A good strategy says what you won’t do.
Strategy is choice under constraints. Most teams feel the constraints every day—too many priorities, too few people, ambiguous goals—but rarely turn those constraints into explicit choices. That’s why I like strategy one‑pagers: they compress the mess into something a busy team can read, remember, and reference when it’s time to say “no.” If it doesn’t fit on one page, it probably won’t fit in anyone’s head.
What a one‑pager is not: a budget, an OKR spreadsheet, or a kitchen‑sink roadmap. Those can be useful artifacts, but they’re not strategy. A strategy one‑pager names the outcomes, the hard tradeoffs, the few bold bets, and the measures that keep you honest. It’s the document you bring to prioritization fights and quarterly planning so that you argue about the right things.
Why one page works
Compression forces clarity. A page boundary makes unspoken assumptions pop: if this goal matters, what are we willing to give up? If that constraint is real, which bets still make sense? Length correlates with cowardice—when we’re afraid to confront tradeoffs, we add more words. One page isn’t a gimmick; it’s a forcing function.
How to write one in 45 minutes
- Set a timer for 15 minutes and draft the four sections below as ugly bullets. Don’t wordsmith yet.
- Spend 15 minutes cutting. Merge overlapping goals, delete “nice‑to‑haves,” and turn ambiguous words into measurable outcomes.
- Spend the last 15 reading it out loud. Anywhere you stumble, simplify. Anywhere you hedge, choose.
Invite one trusted skeptic to do a ruthless pass. If they can’t restate your strategy in two sentences, you have more cutting to do.
What makes this engineering strategy (and not just product fluff)
Engineering strategy translates beliefs into system‑level choices. If the goal is faster time‑to‑value for new customers, the bet might be “guided setup + golden paths + default configs,” and the measure is “TTV median under 10 minutes from sign‑up to first successful API call.” That’s not hand‑wavy. It drives architecture, sequencing, and investment.
Strategy also protects focus. “No major backend rewrites this quarter” can be the most important sentence in the document. Not because backend work is bad, but because every hour spent there is an hour not spent on the outcome we chose.
An example (fictional, painfully familiar)
- Goal: Cut onboarding TTV by 60% and double trial‑to‑paid conversion.
- Constraints: Two squads, one SRE, keep P95 latency under 250ms, no new vendors.
- Bets: Guided setup with sensible defaults; a sandbox dataset; job visibility with retries; instrument everything.
- Measures: TTV median and 90th percentile; conversion rate; first‑14‑day support tickets per new account.
What we won’t do: roll out service mesh, replatform billing, or unify every config file. Important work, not this quarter’s work.
Anti‑patterns I avoid
- Date precision theater: quarters for bets, specific dates only for hard launches.
- Metric theater: if it’s not on a dashboard we review weekly, it doesn’t count.
- People are fungible: they aren’t. Scope to teams you actually have.
- Kitchen‑sink strategy: a grab‑bag of everything interesting is the opposite of focus.
Cadence and governance
Monthly: review actuals versus measures and adjust bets (not goals) if evidence demands it.
Quarterly: prune or double down. If a bet isn’t moving a measure, kill it without ceremony and promote the next bet. Keep the one‑pager updated and versioned—link to ADRs, RFCs, and dashboards.
Annually: revisit beliefs. If the market or constraints changed, say so explicitly and re‑choose. Carry forward measures that still predict success; retire vanity metrics that never did.
Use it to say no (politely)
When a “quick win” arrives that doesn’t advance the plot, point to the one‑pager: “Great idea; it doesn’t move TTV or conversion this quarter. Let’s park it for the next planning pass.” People respect a clear story, especially when it was shared in advance.
One‑pagers don’t make choices painless. They make them explicit. That’s the job.
One page, four parts
- Goals: The outcomes we care about.
- Constraints: Budgets, headcount, non‑goals.
- Bets: What we’ll try and why.
- Measures: How we’ll judge success or stop.
Share it, revisit it quarterly, and delete work that doesn’t fit.