Lean Startup Concepts Spring 2016 - Carnegie Mellon University
Lean startup concepts
description
Transcript of Lean startup concepts
Lean startup concepts
WHAT A STARTUP IS“Existing companies execute a business model; startups search for one.”
Why not a business plan?
• Business plans are full of untested assumptions and rarely survive first contact with customers
• Nobody, aside from venture capitalists and the former Soviet Union, requires five-year plans to forecast a series of unknowns.
• Startups are not smaller versions of large companies. Successful startups go quickly from failure to failure while adapting, testing new iterations, and improving ideas with continual feedback from customers
The lean launch pad approach
Key principles
1. Instead of engaging in months of research, entrepreneurs recognize that all they have is a series of untested hypotheses – good guesses
2. Lean startups “get out of the building” to test their hypotheses and collect evidence about whether they are true or false – this is customer development
3. Lean startups practice agile development by working iteratively and incrementally with the customer – this process leads to the minimum viable product
(a) Favorite quote
"My goal in advocating a scientific approach to the creation of startups is to channel human creativity into its most productive forms, and there is no bigger destroyer of creative potential than the misguided decision to persevere.“- Eric Ries, The Lean Startup (2011: p 149)
Motivation and intent (long version)
• Startups can shorten their product development cycles by adopting a combination of business-hypothesis-driven experimentation, iterative product releases, and "validated learning".
• If startups invest their time into iteratively building products or services to meet the needs of early customers, they can reduce the market risks and sidestep the need for large amounts of initial project funding and expensive product launches and failures.
Motivation and intent (shorter version)
• Build – Test – Learn • Build products (services) iteratively to meet
the needs of early customers – reduce market risks and funding needed to find successes and failures.
• Test hypotheses about customer behavior• Learn from their reactions to your product
(service)
Terms from Lean Startup
• Innovation accounting• Minimum viable product• Continuous deployment• Cohort analysis• Split testing• Actionable metrics• Pivot
Minimum viable product
• "version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least effort."
• Goal of an MVP is to test fundamental business hypotheses (or leap-of-faith assumptions) and to jump start learning process as quickly as possible.
Continuous deployment
• A process “whereby all code that is written for an application is immediately deployed into production,” which results in a reduction of cycle times.
• Ries states that some of the companies he’s worked with deploy new code into production as often as 50 times a day.
Cohort analysis
• A split or A/B test is an experiment in which "different versions of a product are offered to customers at the same time.“
• Goal is to observe differences in behavior between the two groups and to measure the impact of each version on an actionable metric.
• A/B can also be performed in serial fashion where a group of users one week may see one version of the product while the next week users see another
Actionable metrics• Measures that can lead to informed business decisions and subsequent
action. • IMVU used “funnel metrics” that were critical to company growth:
– Customer registration– Download of application– Trial, repeat usage, purchase
• Contrast to 'vanity metrics' - measurements that give “the rosiest picture possible” but do not accurately reflect the key drivers of a business.
• For example, a company specializing in creating web based dashboards for financial markets might view the number of web page views per person as a vanity metric as their revenue is not based on number of page views.
Actionable metrics
Actionable metrics
Pivot
• a “structured course correction designed to test a new fundamental hypothesis about the product, strategy, and engine of growth.”
• E.g., When Groupon first started, it was an online activism platform called The Point.
• After receiving almost no traction, the founders opened a WordPress blog and launched their first coupon promotion for a pizzeria located in their building lobby
Lean Analytics
• Use data to build abetter business faster.
Some background on Lean, analytics, metrics, and segmentation.
Hotmailwas a
database company
Flickrwas going to be an
MMOTwitterwas a
podcasting company
Autodeskmade
desktop automation
Paypalfirst built for Palmpilots
Freshbooks
was invoicing for
a web design firm
Wikipediawas to be written by
experts only
Mitelwas a
lawnmower company
Most startups don’t know what they’ll be when they grow up.
Possible problem space
Product/market
hypothesisTrial
startup
Product/market
hypothesisTrial
startup
Product/market
hypothesis
Trial startup
Product/market
hypothesis
Trial startup
You are
herePIVOT
Kevin Costner is a lousy entrepreneur.
