GO BEYOND CREATIVE & BUDGET TO DRIVE ROI Audience Measurem… · though we “stacked the deck”...
Transcript of GO BEYOND CREATIVE & BUDGET TO DRIVE ROI Audience Measurem… · though we “stacked the deck”...
GO BEYOND CREATIVE & BUDGET TO DRIVE ROI
BRIAN LAVERTYSr. ManagerConsumer Insights Research
SUNIL SOMANSr. DirectorMeasurement Innovation Research
…Creative remains the undisputedchamp in terms of sales drivers…1
1 Nielsen, “When it Comes to Advertising Effectiveness, What is Key?” Oct 10, 2017. http://www.nielsen.com/us/en/insights/news/2017/when-it-comes-to-advertising-effectiveness-what-is-key.html
IDENTIFYING DRIVERS OF ROI
Explore traditional and nontraditional drivers of ROI:
1. Key Brand Measures 2. Implementation Variables
Perform meta-analysis of previous MMMs and ROI analysis
for multiple brands
Analyze drivers by category vertical, including non-CPG
RESEARCH PARTNER METHODOLOGY
• Sourced from Nielsen mix
• Non-CPG & CPG clients:11 categories, 130+ brands from 2015-2017
• BRAND-level ROAS, including all execution
• Structural & executional analysis
• TV IMPACT and ROI drivers for brand
• Sourced from ROI Genome, database of 20,000 studies
• Focused on executional and implementation variables
• Have created best practices and learnings
• Additional multi-platform campaign analysis
• Sourced from KMB Global Link database of 20,000 ads as well as case studies
• Deep-dive on creative drivers
• Ability for creative to break through
• Executional factors
• We applied a Gradient-Boosted Regression Tree model to identify drivers of ROAS
• Each tree in the equation creates splits in our dependent variable (ROAS) based on significant differences in our independent variables (Long-Term & Short-Term factors)
• GBM models identify the factors most important to driving TV ROAS relativeto other variables tested in the data set
MODELING APPROACH
Boosted Regression Trees
Simple Decision Tree
• 11 categories
• 130+ brands
• 2015 - 2017
Category # of Brands
Household cleaners & goods 8
Cosmetics & beauty products 11
Personal hygiene & health 15
Medicine & pharmaceutical 14
Confectionery & snacks 12
Food 34
Beverages 22
Consumer financial services 4
Communications & technology 4
Automotive 6
Retail 3
NIELSEN MMM SAMPLE
FACTORS ACCOUNTING FOR 1%+ OF ROAS VARIATION
Structural
Brand Size
Brand Dollar
Share
Brand Media
Incrementality
Executional (Creative)
Duration
Copy Quality
Executional (Media)
Daypart
Distributor
Reach
Frequency
Position
Pod Placement
Ad Clutter
DEFINITIONS
Variable Tested Description Source
TV ROAS TV Return on Advertising (Revenue/Spend) Nielsen MMM
Brand Dollar Share % Revenue Share of Segment Nielsen Ad Intel
Brand Media Incrementality % Revenue Attributed to Media Nielsen MMM
Brand Size $M Revenue Nielsen MMM
Duration % Execution as 30 Seconds Nielsen Ad Intel
Ad Quality % Memorability and Likeability Nielsen TV Brand Effects
Ad Clutter Number of Ads Per Minute of Programming Nielsen Ad Intel
Pod Placement % Execution in First Pod of Program Nielsen Ad Intel
Position Placement % Execution in First Position of Pod Nielsen Ad Intel
Distributor % Execution in Cable Distributor Nielsen Ad Intel
Daypart % Execution in Prime Time Nielsen Ad Intel
Frequency Average Monthly Household Frequency Nielsen NPower
Reach Average Monthly % Household Reach Nielsen NPower
58% 55%
17%11%
24% 33%
Structural Long-Term Short-Term Creative Short-Term Media
CPG NON-CPG
RELATIVE INFLUENCE OF FACTORS ON TV ROAS
Short Term
Long Term
Less than 50% of ROAS are controllable in the short term through Short-Term shifts
Short-Term
Long-Term
54% 56% 55% 56% 57% 55% 58% 57% 62% 56% 58%
12% 11% 11% 9% 11% 12% 10% 20% 15% 20% 15%
34% 32% 34% 35% 32% 33% 32% 23% 23% 24% 27%
Structural Long-Term Short-Term Creative Short-Term Media
CPG NON-CPG
Beverages Conf. & Snacks
Cosmetics & Beauty
Food HH Cleaners
Medicines & Pharma
Personal Hygiene
Tech & Comms
Auto Retailer Financial Services
RELATIVE INFLUENCE BY CATEGORY
SIZE OF BRANDS’ BUSINESS AND IMPACT ON POTENTIAL ROAS
$1.07
$2.20$2.57
$5.03
< $300 M $300 -$600 M
$600 -$1,300 M
> $1,300 M
YEAR REVENUE OF BRAND
Average TV ROAS by Brand Size Quartile
SALES ($M) AD SPEND ($M) AD LIFT
SMALL BRAND $50,000 $5,000 10%
LARGE BRAND $200,000 $5,000 5%
REVENUE ($M) REVENUE/ SPEND ($M) ROAS
SMALL BRAND $5,000 $5,000/$5,000 $1.00
LARGE BRAND $10,000 $10,000/$5,000 $2.00
For example – if we spent the same amount on advertising, and even though we “stacked the deck” by assigning the Small Brand a greater percentage sales lift (10% vs. 5%), the Large Brand realized a greater ROAS on its advertising investment. A large brand has the ability to leverage fixed costs of advertising across its existing business.
