PadhaikoAll subjects

Lean Startup, Design Thinking and Product Development

IT-Based Innovation and Entrepreneurship (ENCT 436)

Open in the Padhaiko app

5.1 Lean Startup methodology

Formulated by Eric Ries (2011) and grounded in Steve Blank's Customer Development framework, the Lean Startup methodology treats an early-stage venture not as a smaller version of a large company executing a fixed plan, but as a temporary organization searching for a repeatable, scalable business model under conditions of extreme uncertainty.

Definition

Lean Startup Methodology (Eric Ries)

Lean Startup is a scientific approach to creating and managing startups that accelerates validated learning by testing business hypotheses through rapid, minimum-effort product experiments, actionable metrics, and iterative pivots rather than elaborate upfront business plans and full-featured launches.

Five Core Principles of Lean Startup

  1. Entrepreneurs Are Everywhere: Any team building a new product under extreme uncertainty (whether in a garage or an enterprise R&D lab) is a startup.
  2. Entrepreneurship Is Management: Uncertainty requires a distinct management discipline focused on discovery rather than static variance-to-budget tracking.
  3. Validated Learning: Progress is measured by empirically proven truths about what customers actually want and pay for—not by lines of code written or features shipped.
  4. Build–Measure–Learn: Turn ideas into products, measure real user behavior, and decide whether to pivot or persevere.
  5. Innovation Accounting: Replace vanity metrics (total page views, app downloads) with cohort-based actionable metrics (activation rate, retention, referral, revenue).

5.2 Build–Measure–Learn cycle

The Build–Measure–Learn feedback loop is the operational engine of Lean Startup. Crucially, while execution flows Build Measure Learn, planning works in reverse (Learn Measure Build): first decide what hypothesis you need to learn, then define the metric that tests it, and finally build the smallest artifact capable of generating that measurement.

Phases of the Build–Measure–Learn Feedback Loop
PhasePrimary GoalConcrete IT Artifacts & Actions
1. BuildCreate the minimum experiment needed to test a leap-of-faith assumptionLanding page, clickable Figma prototype, concierge workflow, or single-feature MVP
2. MeasureCapture objective behavioral data using cohort metrics rather than opinionsFunnel conversion rates, activation rate, Day-7/Day-30 retention, A/B test lift
3. LearnCompare observed metrics against pre-set success thresholdsPersevere (optimize and scale) or Pivot (change one core hypothesis)

5.3 Hypothesis-driven entrepreneurship

Instead of treating a startup idea as an unquestioned blueprint, hypothesis-driven entrepreneurship breaks the venture down into explicit, falsifiable assumptions tested in order of highest risk:

  1. Value Hypothesis: Does the product actually deliver value to target customers once they try it? (Measured by engagement, retention, and willingness to pay.)
  2. Growth Hypothesis: How will new customers discover and adopt the product sustainably? (Viral referrals, paid acquisition with , or SEO/partnerships.)

Anatomy of a Falsifiable Experiment Card


5.4 Design Thinking

Popularized by IDEO and Tim Brown (2009), Design Thinking is a human-centered, iterative innovation process that balances desirability (human need), feasibility (technical capability), and viability (business sustainability).

The Five Stages of Design Thinking
StageMindsetCore Activities
1. EmpathizeObserve without judgmentField shadowing, contextual inquiry, user interviews, empathy maps
2. DefineSynthesize insights into focusAffinity clustering, user personas, Point-of-View (POV) problem statements
3. IdeateDivergent explorationBrainstorming, Crazy 8s sketching, "How Might We..." (HMW) prompts
4. PrototypeBuild to think & learn cheaplyPaper sketches, wireframes, interactive Figma flows, role-play mockups
5. TestGather real user reactionsUsability testing, task completion observation, refining problem/solution understanding

5.5 Customer discovery and validation

Aligned with Steve Blank's Customer Development model, startups go through discovery and validation before building full sales teams:

  1. Customer Discovery: Translate founder hypotheses into concrete briefs, get "out of the building" to test whether the problem exists and resonates with early evangelists.
  2. Customer Validation: Verify that a repeatable, scalable sales and onboarding process works—proven by paying early customers rather than free pilot favors.
  3. Transition Check: Only after Customer Validation succeeds does the startup move to Customer Creation (scaling demand) and Company Building (formalizing departments).

