Someone senior has asked 'what's your plan?' and I haven't got one yet.

What to do when someone senior asks for your plan and you haven't got one

The Bay Area, early twenty-tens. A startup called SkyRocket. The founder is exceptionally good at selling.

Good enough to walk into enterprise procurement meetings with a PowerPoint and walk out with signed contracts. Not pilot agreements — purchase orders. Tens of thousands of dollars at a time, on the strength of slides. There is no product behind the deck. There is no architecture behind the product. There is no team behind the architecture.

The slides describe what the software will do. The customers nod, sign, and wait for delivery.

When delivery becomes the question, the founder hits the gap. He is not a technical founder. There is no technical founder. The development team he has assembled in the months between the deals being signed and the deals being shipped is small, under-supported, and looking at a backlog that grows every week the founder closes another deal.

He takes a short leave of absence. Stress, exhaustion, the visible cost of selling something into existence faster than it can be built.

In his absence, the development team quits. Not one by one. As a group. They take the contracts with them, set up under their own banner, and the customers — who care about the working software, not the founder — go with the contracts.

The company collapses overnight.

The founder writes the post-mortem himself, eventually, and it is unsparing. He was good at the front end. He was absent from the back end. The plan, as such, was the deck. There was nothing underneath it.

So let's go to the office and work through it.

"Someone senior has asked 'what's your plan?' and I haven't got one yet."

The feeling is caught.

The email arrived an hour ago. The reply is due before the next meeting, and the next meeting is tomorrow morning, and the question itself is doing more work than the words on the screen suggest.

First work out whether they want to help or to be reassured

Two choices. The wrong one costs more than the question itself.

When they want to be part of the answer

Choice one: they want to be part of the answer. The senior who'd rather shape a plan than read one. The question isn't a test; it's an opening. They've spotted you don't have one yet, and they'd quite like to help. People at that level know what it cost them not to be asked.

If that's the trap, reply by asking for their advice on two genuinely different approaches you've been weighing up. You don't need to have been weighing them up before the email arrived. The approaches you invent on the spot are fine, as long as they're real options. What matters is the question converts them from interrogator to collaborator. People rarely interrogate a plan they helped build.

When they just want reassurance it exists

Choice two: they want reassurance. The subtext is tell me you've got this. The question isn't an opening. It's a stress test of whether you have a credible answer at all. They don't want to shape your plan; they want to know it exists and that you can describe it in a sentence.

If that's the trap, name the single most obvious, dull first step you're taking today, and say it with the confidence of someone who has thought about it for longer than five minutes. We're spending the next two weeks talking to fifteen of our existing customers about how they're using the product, and what they tell us will write the first version of the plan. The sophisticated answer is for later. In high-pressure environments, stating the obvious plainly is almost always read as clarity.

Get this wrong in either direction and the cost is worse than the question. Treat a collaborator as an interrogator and you've turned a useful conversation into a defensive briefing. Treat an interrogator as a collaborator and you've confirmed their suspicion that you don't have a plan, by asking them for one.

How to commit to a direction you can still walk back

Three tools. The discipline is to assemble a defensible direction without committing to anything you can't walk back.

Mark which assumptions the plan rests on

The first is

Lean Canvas.

Maurya adapted the Lean Canvas from Alex Osterwalder's Business Model Canvas in twenty ten, in Running Lean. The version most teams use today is Maurya's — Osterwalder's underlying structure retained, the boxes reworked for early-stage product work.

Nine boxes on a single page: problem, customer segments, unique value proposition, solution, channels, revenue streams, cost structure, key metrics, unfair advantage.

The reason the tool exists is to distinguish what's load-bearing from what's derivative. Most strategy work treats every claim as equal weight. The canvas treats some claims as foundations and others as consequences — and the order of filling the boxes makes that explicit.

The unique insight is that the boxes you can't fill yet are the ones that matter. Assumptions you can see are assumptions you can test. Assumptions you can't see are the ones that kill you, because you don't know they're load-bearing until something breaks underneath them.

What you get is the structure the plan will harden into, with the assumptions visibly marked. Not the plan itself — the scaffold.

So. How to run it.

Setup. The canvas on a wall, ninety minutes, the people whose work the plan would commit to in the room.

Load-bearing. Fill four boxes first, in this order: Problem, Customer Segments, Unique Value Proposition, Solution. Everything else depends on them. If you can't name the problem in a sentence and the customer in a sentence, the solution box is guesswork dressed as strategy.

Derivative. Now the other five: Channels, Revenue Streams, Cost Structure, Key Metrics, Unfair Advantage. These follow from the load-bearing four. If they don't, the load-bearing four aren't right.

Assumptions. The boxes you couldn't fill, or filled with low confidence, get marked. A circle, a sticky note, whatever. Those are what the plan rests on. The senior who asked for the plan needs to see them as assumptions, not commitments.

The canvas is what you put in front of the senior. Not the plan. The scaffold.

Commit to direction, not to dates you can't keep

The second is

Roadmap Planning.

