search could learn from display and improve campaigns by enhancing keywords with other signals?

Inspired by programmatic display, this chapter covers how you can enhance keywords with other signals to maximise relevance and performance through three simple steps:

1. Gather audience signals that matter

2. Target based on discovered insights

3. Engage with a tailored ad

A framework to plan more relevant campaignsAudience signals such as location, time of day, device and Remarketing Lists for Search Ads can help advertisers serve a more relevant search ad to their customers. But having to deal with more than keywords can increase complexity.

It helps to have a framework to use in order to decrease such complexity. We lay out three simple steps to succeed: Gather, Target, Engage. Here are the main steps in more detail.

13

74

GATHEREN

GAGE TARGET

Android users

Purchased laptop

Close to store-10% To Buy A Laptop Now

www.laptops.com

Get -10% If You Come In

Store In the Next 30'

Call

Close to store

The Chromebook Laptopwww.google.com/chromebook

Autosync Your Laptop

And Your Android Mobile

Android users

Need A Laptop Case?

www.dslaptops.com/cases

Look At Our Five Star

Reviews For Laptop Cases!

Purchased laptop

WHO

WHAT

INTERESTS

AUDIENCE SIGNALS THAT MATTER

BASED O

N IN

SIGH

TS

WITH A TAILORED AD

laptop

BR

AND

PURCHASE CONSUMEDCONTENTHISTORY

FAMILI

ARITY

Audience Planning in Search 13

75

1. Gather audience signals that matterThis step is about identifying relevant signals for use in your search campaigns - an essential step for an advertiser to develop insights on who its audiences are, what context they are in and what their interests may be. Here are a few examples of types of signals you can identify:

Who: What is the user’s relationship with you? For example: Have they ever visited the site? Have they ever purchased?

What: Time, location and device used.

Interests: Interests in specific categories based on consumed content.

2. Target based on discovered insightsSecond, you can combine those signals, creating separate segments based on the insights you discover. Each segment can represent a different campaign. For example, here are a few different segments you could create if you were selling laptops:

a. “Close to store”, based on device and location signals.

b. “Android users” might be more inclined to get a Chromebook.

c. “Interested in bags”: If a user has bought a laptop through your website, he might now need a laptop case rather than a laptop.

13

76

3. Engage with a tailored adIn this step, you will need to match each segment with a tailored creative message. Messaging can be optimised for each segment through A/B testing to achieve best results. Post engagement, and through measuring results, you can reassess if there are any new signals you may be able to gather, returning back to the first step.

Brands at the cutting edgeSearch marketing is a trusted channel among marketers because it conveys intent information. Still, marketers who do not enhance keywords with audience signals are missing out on results. Our advice would not be to change everything you currently do with search marketing, but you should start enhancing keywords with audience signals now via a test-and-learn approach.

After all, as one of our clients noted, “Today it’s nascent and because only brands at the cutting edge are leaning into this, you’re not going to lose out if you take it or leave it. But in six months’ time, this isn’t something you can’t be doing.”2

Sources

(1) Average performance noted as per “Adding Data, Boosting Impact: Improving Engagement and Performance in Digital Advertising”, Boston Consulting Group, September 2014

(2) “What opportunities are there in search marketing with audience signals?”, DoubleClick, February 2015

Audience Planning in Search 13

77

by Johanna Deschamps, Regional Marketing Manager - UK and Southern Europe, Marin Software

Jargon Buster

14

14

The Ultimate Search Acronym Buster

AOV Average Order Value: A customer’s average order value is the customer’s total order value divided by the customer’s total number of orders. The cost of items ordered, any taxes and shipping/handling fees added to orders, and any discounts applied to orders, are generally included in the average order value.

API Application Program Interface: A set of programming instructions and standards for accessing a web-based software application or web tool. A software company releases its API to the public so that other software developers can design products that are powered by its service. For example Marin Software uses the APIs of publishers such as Google, Yahoo, Bing and Facebook to build their ad management tools on top of.

ATD Agency Trading Desk: A service-based hub that serves as a managed service layer, normally above a licensed demand-side platform (DSP) and other audience buying technologies. They manage programmatic, bid-based media and audience buying. Operates as an agency’s internal “centre of excellence,” working with agency teams wanting to tap into programmatic buying for their clients. Examples include Accuen (Omincom), Audience On Demand (Publicis) and Xaxis (WPP).

CLV Customer Lifetime Value, sometimes known as Lifetime Value (LTV): A prediction of all the value a business will get from their entire relationship with a customer. As they don’t know how long each relationship will be, they make a good estimate and state CLV as a periodic value — Usually stating “this customer’s 12-month CLV is £x”.

CMS Content Management System: An application allowing publishing, editing and modifying content, organising, deleting as well as maintenance from a central interface. These systems provide procedures to manage workflow in a collaborative environment. Examples include Drupal, Joomla and WordPress.

CPA Cost Per Acquisition/Action: A pricing model where the advertiser pays for each specified conversion - for example a contact request, newsletter sign up, registration or purchase.

CPC Cost Per Click: A pricing model where the advertiser pays for a click on their advert, typically used in paid search and programmatic display.