5.6 Minimum Viable Product (MVP)

Definition

Minimum Viable Product (MVP)

A Minimum Viable Product (MVP) is that version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least engineering effort and time. An MVP is not a buggy, broken half-product; it is a focused, reliable slice of the core value proposition.

AspectTraditional Full ReleaseMinimum Viable Product (MVP)
ScopeComplete feature checklist built upfrontSingle core workflow solving the primary customer pain
Launch TimingMonths or years after initial ideaDays or weeks after problem validation
Primary GoalDeliver finished product assuming demand existsTest demand, pricing, and core usability with real users
Failure CostHigh sunk engineering and capital lossLow sunk cost; rapid pivot capability

5.7 MVP development strategies for IT products

IT founders choose different MVP patterns depending on whether the main risk is demand, workflow complexity, or algorithmic feasibility:

Common MVP Strategies for IT Products
MVP TypeHow It WorksBest Used When
Smoke Test / Landing Page MVPA crisp value-proposition webpage with pricing and a "Start Trial / Pre-Order" button measuring click-through and sign-up intentTesting demand and pricing messaging before writing backend code
Explainer Video MVPA short 2–3 minute demo video simulating how the product will work (famously used by Dropbox)Complex infrastructure or cross-platform sync is expensive to build before proving interest
Concierge MVPFounders manually deliver the exact end-to-end service behind the scenes for the first 5–10 customersLearning subtle customer edge cases before automating workflows
Wizard of Oz MVPFront-end looks like an automated AI/software app, while humans quietly perform processing behind the scenesTesting whether users trust and adopt an AI/matching output before training expensive models
Single-Feature MVPWorking software executing only the single most critical job-to-be-done cleanlyCore technical execution must be experienced directly (e.g., audio tool, developer API)

5.8 Prototype development and user testing

A prototype tests usability and design assumptions (low-fidelity sketches mid-fidelity clickable wireframes high-fidelity interactive mockups), whereas an MVP tests real-world market and business-model viability with live users.

Structured User Testing Protocol

  1. Task-Based Scenarios: Give test users realistic goals ("Book a follow-up lab test for Friday morning") without coaching them where to click.
  2. Think-Aloud Observation: Ask users to narrate expectations and hesitations while interacting with the prototype.
  3. Quantitative Usability Signals: Track task completion rate, time on task, error/backtrack count, and System Usability Scale (SUS) score.

5.9 Product-market fit

Definition

Product-Market Fit (PMF)

Product-Market Fit (PMF) (Marc Andreessen) means "being in a good market with a product that can satisfy that market." Operationally, PMF occurs when a clearly defined customer segment organically adopts, retains, and recommends the product because removing it would cause genuine disruption.

Measuring Product-Market Fit

  1. Sean Ellis 40% Survey Rule: Ask active users: "How would you feel if you could no longer use this product?" If answer "Very disappointed", the product shows strong PMF.
  2. Flattening Retention Curves: Plot percentage of a monthly signup cohort still active in Month 1, 2, 3, . If the curve flattens above zero (e.g., stabilizing at after Month 4 instead of decaying to ), the product has found a loyal core segment.
  3. Organic Word-of-Mouth & Net Revenue Retention (NRR): Customers expand seats/usage faster than churn erodes revenue ().

5.10 Iterative product development

After MVP launch, IT teams evolve the product through short, data-informed iterations (typically 1–2 week Agile sprints):

  1. Instrument Events: Log key funnel actions (signup aha-moment repeat use payment).
  2. Cohort Analysis: Compare newer user batches against older batches so overall user growth never hides worsening retention.
  3. Prioritize Bottlenecks: Fix the biggest drop-off step in the user journey rather than adding new peripheral features.

