The first slide of a typical pitch deck is a logo. The second is a value proposition. By the eighth slide, we're showing a methodology with arrows. By the twelfth, a price. None of it is the work. All of it is a description of work, written by people who have not yet done the work, for people who do not yet know what the work will look like.
This is the part of consulting we found we could no longer defend. Not on principle — pitches are how the industry has always moved — but in practice. Every time a deck closed a deal, the first six weeks of delivery were spent unwinding the deck's promises. Every time a deck failed, we'd find out, three months later, that the team picked someone whose deck looked exactly like ours but who didn't ship.
What we send instead.
The current first deliverable in any conversation is what we call a working draft. It is not a proposal. It is a small, real piece of the work we'd do — built before the contract is signed, on our own time, against the actual problem.
For a checkout audit, that's a recorded screen-share where we walk through the prospect's mobile checkout, time the bottlenecks, and pull three specific changes that would lift conversion by an estimated 2–4%. Not "we'd recommend a holistic CX review." Three specific changes. With numbers.
For a data architecture engagement, it's a one-page diagram of the prospect's current stack, drawn from public job postings, vendor logos in their footer, and a 30-minute call with their head of data. It includes the three contracts we'd write first and the order we'd write them in.
A working draft costs us a day. A pitch deck costs us four. The draft wins more, and the engagements that follow start six weeks ahead of where they used to.
— Internal memo, Q1 2024
Why this works better.
Three reasons, and they're all unflattering to the way pitches usually work.
It pre-qualifies the prospect. A team that won't share enough information for a working draft is a team that won't share enough information once the contract is signed. We've stopped pursuing those engagements. They were always the ones that produced bad work.
It pre-qualifies us. If we can't produce a useful draft in a day, we don't understand the problem well enough to take the engagement. The constraint forces us to either learn the domain quickly or walk away — both better outcomes than a deck full of confident-sounding nothing.
It changes the conversation. A pitch is a sales meeting. A working draft is a working session. The prospect's first reaction is not "do I trust these people" but "is this right? Did they get the parts wrong?" That's a much better question to be answering.
What we lose.
This is not free. We turn down roughly 60% of the engagements we used to win on a deck. Some of those were probably worth winning. We've made our peace with that — the remaining 40% start better, run better, and end better, and that's the trade we're making.
We also can't compete in formal RFPs that mandate a deck. We don't try anymore. We've never won one we shouldn't have lost, and we've lost a few we should have won, but the math still works.
If you'd like a working draft for a real problem at your company, send us the problem. Not the brief — the actual problem. We'll send back something useful within five business days, free, no obligation. If it's not useful, you've lost nothing. If it is, we'll talk about whether to do more of it together.
The deeper thing.
The reason this matters is not really about decks. It's about what the deck represents — a way of operating where the description of work is treated as more important than the work itself. Where rehearsing what you'd do is mistaken for the act of doing it.
That habit is everywhere in technology consulting, and it's why so many engagements feel hollow. Three months of discovery with no artifacts. Six months of "alignment" before any code ships. A 200-page strategy document handed off to an in-house team with no idea how to build any of it.
We've had to be honest with ourselves about how much of that we used to do. The pitch deck was a small part of a bigger pattern. Removing it forced us to look at the rest.
A description of work is not work. A roadmap is not a product. A methodology is not a delivery. The thing has to actually exist, or it doesn't exist.
— Working principle · prodct, 2025
What's next.
We're applying the same logic to a few other places where description has crowded out delivery. Annual planning is one — we now plan in 8-week increments, with a working version of the next thing always in flight. Status reports are another — we replaced ours with a public Loom every Friday, three minutes, what shipped.
If any of this is useful to you, take it. We're not protecting it. The point is that there's more work to do, and less time spent describing work yet to be done.
