A3: Lean For Social Impact: Innovating for Greater Impact and Scale, Steve Nagai-Ma (1 of 2)
1 like791 views
The document outlines a 5-step Lean process for social innovation: 1) Define guideposts like target users and goals, 2) Propose 3-factor solutions addressing value, scale, and impact, 3) Develop success metrics, 4) Quickly test solutions through the Test-Learn-Respond loop, and 5) Track learnings through iterative testing. It provides an example of using this process to develop an English learning program for immigrants, iterating based on testing until a validated model was found that served over 1 million users at minimal cost. The key is embracing an experimental mindset to fail fast, learn from failures, and improve solutions.
A3: Lean For Social Impact: Innovating for Greater Impact and Scale, Steve Nagai-Ma (1 of 2)
- 1. Steve Nagai-Ma steve@flyingbird.us 1 Lean For Social Impact: Innovating for Greater Impact and Scale
- 2. Session Objectives: • Gain an understanding of how to use the Lean innovation process to achieve more scale and impact • Learn how other orgs used the Lean innovation process successfully • Begin applying the Lean innovation process to your own organization 2
- 3. Poll: What is your comfort level with risk? (Willingness to invest time & resources on projects that are new, unproven, and have a high chance of failure) 3 Comfortable with Risk Not comfortable with Risk Somewhere in the Middle
- 4. The Lean Startup/Lean Impact Process: • Lean is about reducing wasted effort, not frugality • Involves small scale tests, revising and improving our proposed solutions based on feedback from real users • Applies to new initiatives and/or improvements to existing programs 4
- 5. Focus: South Africa... 40% of young people not employed Early Learnings (Customer Discovery): • Young people didn’t know how to connect with jobs or what was expected of them • Employers had poor performing staff and high attrition 5
- 6. Initial Solution: Connect young people with jobs that match their skills. Get employers to pay for placements. Result: Few young people met the minimum requirements for any job. Iteration 2: Train people to be work ready. Result: Employers paid for trained recruits. But many employees quit because they weren’t prepared for the rigors of the job. Iteration 3: Train people in a way that simulated real life job situations. Result: Retention improved but many quit after a month (folks ran out of savings). 6
- 7. Iteration 4: Encourage employers to pay earlier. Result: Retention improved but not for people who travelled long distances. Iteration 5: Only match youth to jobs that are nearby. Result: Improvement for some but others were isolated or living in desperately poor towns. Iteration 6: Partner with employers that are not location specific (eg. international cruise lines). Result: Success! Youth placed in jobs they liked. Employers also satisfied. 7
- 8. Results: • Partnerships with 450 employers • Successfully placed more than 45,000 youth • Expanded to Rwanda 2022 Goal: Support 1.5 million work-seekers and place 500,000 (represents 5% of the target population) Current Priority: Scale & Sustainability (high cost per trainee is not entirely offset by employers) 8
- 9. “We had promised one set of deliverables and found ourselves delivering something else. Being very proactive with funders about why a current solution is not working and what new learnings are emerging from iteration is the key.” 9
- 10. “After a few different attempts, we discovered that when we simply provided a piece of fruit and a peanut butter sandwich, the test scores went up by a whopping 30%. Harambee has made over 1.2 million peanut butter sandwiches as a result of this accidental iteration.” 10
- 11. Step 1: Establish Guideposts • Define your target constituency • Describe the societal problem you are focused on solving • Set ambitious (but possible) goals for impact and scale (Note: Do not reference a solution at this point) 11 Steve Nagai-Ma steve@flyingbird.us
- 12. 12 Target constituency: ~17 million Spanish-speaking, working-age, immigrants in the US Problem: Spanish-speaking immigrants face a number of challenges including lower incomes, housing insecurity, and health inequities Transformational Goal: Provide services that improve the lives of more than 1 million immigrants and activate them to drive policy changes
- 13. 13 Think about your organization… • Who is in your target constituency? • What broad problem are you trying to solve? • What are your ambitious, specific goals for scale and impact?
- 14. Step 2: Propose Solutions With your constituents in the room (or with their direct feedback), come up with 3- factor solutions Value Scale How will it grow & sustain? Do people want it? 14 Steve Nagai-Ma steve@flyingbird.us Impact What positive change will result?
- 15. Do people want it? Does it solve a real problem for your users or enhance their lives? Who would pay for it? Value 15
- 16. Can it scale & sustain? How will reach your target constituents and how will your innovation pay for itself at scale? Scale 16
- 17. Sustainable Revenue Sources • Paying Customers • Cross Subsidy • Freemium • Optional Payments from Users/Pay what you can • Referral Fees • Government Policy 17
- 18. Does it make the world better? Will your innovation actually deliver lasting positive impact? Are there unintended consequences? Impact 18
- 19. Proposed 3-factor solution: Offer a digital, English learning program that works as well (or better) than traditional classes but costs significantly less 19
- 20. 20 Think about your organization… • Come up with 3-factor solutions that can potentially offer value, scale, and impact (these can be improvements on what you currently offer). NOTE: Ideally, the process of developing solutions should happen with your constituents
- 21. Step 3: Develop Success Metrics Define what success looks like for your solution through numerical goals around value, scale, and impact. When developing key metrics, focus on per user rates rather than gross numbers 21
- 22. • Conversion rate • Engagement rate • Retention/Renewal Rate • Viral Growth Rate • Cost per person served (includes all operational & acquisition costs) • Revenue per person • What positive impact will our solution make? • How deep, wide, and lasting will the impact be? Value ImpactScale Metrics That Matter 22
- 23. 23 Sample Success Metrics for Value: • When offered, 50% of immigrants will subscribe to free lessons from Revolution English • 25% of subscribers will pay for lessons from Revolution English
- 24. Sample Success Metrics for Scale: • People will pay enough to cover 100% of costs (includes content production costs, marketing, and collecting payments) 24
- 25. Sample Success Metrics for Impact: • Learners who complete 3 months of Revolution English will improve their English as much as a high-quality, 3-month class 25
- 26. 26 Think about your organization… • Come up with success metrics for value, scale, and impact for your proposed solution.