5.11 Pivoting and experimentation

Definition

Pivot (Structured Course Correction)

A pivot is a structured course correction designed to test a new fundamental hypothesis about the product, business model, or engine of growth while preserving the validated learning accumulated so far.

Major Types of Pivots in IT Startups


5.12 Technology adoption and scalability

Crossing from early enthusiasts into mainstream adoption requires addressing Geoffrey Moore's Chasm (between Visionary Early Adopters and Pragmatic Early Majority) and engineering technical + operational scalability:


Worked Examples

Example 1

1 (8 marks)

A SaaS startup runs an A/B pricing and onboarding experiment on landing-page visitors split evenly ( in Control Group A, in Variant Group B):

Calculate the conversion rate of each group, the absolute lift (percentage points), and the relative lift (), and explain how the Build–Measure–Learn loop guides the team's next decision.

Conversion Rates for Control (A) and Variant (B):

Absolute Lift and Relative Lift:

Build–Measure–Learn Decision: Because Variant B improved trial starts by relative to baseline, the team learns that guided onboarding removes early friction. Rather than stopping at trial signup, the next loop measures Day-30 paid retention of Cohort B vs. Cohort A to verify whether concierge-onboarded users convert into long-term paying subscribers.

Answer:

, , absolute lift percentage points, and relative lift . The team adopts guided onboarding for the next cohort while tracking Day-30 paid retention.

Example 2

2 (6 marks)

An ed-tech startup polls active users using the Sean Ellis Product-Market Fit question and gets the following responses:

Meanwhile, the startup holds in bank cash, earns in recurring net revenue, spends in total operating expenses, and currently keeps reserved as a minimum safety buffer. Calculate (a) the PMF score (), (b) monthly net burn rate, and (c) usable runway (in months) before hitting the safety buffer.

Sean Ellis Product-Market Fit Score:

Since , the product shows strong evidence of Product-Market Fit.

Monthly Net Burn Rate and Usable Runway:

Usable cash above the safety buffer :

Answer:

(a) PMF score (exceeds the threshold); (b) Monthly net burn ; (c) Usable runway .

Example 3

3 (6 marks)

A student team wants to build an AI-powered career-counseling and resume-matching portal for fresh engineering graduates. Recommend an appropriate MVP strategy so they can validate willingness to pay within 3 weeks without training custom ML models upfront, and state what signal would justify a Customer Segment Pivot.

Recommended MVP Strategy (Wizard of Oz + Concierge MVP): Build a simple landing page and intake form (Google Form / lightweight React page) promising personalized resume optimization and curated job matches within 24 hours for a small fee (e.g., ). Behind the scenes (Wizard of Oz / Concierge), the founders manually review resumes and match openings using existing tools and checklists, measuring conversion rate, satisfaction, and repeat referrals.

Criteria for a Customer Segment Pivot: If fresh graduates refuse to pay (conversion ), yet HR recruiters or college placement cells repeatedly ask to pay for filtered candidate shortlists, the team executes a Customer Segment Pivot from B2C students to B2B recruiters/colleges.

Answer:

Use a Concierge / Wizard of Oz MVP (simple intake frontend + manual matching behind the scenes) to test real payment conversion in 3 weeks. If graduates resist paying while HR teams or placement offices offer to pay for vetted shortlists, execute a Customer Segment Pivot toward B2B hiring.


Exercises

Practice Exercises — Lean Startup, Design Thinking & MVP

Question 1

Explain the Build–Measure–Learn feedback loop of the Lean Startup methodology. Why does Eric Ries emphasize planning this loop in reverse (Learn Measure Build), and what is the difference between vanity metrics and actionable metrics?

