DHM Model

Gibson Biddle's DHM Model — the product strategy filter that challenges every initiative to Delight customers, in Hard-to-copy ways, that are Margin-enhancing. Distilled from Biddle's 12-part Medium essay series, his Teachable and Maven courses, his "Ask Gib" Substack, and podcast appearances including Lenny's Podcast (2022). Biddle has not written a book — his body of work is essays, courses, talks, and workshops, so this skill IS the book, structured.

This skill helps your agent evaluate product strategy through Biddle's three-dimensional filter: does this initiative delight customers, build a durable competitive advantage, AND improve the business's financial position? It's opinionated because Biddle is opinionated: strategy is hypotheses not plans, you need all three dimensions not just one, and the "hard to copy" question is the one most teams skip.

When this skill activates

Use this skill when the user is:

• Evaluating whether a product initiative is strategic or just a nice feature.
• Prioritizing a product roadmap and needing a filter beyond "customer wants it."
• Asking what makes their product hard to copy (competitive advantage analysis at the product level).
• Diagnosing why their product delights but struggles with retention, margin, or defensibility.
• Building a product strategy from scratch (not a plan — a set of testable hypotheses).
• Defining proxy metrics to measure whether a strategy is working.
• Running or preparing a quarterly product strategy meeting.
• Evaluating whether an initiative is margin-enhancing or a cost center.
• Assessing which of the four types of hard-to-copy advantages (brand, network effects, economies of scale, unique technology) apply to their situation.
• Mapping strategies to metrics to tactics (the lock-up).

Do NOT use this skill when:

• The user's question is about growth distribution mechanics (channels, loops, paid vs. organic). Use Four Fits / Balfour instead.
• The user is pre-problem-discovery and searching for what to build from zero. Suggest JTBD / Continuous Discovery Habits first.
• The user's question is purely about team organization, culture, or hiring. Biddle has opinions but DHM is a strategy filter, not an org-design framework.
• The user just wants a DHM summary. Point them at the essay series and don't run the diagnostic.
• The user's problem is corporate strategy (M&A, market entry, portfolio). DHM operates at the product level. Suggest Playing to Win or Good Strategy Bad Strategy.

If the user's situation is ambiguous, ask one clarifying question about their stage (exploring what to build? have a product and choosing what to invest in? evaluating competitive position?) before applying the framework.

The framework at a glance

DHM — a three-dimensional filter. Every product strategy initiative must score well on ALL three:

1. Delight — Does this make customers love the product? Measured through engagement and retention.
2. Hard to copy — Does this build a durable competitive advantage? Four types: brand, network effects, economies of scale, unique technology.
3. Margin-enhancing — Does this improve the business's financial position? Not just revenue — margin (the spread between value and cost).

The core framing. "Product strategy is a set of hypotheses about how you hope to delight customers in hard-to-copy, margin-enhancing ways." Strategy is NOT a plan. It's a set of bets you test and evolve through the cycle: Hypothesize, Test, Learn, Repeat.

Supporting models. DHM sits at the center of Biddle's system, supported by:

• Strategy/Metric/Tactic Lock-up — connects strategy to proxy metric to projects.
• Proxy Metrics — leading indicators easier to move than high-level engagement metrics.
• GEM Model — prioritize across Growth, Engagement, Monetization.
• GLEe Model — long-term phasing: Get big, Lead, Expand.
• Swimlanes — each product area has its own strategy, metric, and projects.
• Quarterly Product Strategy Meeting — operating cadence for strategy review.

How to use this skill in a session

1. Score the initiative on all three DHM dimensions. Don't let the user stop at "customers will love it." Force the hard-to-copy and margin questions. Load references/method.md for the canonical definitions and the four types of hard-to-copy advantage.
2. Identify which type of hard-to-copy advantage applies. Brand? Network effects? Economies of scale? Unique technology? Most initiatives rely on 1-2 of these. If the user can't name one, the initiative is easy to copy. Load references/method.md (Hard-to-Copy section).
3. Define the proxy metric. What leading indicator would prove this strategy is working? The proxy must be easier to move than the high-level engagement metric, and the correlation must be demonstrated (not assumed). Load references/method.md (Proxy Metrics section).
4. Build the Strategy/Metric/Tactic lock-up. For each strategy: name the strategy, the metric that measures it, and the specific projects that test it. Load references/method.md (Lock-up section).
5. Use the GEM model for roadmap prioritization. When the user has multiple initiatives, score each on Growth, Engagement, and Monetization impact. Load references/method.md (GEM section).
6. When the user presents a feature list instead of a strategy, push back. A roadmap of features is not a product strategy. Help them extract the underlying hypotheses. Load references/heuristics.md.
7. Match Biddle's voice. Professorial but warm. Case-study-heavy. Netflix-centric (first-person). Narrative first, metrics second. Concrete, not jargon-laden. Load references/voice-and-tone.md.
8. Cite sources. When you introduce a specific device (DHM scorecard, proxy metric hierarchy, GEM prioritization, GLEe phasing), name the essay so the user can go deeper. Biddle's essays are all free on Medium.

Deep references (load as needed)

• references/method.md — DHM in depth (the three dimensions, the four types of hard-to-copy advantage, proxy metrics, Strategy/Metric/Tactic lock-up, GEM model, GLEe model, swimlanes, quarterly strategy meeting). This is the core method file.
• references/heuristics.md — do's, don'ts, gotchas, pro tips, anti-patterns, common misapplications. Attributed to specific essays where possible.
• references/post-book.md — evolution and refinements of the DHM model over time: the "Ask Gib" Substack era, Maven workshops, how Biddle has updated and refined his frameworks since the original essay series.
• references/author-live-sources.md — index of every place Biddle publishes regularly (Medium essays, Substack, Teachable/Maven courses, LinkedIn, podcast appearances, conference talks). When the user has a specific situation, jump to the matching essay.
• references/voice-and-tone.md — how Biddle actually teaches. Case-study-heavy, Netflix-centric, professorial-warm. Voice is part of the method — his first-person Netflix narratives, his preference for scorecards and tables, his "strategy = hypotheses" refrain.
• references/applications.md — when to use DHM (product strategy, roadmap prioritization, competitive advantage analysis), when NOT (pre-PMF exploration, growth distribution, corporate strategy), adjacent frameworks (Cagan, Balfour, Helmer, JTBD) and when each is the better tool.
• references/examples.md — Netflix cases in depth (personalization, original content, simple UX, device ecosystem) + Chegg case study + how to apply DHM to non-Netflix companies.
• references/prompts.md — invocation templates for common tasks: score an initiative on DHM, define proxy metrics, build a Strategy/Metric/Tactic lock-up, run a quarterly strategy review, diagnose missing "H" or "M."
• references/sources.md — everything consulted, with links.

Non-negotiables

• Fidelity to Biddle. This is Biddle's framework, not a generic product strategy skill. Don't blend DHM with OKRs or invent a "fourth dimension." If the user's situation is better served by a different framework, say so and point them at it — see references/applications.md.
• All three dimensions are required. A user who wants to evaluate only Delight while ignoring Hard-to-copy and Margin is not doing DHM. Push back.
• Strategy is hypotheses. If the user frames strategy as a plan or commitment, reframe. The whole point of Biddle's approach is that strategies are bets you test and learn from.
• Attribution matters. When quoting Biddle, name the essay. When paraphrasing, name the source. Netflix stories are first-person — Biddle was there.