‘Thought Experiment’ to Demonstrate Brand Size Impact on Potential ROAS
5.2%
6.0%
3.0%
3.8%
4.3%
5.1%
4.7%
6.0%
6.5%
Duration
Ad Quality
Ad Clutter
Pod Placement
Position Placement
Frequency
Daypart
Distributor
Reach
Relative Influence on TV ROAS – CPG
4.3%
13.1%
2.0%
2.0%
1.8%
2.4%
2.9%
6.3%
6.9%
Duration
Ad Quality
Ad Clutter
Pod Placement
Position Placement
Frequency
Daypart
Distributor
Reach
Relative Influence on TV ROAS – NON-CPG
RELATIVE INFLUENCE ON TV ROAS
ACROSS CATEGORIES, EXECUTIONAL DRIVERS OF ROI VARY SIGNIFICANTLY
BEVERAGESCONFECTIONERY
& SNACKS
COSMETICS & BEAUTY
PRODUCTS FOOD
HOUSEHOLD CLEANERS &
GOODSMEDICINES &
PHARMACEUTICAL
PERSONAL HYGIENE &
HEALTH TECHNOLOGY &
COMMUNICATIONS AUTOMOTIVE RETAILER
CONSUMER FINANCIAL SERVICES
Reach 1 3 2 1 1 2 1 3 2 2 2
Ad Quality 2 2 1 7 3 1 6 1 1 1 1
Distributor 3 1 3 2 2 3 2 2 3 3 3
Duration 4 4 6 4 4 4 3 4 4 4 4
Frequency 5 6 4 3 5 5 5 7 6 9 6
Daypart 7 5 5 5 6 6 4 5 5 5 5
Position Placement 6 7 7 6 7 7 7 9 9 7 7
Pod Placement 8 8 8 8 8 8 8 8 7 6 9
Ad Clutter 9 9 9 9 9 9 9 6 8 8 8
INFLUENCE OF REACH BY CATEGORY
6.0% 6.1% 6.1% 6.4% 6.4% 6.6% 6.7% 6.8% 7.3% 7.4% 7.8%
Relative Influence of Reach on TV ROAS
Conf. & Snacks
HH Cleaners
Tech & Comm
Personal Hygiene
Medicines & Pharma
Cosmetics & Beauty
Auto Beverages FinancialServices
Retailer Food
CPG Norm = 6.5%
Non-CPG Norm = 6.9%
Driver Rank 1 1 2 1 2 2 2 3 2 2 1
Relative Influence -7% -5% -11% -1% -1% +2% -3% +5% +6% +7% +21%
Median Value +11% +26% -47% +5% +12% +7% +3% -8% +32% +19% -6%
BRANDS SHOULD MAXIMIZE COST-EFFECTIVE REACH
Case StudyTV Avg. Frequency Levels
By Reach Level
Reach Levels
Case StudyTV Communications
Cost of Reach
W18-34 Reach
INFLUENCE OF DISTRIBUTOR BY CATEGORY
5.0%5.8% 6.0% 6.0% 6.1% 6.3% 6.3% 6.6% 6.6% 6.8% 7.1%
Auto Medicines & Pharma
Food Beverages HH Cleaners
Personal Hygiene
Retailer Conf. & Snacks
Cosmetics & Beauty
Tech & Comm
Financial Services
Driver Rank 3 3 2 3 2 2 3 1 3 2 3
Relative Influence -20% -3% 0% +1% +2% +5% +0% +10% +6% +8% +12%
Median Value +45% -20% +5% +13% -6% 0% -2% +12% -7% -67% -45%
CPG Norm = 6.0%
Non-CPG Norm = 6.3%
ADVERTISERS SHOULD BALANCE CABLE/NETWORK MIX TO MAXIMIZE COST-EFFECTIVE REACH
TV AdvertisingCable vs. Network
ROIResponsive IndexCost Index
Cable Network
200
150
100
50
0
INFLUENCE OF AD QUALITY BY CATEGORY
4.1% 4.1%6.0% 6.1% 6.4% 6.5% 6.6%
10.3% 10.6%
15.7% 16.0%
Food Personal Hygiene
HHCleaners
Beverages Conf. & Snacks
Medicines & Pharma
Cosmetics & Beauty
Auto Financial Services
Tech & Comm
Retailer
Driver Rank 7 6 3 2 2 1 1 1 1 1 1
Relative Influence -31% -32% +1% +8% +2% +9% +11% -22% -19% +19% +22%
Median Value +18% -18% +55% -9% +27% +32% -41% -9% +55% -18% +23%
CPG Norm = 6.0%
Non - CPG Norm = 13.1%
MESSAGING & CREATIVE QUALITY ARE KEY – PARTICULARLY FOR VIDEO
Moving creative quality from Average to Good creates similar increased efficiency for both CPG and non-CPG products. However, further improving creative quality to Very Good does not generate as much lift in ROIs for CPG brands.