Build–Measure–Learn & Reverse Planning: Build–Measure–Learn converts startup assumptions into empirical evidence through rapid iterations. Planning happens in reverse—first identifying what must be learned (the riskiest hypothesis), next deciding what behavioral metric measures it, and only then building the smallest MVP needed to run that test—so teams never write wasted code.

Vanity vs. Actionable Metrics:

  • Vanity metrics (cumulative signups, total page hits) look impressive on slides even when users churn immediately.
  • Actionable metrics (cohort retention, conversion rate, , ) isolate cause-and-effect so teams can verify whether product changes truly improve performance.
Answer:

The Build–Measure–Learn loop accelerates validated learning; planning in reverse (Learn Measure Build) ensures every MVP is built strictly to test a critical hypothesis. Actionable cohort metrics measure real retention and conversion, whereas vanity metrics mask churn behind cumulative totals.

Question 2

Compare the five stages of Design Thinking (Empathize, Define, Ideate, Prototype, Test) and explain how Design Thinking complements the Lean Startup methodology.

Five Stages of Design Thinking:

  1. Empathize: Observe and interview users to understand real behaviors and emotions.
  2. Define: Synthesize findings into a clear human-centered problem statement.
  3. Ideate: Generate diverse potential solutions before narrowing down.
  4. Prototype: Build low-cost tangible artifacts (wireframes, mockups) to test ideas.
  5. Test: Observe users interacting with prototypes and refine understanding.

Complementarity with Lean Startup: Design Thinking uncovers deep customer empathy and creative product concepts during problem framing; Lean Startup validates whether those concepts work commercially through MVPs, actionable cohort metrics, and business-model pivots.

Answer:

Design Thinking moves iteratively through Empathize, Define, Ideate, Prototype, and Test to ensure human desirability, while Lean Startup tests commercial viability through MVPs and Build–Measure–Learn experiments.

Question 3

An IT team tests two onboarding flows for a mobile bookkeeping app on new installs ( users in Control Flow A, users in Variant Flow B). In Flow A, users record their first 5 transactions within 48 hours; in Flow B, users do so. Calculate the activation rate for each flow, the absolute lift in percentage points, and the relative lift ().

Activation Rates:

Absolute and Relative Lift:

Answer:

Flow A activation , Flow B activation , absolute lift percentage points, and relative lift .

Question 4

A B2B SaaS startup surveys active users with the Sean Ellis PMF question: report they would be "Very disappointed" if the product disappeared. The startup currently has in cash, monthly revenue of , and monthly operating costs of . Compute: (a) the PMF score (), (b) monthly net burn, and (c) total runway in months.

PMF Score:

Monthly Net Burn and Runway:

Answer:

(a) PMF score (strong PMF); (b) Net burn ; (c) Runway .

Question 5

Define a Minimum Viable Product (MVP) and contrast Concierge MVP, Wizard of Oz MVP, and Smoke Test MVP. Also explain three common types of pivots when an MVP experiment falsifies an initial hypothesis.

MVP and Three Strategies: An MVP maximizes validated learning with minimal engineering effort.

  • Smoke Test: Landing page/ad testing interest before building software.
  • Concierge MVP: Founders visibly perform the service manually for early users to learn workflow nuances.
  • Wizard of Oz MVP: Users see an automated software UI while humans execute processing behind the curtain.

Three Common Pivots:

  1. Zoom-In Pivot: Turning one high-demand feature into the entire product.
  2. Customer Segment Pivot: Keeping the solution while targeting a more urgent buyer segment.
  3. Revenue Model / Channel Pivot: Changing monetization (e.g., consumer ads B2B subscription) or distribution channel.
Answer:

An MVP delivers core learning at minimum cost via Smoke Tests (demand check), Concierge (manual service), or Wizard of Oz (manual backend behind automated UI). When experiments invalidate assumptions, founders execute structured Zoom-In, Customer Segment, or Revenue Model pivots.

← Business Models and Competitive Advantages (part 2: 4.6–4.9)Business Planning and Emerging Technologies →