Gibson Biddle on Product Strategy

Source: Lenny’s Podcast Speaker: Gibson Biddle Source URL: https://www.lennysnewsletter.com/p/gibson-biddle

Key ideas

DHM model: product strategy = hypotheses about how to delight customers in hard-to-copy, margin-enhancing ways. All three dimensions must be satisfied simultaneously. Any proposed feature or strategy is a hypothesis on all three axes.
Test the ideal, then do the maths: the Perfect New Release Test at Netflix showed customers said they wanted faster DVDs, but the retention impact was negligible ($1M value vs. $5M cost). Never substitute stated preference for observed behaviour.
JAM force-ranking: the #1 source of startup misalignment is unresolved disagreement about whether Growth, Engagement (quality), or Monetisation comes first. Force-ranking these three clears the air.
SWAG → iterate → present: take a rapid provisional position (Stupid Wild-Ass Guess), share it one-to-one with better-informed people, iterate, then present. Beats paralysis or outsourcing the question.
Two-way door discipline: most product decisions are reversible and small relative to revenue. Ask “what’s the magnitude?” and “is this reversible?” before treating any decision as high-stakes.

DHM model

“Delight customers in hard-to-copy, margin-enhancing ways.” Learned from Reed Hastings; distilled from his reference check question: “Can Gib delight customers?”

The three dimensions:

Delight: aim for 10X better, not satisfactory. Peter Thiel’s framing from Zero to One applied to product.
Hard-to-copy: structural advantages competitors cannot replicate quickly. Netflix examples: personalisation (taste graph of 1B+ profiles), brand trust (customers give their credit card monthly), original content, economies of scale. Easy-to-copy tactics (e.g., hero image on sign-up page) work briefly but create no lasting advantage.
Margin-enhancing: does this build a better business? Personalisation enables right-sizing content investment; subscriptions produce predictable unit economics.

The delight–margin tension is the hardest trade-off. If Netflix dropped its price from $20 to $5/month tomorrow, delight would spike but the business would collapse. The word-of-mouth factor (how many additional customers each retained customer brings) is one lever for resolving this: a higher WOM multiplier justifies more product investment.

The portfolio reality: Gibson had ~10 strategy hypotheses, ~6 failed. Netflix social (movie recommendations from friends) failed twice. The model does not guarantee success; it ensures that failures generate useful information.

Netflix case studies

Perfect New Release Test

Hypothesis: shipping new-release DVDs next-day (vs. average 2-week wait) will meaningfully improve retention.

A/B test (~2005): 10,000 members received next-day delivery; rest continued as normal. Result: churn dropped from 4.5% to 4.45% — saving ~5,000 customers against 1M subscribers if rolled out.

DHM maths:

Value: 5,000 customers × $100 LTV × 2X word-of-mouth = $1M
Cost: $5M in additional DVD inventory

Decision: do not roll out. Delight is real but margin negative at 2X WOM. At Amazon’s 10X, the maths would flip.

Key lesson: customers universally said “faster DVDs” was their top request. The A/B test revealed the real impact was negligible. Stated preference ≠ retention behaviour.

Auto-cancelling inactive members

During COVID, ~0.5% of Netflix members had not used the service in a year. A PM (Eddie Woo) proposed auto-cancelling them and refunding money.

Delight ✓ — customers feel respected; builds brand trust
Hard-to-copy ✓ — brand halo; competitors won’t do this
Margin ✗ — $100M annual revenue loss

Decision filter: magnitude ($100M / $30B revenue = 0.3%, low) + reversible (can stop at any time). Proceed.

Lesson: most decisions that feel high-stakes are reversible and small in proportion. Explicitly asking “magnitude?” and “reversible?” unlocks action.

Ad-supported plan

Context: Netflix earnings Q1 2022 — first customer loss in 10 years; ~30M accounts sharing passwords outside the home.

Reed Hastings killed advertising in 2008 (after two profitable years) to maintain simplicity and stay focused on personalisation. His Socratic logic: “Who’s going to be best in the world at advertising?” (Google). “Who needs to be best at personalisation?” (Netflix).

By 2022, the calculation changed. Reed framed it as “customer choice” outweighing complexity of ads; also, advertising infrastructure had matured such that Ads revenue could be outsourced to partners, keeping Netflix focused on its core.

The lesson: strategic principles are contextual. “No advertising” was right at $20M operating income with Google as the benchmark; it was revisited when growth stalled and infrastructure commoditised.

~30M of Netflix’s estimated 100M US accounts were shared outside the household. Netflix had unintentionally incentivised this by selling “multiple streams” as the primary reason to upgrade to $20/month.

Fix: reframe the $20/month plan around resolution and audio quality; add fine print clarifying household-only sharing. Test charging for external profiles in smaller markets (Peru, Caribbean).

Key question Gibson poses: what % of password-sharers will convert to paid when prompted? Anecdote: Netflix free trial conversion was 90%. External sharer conversion is lower (perhaps 5-10%), but even 5% of 30M = 1.5M new subscribers.

JAM model: Growth, Engagement, Monetisation

The most common source of leadership misalignment Gibson observes across startups: unspoken disagreement about the relative priority of Growth, Engagement, and Monetisation.

How to use it:

CEO, CPO, and CFO each privately rank the three.
Compare. If misaligned, resolve before setting product priorities.
Revisit when strategy changes.

At Netflix (2005): Growth → Engagement → Monetisation. At Chegg (2010): CEO wanted Growth first; CFO wanted Monetisation first. After failed resolution, CFO departed. Not a personality clash — a structural misalignment.

The hardest part: defining the engagement metric. At Netflix, Gibson used monthly retention (churn rate). Agreeing on this forces the company to specify what “product quality” means in measurable terms.

SWAG: start with a provisional position

Don’t wait for perfect data before forming a product strategy view.

The practice:

Take a SWAG — a rapid personal position based on available knowledge.
Share it one-to-one with the 4-6 people who have the most relevant context. “This is my best thinking — what am I missing?”
Incorporate their refinements.
Present the evolved version to a broader group.

This approach avoids: (a) endless analysis paralysis, (b) starting group discussions with no anchor, (c) hiring consultants to do the thinking.

Gibson applied this on his first two weeks at each company — developing a product strategy SWAG, then iterating via individual meetings before any group presentation.

Personal board of directors

A deliberately maintained network of peers and senior mentors consulted on career and strategic decisions.

Peers: former colleagues and LinkedIn connections at similar levels. Useful for specific tactical questions (“mobile vs. desktop investment right now?”).
Senior mentors: people who can see around corners. Built by finding ways to be genuinely useful, not by asking to be mentored.

Gibson credits his personal board with all three major company-joining decisions (EA 1991, Netflix 2005, Chegg 2010). His test: asking a VC friend, “Would you invest in this company?” — proxies the “should I invest my time?” question.

Rule: do not ask someone to be your mentor. Instead, find what they need and provide it. Gibson’s example: a data analyst who wanted a product career built Gibson’s website (Squarespace), which Gibson needed; Gibson provided introductions in return.

Career as product

Gibson’s meta-framework: treat career decisions as product hypotheses.

Define a hypothesis (“I might enjoy teaching outside formal employment”).
Run a minimum-viable experiment (taught a Stanford course; liked but didn’t love it due to location constraints).
Measure engagement (did the experience produce genuine satisfaction and learning?).
Iterate or pivot (moved to global talks and workshops with full flexibility).

Applies the same SWAG → test → learn loop to career as to product.

Key advice:

Optimise for learning, especially early career.
Pick the right company: use your personal board of directors.
Be bold: incremental safe choices produce slow compounding; occasional high-risk experiments produce step-changes.