- 27. Metrics are less risky when they are: • Co-designed with constituents/based on customer discovery • Based on relevant research, and/or • Similar to a comparable analog that is already successful 27
- 28. Step 4: Test/Learn/Respond Identify your solution’s riskiest success metric, then quickly move through the test, learn & respond loop. 28 Steve Nagai-Ma steve@flyingbird.us TEST LEARN RESPOND
- 29. Build a Minimum Viable Product (MVP) to quickly assess whether you can achieve your riskiest success metric. 29 TEST
- 30. Key Metrics to Test: • People will pay enough to cover 100% of costs (includes content production costs, marketing, and collecting payments) Testing Plan Run $150 worth of Facebook ads. Send 10 days of content to learners. Try to convert people into paying customers. 30
- 31. Get target users to engage with, use, and/or buy your MVP. Measure how they behave. 31 Steve Nagai-Ma steve@flyingbird.us LEARN
- 32. Test Results: 32 Pitched 156 learners to sign up for paid tutoring sessions. • 53 said they wanted to sign up • 49 got to the online payment processing page • 0 paid We contacted 21 folks who didn’t pay • All 21 seemed legitimately interested in paying • 12 said they didn’t have a credit card • 9 said they were uncomfortable paying online
- 33. Respond to your data in one of three ways: • ITERATE: make improvements based on what you learn, then Test/Learn/Respond again • PIVOT: move on to a new solution because this one won’t fly • SCALE UP: do this when you’ve hit all of your success metrics 33 RESPOND
- 34. Test Results: 34 Based on these data, would you iterate, pivot, or scale up? Pitched 156 learners to sign up for paid, live lessons. • 53 said they wanted to sign up • 49 got to the online payment processing page • 0 paid We contacted 21 folks who didn’t pay • All 21 seemed legitimately interested in paying • 12 said they didn’t have a credit card • 9 said they were uncomfortable paying online
- 35. When you engage in the Test/Learn/Respond loop… • Move quickly (your success rate is correlated with the speed at which you Test/Learn/Respond) • Learn what the solution is through experiments, data, and actual user behavior 35 Steve Nagai-Ma steve@flyingbird.us
- 36. 36 Think about your organization… • Describe a quick test to determine whether your riskiest success metric is achievable.
- 37. Step 5: Innovation Accounting For each iteration of your solution, document what you tested, your learnings, and what iteration(s) you plan to test next. 37 Guideposts • Constituents • Problem • Goal Solution Description of proposed solution Metrics • Value • Scale • Impact Test 1 • Metric to test • MVP needed Results Response What was learned? How will you iterate? Test 2 • Metric test • MVP needed Results Response What was learned? How will you iterate? Test 3 • Metric to test • MVP needed Results Response What was learned? How will you iterate? Test 4 • Metric to test • MVP needed Results Response What was learned? How will you iterate?
- 38. Keys to Innovation Accounting • Create a dashboard that makes it easy to see your progress • Gather reliable quantitative data on actual user behavior • Gather qualitative data to shed light on your numbers • Run A/B tests and do some math • Track your learnings so each test builds on what you learned before 38
- 39. 39 Iteration 1: Generate revenue through ads Test results: Projected revenue/ user: $0.43 Iteration 2: Improve on- boarding experience Test results: Projected revenue/ user: $0.57 Iteration 3: Offer 3 lessons per day Test results: Projected revenue/ user: $0.73 Iteration 4: Include interactive flipcards Test results: Projected revenue/ user: $1.06 Iteration 5: Optimize format/design of lessons Test results: Projected revenue/ user: $1.18 Key Metric: Generate $1.25 per user (average over the lifetime of the user) Iteration 5: Identify best content and offer it early Test results: Projected revenue/ user: $1.28 * These numbers are not exact
- 40. Before Innovation Students Served Per Year: 1,200 Cost covered by philanthropy: $790,000 After Innovation Students Served Per Year: 1,000,000+ Cost covered by philanthropy: $0 The impact of both programs is a significant increase in English language proficiency 40 Steve Nagai-Ma steve@flyingbird.us
- 41. 1) Define your guideposts 2) Propose 3-factor solutions 3) Develop numerical success metrics 4) Quickly move through the Test, Learn & Respond loop 5) Track your learnings until you’ve validated your model 5-10% of the work 90-95% of the work 41 Lean for Social Innovation: The 5-Step Process
- 42. A Note About FailureA Note About Failure Expect that your proposed innovation will fail. When you fail… • Fail fast • Fail forward (learn from failures to get you closer to a successful model) 42
- 43. Key Takeaways: • Propose solutions that offer value, scale, & impact (at the same time) • Love your problem, not your solution • Embrace an experimental mindset and iterate based on what you learn • Rate of success is correlated with speed of testing and learning 43
- 44. Think of a solution Build, launch & scale up your solution Raise funds to build that specific solution Pray that your solution works Identify a problem Raise funds to experiment with various solutions Test potential solutions The Traditional “Waterfall” Approach The Lean for Social Impact Approach Keep deploying solution despite flaws Improve your best solution based on learnings Scale your validated solution to maximize impact