CPL Cost Per Lead: A pricing model where the advertiser pays for a lead a publisher provides. It’s used interchangeably with CPA, however, CPL is more specific as it focuses on driving enquiries which aren’t immediately converted to paying customers. Typically found in the B2B and finance verticals.

CPM Cost Per Thousand: A pricing model, mostly found in brand focused display advertising, where the advertiser pays for a thousand impressions or views of their advert.

CRO Conversion Rate Optimisation: The method of creating an experience for a website or landing page visitor with the goal of increasing the percentage of visitors that convert into customers. Similar to Customer Experience (CX) or User Experience (UX) but with a specific focus on driving conversions.

CTR Click-Through Rate: A metric to judge the success of an online advertising campaign or creative based on the ratio of clicks to impressions.

CX Customer Experience: All the interactions a person has with your brand. It might be measured in: overall experience, likelihood to continue using, and likelihood to recommend to friends. User Experience (UX) is part of a broader CX, but CX contains some aspects outside of a product that UX does not.

DKI Dynamic Keyword Insertion: A feature provided by Google Adwords, Microsoft adCenter and Yahoo! Search Marketing to dynamically insert a keyword that is most relevant to the ad to be displayed. It helps advertisers to use relevant keywords on their ads while making it easier for them to create unique ads for a large number of keywords.

80

Jargon Buster 14

DMP Data Management Platform: A data warehouse that ingests, structures and stores information, and exports it in a way that marketers, publishers and other businesses can action to optimise advertising efforts. Commonly used to integrate audience or demographic information with paid search advertising. Examples include Aggregate Knowledge, BlueKai and X+1.

DSP Demand Side Platform: A software platform allowing buyers of display advertising inventory to manage multiple ad exchange and data exchange accounts through one interface. Examples include DataXu, MediaMath and Turn.

UI User Interface: The interface where interactions between humans and machines occur.

MVT Multivariate Testing: A process by which more than one component of a website may be tested in a live environment. A/B tests are usually performed to determine the better of two content variations; multivariate testing can theoretically test the effectiveness of limitless combinations. The only limits on combinations are the time it will take to get statistically valid samples of visitors and computational power.

PPC Pay Per Click: Similar to Cost Per Click (CPC), PPC is a pricing model where an advertiser pay the publisher when their ad is clicked. A commonly used acronym to describe Paid Search on Bing, Google and Yahoo.

PPV Pay Per View: Sometimes called Cost Per View (CPV) this is a pricing model where an advertiser pays every time a consumer views their video advertisement.

QS Quality Score: An estimate of the quality of your paid search ads, keywords and landing page. Higher quality ads can lead to lower prices and better ad positions.

ROI Return On Investment: The rate of revenues received for every pound invested in an item or activity. In a marketing sense, knowing the ROI of your advertising and marketing campaigns helps you to identify which techniques are most effective in generating income for your business. For example, if you spend £1,000 per month on PPC and generate £2,000 in revenues from your PPC campaign, your ROI is £2,000 but is more often referred to as a ratio such as £2 or 2 to 1.

RTB Real-Time Bidding: When ad inventory is bought and sold on a per-impression basis, via an instantaneous programmatic auction. Advertisers bid on an impression and, if the bid is won, the buyer’s ad is instantly displayed on the publisher’s site.

SaaS Software as a Service: A software model in which software is licensed on a subscription basis and centrally hosted. It is sometimes referred to as “on-demand software”. SaaS is typically accessed by users via a web browser.

SEM Search Engine Marketing: The marketing of a website on search engine results pages through optimisation and advertising. SEM encompasses Search Engine Optimisation (SEO) and Pay Per Click (PPC) listings.

SEO Search Engine Optimisation: The process of affecting the ranking of a web page in a search engine’s “natural” or un-paid (“organic”) search results. Using on-page optimisation and off page influence practitioners look to drive a page up the rankings to drive more relevant traffic to a website.

SERP Search Engine Results Page: The listing of results returned by a search engine in response to a keyword query. The results normally include a list of items with titles, a reference to the full version, and a short description showing where the keywords have matched content within the page.

SSP Supply-Side Platform: A supply-side platform is a software platform used to sell advertising in an automated fashion. SSPs are mostly used by online publishers to help them sell display, video and mobile ads. Whilst Demand Side Platforms (DSPs) are for advertisers to buy ad impressions from exchanges cheaply and efficiently, SSPs are for publishers to do the opposite, by maximising the price of their ad impressions.

UX User Experience: Deals with people interacting with your product and the experience they receive from that interaction. UX is measured with metrics like: success rate, error rate, abandonment rate, time to complete task, and clicks to completion. It’s part of Customer Experience (CX).

81

The production of this handbook has been made possible by the Search Council. Its members have produced the majority of the content and sponsored the production.

Special thanks to our sponsor Tug

AbakusAgenda 21 Marin SoftwareMicrosoftOban DigitalPHDSizmekStarcom MediaVest GroupTugUnique Digital

For more information or to participate in the Council please contact Steve@iabuk.net or visit www.iabuk.net/disciplines/search-marketing

April 2015

Internet Advertising Bureau 14 Macklin Street London WC2B 5NF

0207 050 6969 info@iabuk.net @iabuk

iabuk.net