The name is unhelpful — it sounds like a Gantt chart with deadlines on it. The version that earns its place was formalised by Roman Pichler and the ProductPlan crowd in the early twenty-tens, building on a longer agile-product lineage.

Three columns: now, next, later. Themes, not features. Each theme names a direction the product is moving in, without committing to a specific deliverable on a specific date.

The reason the tool exists is that feature roadmaps with dates rent credibility against the slipping. The day a feature slips, the credibility of the document slips with it. Themes commit to direction without renting credibility against dates that haven't earned themselves yet.

The unique insight is the shift from features to themes. A feature roadmap promises specifics. A theme roadmap commits to direction. When you don't have a plan yet — and the question is whether you have one — themes are what you can credibly put on the page. Features would be lying.

What you get is a chart that's honest. It also gives the senior asker something to react to, which is what they were really after.

So. How to run it.

Now. The work in flight this quarter. Themes, not features. Reducing onboarding friction. Closing the analytics gap. Direction, not deliverable.

Next. The directions discovery is closing on. Where you'd commit if the discovery work confirms what it currently suggests. Less specific than Now, more specific than Later.

Later. Directions you've recognised but haven't started on. Honest about uncertainty. Senior stakeholders hate this column most; it's also the most useful, because it shows you're tracking what you're not yet building.

The chart goes to the senior. They'll ask where the dates are. The honest answer is that the dates are made up in any roadmap until the team has done the discovery underneath them — and a roadmap with made-up dates costs you twice, once when you commit and again when you slip. Themes cost you nothing on the way in.

Time the commitment to the information, not the calendar

The third is

Real Options Theory.

Stewart Myers, MIT, nineteen seventy-seven — applying the financial-options mathematics from Black and Scholes to decisions that aren't financial. The mechanics are heavier than most managers need; the discipline underneath them is essential.

Three questions, run against any decision the room is about to commit to.

First — is this reversible? If the answer is yes at low cost, treat it as an experiment, not a decision. The conversation is whether to try, not whether to commit. The room argues less when the decision is recoverable.

Second — what does it cost to hold? Some decisions get cheaper the longer you wait, as information arrives. Others get more expensive — competitor moves, capacity loss, team morale. The question isn't whether to wait; it's what the wait costs and what arrives during it.

Third — what information will I have in a week, a month, a quarter, that I don't have today? Decisions made before the information arrives are decisions that will be revised. Decisions made just after it arrives are decisions that hold. The discipline is to time the commitment to the information, not to the calendar.

The senior asking for your plan is, almost always, asking for a commitment. Real options is the discipline of distinguishing the commitments worth making today from the ones worth deferring until the information they depend on actually arrives. Not all decisions are equally reversible. Treating them as if they were is what produces plans that have to be rewritten the moment they meet contact with reality.

The harder discipline is the one underneath. Most teams treat decisions as more irreversible than they are, because we have decided feels final, and finality feels like progress. Real options names that as an error. Reversibility is a property of the decision, not of how the room talks about it.

That's the toolkit. One more story before we close.

A precedent: ship the present so the plan reveals itself

The SkyRocket founder sold a future that didn't exist, and the present collapsed underneath him. The story we close with is the inversion: someone who shipped a present so specific that the future arrived around it.

September twenty ten. Washington DC. Elizabeth Warren has just been asked by President Obama to set up the Consumer Financial Protection Bureau.

The bureau exists, at this point, as a few lines of authorising language in the Dodd-Frank Act, signed into law two months earlier. There is no staff. There is no premises. There is no budget code. There is not even a settled view inside the Treasury or the Federal Reserve about what the new agency is actually for.

Warren had written the conceptual paper that proposed the idea — Unsafe at Any Rate, in Democracy: A Journal of Ideas, summer two thousand and seven. The paper was an argument. It was not a plan. Nor did the agency have a plan.

The standard move, at this point, would be the org chart. Hire the senior team, define the divisions, draft the regulatory remit. Build the institution and ship the products it produces, in that order.

Warren does the opposite.

She starts not with the org chart but with the first user-facing product the agency will ship: a simplified mortgage disclosure form. The form will replace the two-page legal document homebuyers had been signing for decades — the document everyone signed without reading — with a single page in plain language. The initiative gets a name before the bureau does. Know Before You Owe. It launches on the eighteenth of May twenty eleven. Before the bureau has finished hiring. Before it has a confirmed director. Before much of its regulatory structure has been built.

The product pulls the organisation into existence around it. The teams form because the teams are needed to ship the form. The legal opinions get written because the form needs them. The regulatory approach hardens because the form has to be defensible. The hiring gets done because someone has to staff the work.

The plan didn't precede the agency. The first thing the agency shipped did, and the plan grew up around it.

So. The SkyRocket founder sold the future and the present collapsed. Warren shipped the present and the future formed around it. What's your plan? is almost never a question about the plan. It's a question about whether you have one. The useful answer is rarely the full plan; it's the first specific, visible thing you're about to ship that will force the rest of the plan to reveal itself.

So. The question from this Playbook.

Which decision in your next meeting is genuinely irreversible — and which one are you about to treat as irreversible because someone wants closure?