• Don’t sell what you can make.Make what you can sell.
5 things you need to know about metrics
Qualitative or QuantitativeExploratory or ReportingVanity or ActionableCorrelated or CausalLeading or Lagging
Qualitative Quantitative
Unstructured, anecdotal, revealing, hard to aggregate.
Numbers and stats; hard facts but less insight.
http://www.flickr.com/photos/zooboing/8388257248/ http://www.flickr.com/photos/x1brett/4665645157/
Warm and fuzzy. Cold and hard.
Exploratory ReportingSpeculative, trying to find unexpected or interesting insights.
Predictable, keeping you abreast of normal, managerial operations.http://www.flickr.com/photos/elwillo/4737933662/http://www.flickr.com/photos/50755773@N06/541529
5449/
Donald Rumsfeld on analytics
(Or rather, Avinash Kaushik channeling Rumsfeld)
Things we
know
don’tknow
we know
we don’tknow
we know
we don’tknow
Are facts which may be wrong and should be checked against data.
Are questions we can answer by reporting, which we should baseline & automate.
Are intuition which we should quantify and teach to improve effectiveness, efficiency.
Are exploration which is where unfair advantage and interesting epiphanies live.
Vanity
Makes you feel good, but doesn’t change how you’ll act.
Picks a direction.
Actionable
http://www.flickr.com/photos/lostseouls/807253220/ http://www.flickr.com/photos/aussiegall/6382775153/
Hits A metric from the early, foolish days of the Web. Count people instead.
Page views Marginally better than hits. Unless you’re displaying ad inventory, count people.
Visits Is this one person visiting a hundred times, or are a hundred people visiting once? Fail.
Unique visitors This tells you nothing about what they did, why they stuck around, or if they left.
Followers/friends/likes
Count actions instead. Find out how many followers will do your bidding.
Time on site, or pages/visit
Poor version of engagement. Lots of time spent on support pages is actually a bad sign.
Emails collected
How many recipients will act on what’s in them?
Number of downloads
Outside app stores, downloads alone don’t lead to lifetime value. Measure activations/active accounts.
Correlated CausalTwo variables that change in similar ways , perhaps because they’re linked to something else.
An independent factor that directly impacts a dependent one.
Summer
Ice creamconsumption
DrowningCorrelatedCau
salCausal
Causality is a superpower, because it lets you change the future.
Correlation lets you predict the future
Causality lets you change the future
“I will have 420 engaged users and 75 paying customers next month.”
“If I can make more first-time visitors stay on for 17 minutes I will increase sales in 90 days.”
Find correlation
Test causality
Optimize the causal
factor
Leading LaggingNumber today that shows metric tomorrow—makes the news.
Historical metric that shows how you’re doing—reports the news.
A leading indicator for e-commerce
Acquisition
Hybrid
Loyalty
70%of retailers
20%of retailers
10%of retailers
You are just like
How many of your customers buy a second
time in 90 days?
Once
2-2.5per year
>2.5per year
Your customers will buy from
youThen you are in
this mode
1-15%
15-30%
>30%
Low CAC, high
checkout
Increasing returns
Loyalty, inventory expansion
Focus on
(Thanks to Kevin Hilstrom for this.)
Segments, cohorts, A/B, and multivariates
Segment:Cross-sectional comparison of
all people divided by some
attribute (age, gender, etc.)
☀
☁
Cohort:Comparison of similar groups along a timeline.
A/B test:Changing one thing (i.e. color) and measuring the result (i.e. revenue.)
MultivariateanalysisChanging several things at once to see which correlates with a result.
☀☁☀☁
Why use cohorts? Here’s an example.
January February March April May
Rev/customer $5 $5 $4 $4 $5
Is this company growing
or stagnatin
g?
Cohort 1 2 3 4 5
January $5 $3 $2 $1 $1
February $6 $4 $2 $1
March $7 $6 $5
April $8 $7
May $9
How about now?
Why use cohorts? Here’s an example.
Cohort 1 2 3 4 5
January $5 $3 $2 $1 $1
February $6 $4 $2 $1
March $7 $6 $5
April $8 $7
May $9
Averages $7 $5 $3 $1 $1
Look at the same data in cohorts