70100
123143
Poor Average Good Very Good
Impact of STSL on
ROIs – CPG
53
100137
169
Poor Average Good Very Good
Impact of STSL on
ROIs – Non-CPG
10% of Ads
26% of Ads
30% of Ads
32% of Ads
12% of Ads
41% of Ads
39% of Ads
10% of Ads
Based on ~3,500 Copy tested ads in Link Validation database
IMPACT OF CREATIVE SCORE ON SALES ROI’S VARIES BY CPG
VS. NON-CPG CATEGORY
CREATIVE QUALITY IS A MULTIPLIER FOR MEDIA EXPOSURE TO DELIVER ROI’S
Stronger ad requires lower media weight to achieve same in-market sales performance with good creative quality; it requires 18% lower GRPs to achieve similar ROIs with an average ad.
Average Sales ROIs by Creative Quality and Media Weight
11 studies across 514 ads – Link Validation DatabaseGood quality: 10% above category norm
BETTERCREATIVE
100ROI Index
90ROI Index
AVERAGECREATIVE
+18% GRP weight
125ROI Index
102ROI Index
GOODCREATIVE
+18% GRP weight
/ = GRP “weight”
AD REQUIRES SUFFICIENT AIRING TIME TO CREATE IN-MARKET EFFECT, BUT NEEDS TO AVOID BEING AIRED FOR TOO LONG WHEN DIMINISHING RETURN OCCURS
85
100
71 69
4-6 weeks 8 weeks 8-12 weeks >12 weeks
Inde
x of S
ales
ROI
Impact of # of Weeks on Sales ROIs
From 3 case studies – Link Validation Database
FLIGHTING PATTERNS MATTER
Testing a few flighting scenarios indicates Tapered flighting to yield 5% incremental sales ROIs compared to Uniform or On-Off 3-week flighting schedule.
Tapered flighting is optimal because it enables the copy to break through clutter with high GRPs in the early weeks and then sustains continued momentum weeks after due to ad stock effects.
However, the lowest weekly GRPs with Tapered flighting should remain above threshold level.
We would recommend using On-Off Tapered flights over 8-10 weeks for TV.
Based on one CPG and one Non-CPG study – ROI database
0
20
40
60
80
100
120
140
160
Uniform Tapered 10 OnOff3 Tapered8
Wee
kly
GRPs
On-Off 3 Tapered 8
Flighting ScenariosRaw GRPs
COPY WEAR-OUT CAN REDUCE ADVERTISING RESPONSE BY +30%
Air higher quality copies longer than low quality copies• High-quality commercials can be aired well-after wear-out has
begun and still achieve a higher response than lower-quality copies• Remember, airing beyond the point of wear-out still drives sales
and brand equity, just at a lower rate
Wear-out occurs as frequency becomes excessive• Broadly speaking, this is typically at around 2,500 GRPs for General
Market and 1,500 GRPs for Hispanic in the US, though good copy can be aired beyond that
• Target size and “purchase” cycle frequency determine how much support it will take to reach your target audience before wear-out occurs
Consider copy type when planning copy rotation• Wear-out typically occurs faster for ads focused on new items/initiatives• Seasonal ads with nostalgic or cherished holiday associations tend
to see lower wear-out than nonseasonal ads
Consider production costs• Re-airing stronger copies with moderate wear-out saves on production costs• Copy refreshes (subtle changes to main concept) may help minimize wear-out
Wear-out Principles
Creative Optimization goes beyond Copy Quality. In addition to making sure that creative resonates with the desired target segment, it is equally important to make sure that proper flighting is employed to maximize creative impact.
CPG and non-CPG. Long-term (brand characteristics) and short-term (controllable media) factors have similar influence on ROAS variation across both CPG and non-CPG brands with long-term accounting for an average of 56% for CPG and 58% for non-CPG. This suggests that ability to control short-term ROI at a higher level is comparable in both groups.
Short-term factors can drive ROAS. While brand Long-term factors account for more than half of variation of ROAS, opportunities exist for brands and categories to optimize ROAS 5+% through adjusting key factors such as ad quality (non-CPG) and reach (CPG) in the short term.
Category differences. ROAS variation is best explained by brand presence factors (including size, category share, media incrementality), not necessarily by the industry they play in. Focus on individual brand, not industry, for optimization opportunities.
SUMMARY