The SprintCompliance ClubProposal developmentPursuit memoryCaptureGraphicsGraduationCompareAboutResultsLearnDoThree betsPricingWorkbenchSecurityFAQBook a call →

Three bets this company is built on.

Every company is a wager on something being true. These are ours, written plainly enough that you can disagree with them. They are bets rather than principles, because a principle cannot be wrong and these can.

One. Experts should drive their own field forward

Human experience is the core value. Not the process, not the tooling. Someone who has done a thing for years knows things that were never written down anywhere, and those are exactly what decide a hard call.

This is why frwrd is one senior person on your account rather than a team of juniors working through a checklist. The large consultancies sell process, and they sell it because process is the thing that scales headcount. We are betting that the scarce ingredient is judgement, and judgement does not come out of a binder.

It is worth saying that this is not a proposals claim. It is as true of a good HR lead, a structural engineer, a chef or a painter. Most industries have organised themselves in a way that dilutes their best people, by surrounding them with process designed to make everyone else adequate.

Two. The gap in AI is not the technology. It is people on the ground.

The models are good enough. The tooling exists and most of it is cheap. Deployment still fails constantly, and it fails in a specific way: someone technical builds a general capability, hands it to a domain they have never worked in, and it becomes shelfware. It was designed from the angle of what the model can do rather than from the angle of what actually breaks in the work.

What closes that gap is a person who knows the domain, sitting inside the business, looking at the real process. They can tell which parts are genuinely mechanical and which parts only look mechanical from outside. That distinction is invisible to someone who has not done the job, and it decides whether any of it works.

Concretely, here. Pulling every requirement out of a 300-page solicitation is mechanical. Deciding which of those requirements are actually discriminators is not, and it looks like the same task if you have never written a proposal. Build for the first and you save a week. Assume the second is the same kind of problem and you produce something confident and wrong, which in this field is worse than producing nothing.

This one is not a forecast. Five years at AWS Public Sector, the last of them as the AI Acceleration lead, building the generative-AI tooling and agents their proposal teams use. That worked because of domain knowledge, not engineering. I knew where the proposals broke.

The bet is that this holds for virtually every non-technical field of expertise. HR, legal, finance, procurement, clinical documentation. Anywhere the expertise is judgement and the output is documents, the same gap is sitting open, and the thing missing is not a better model. It is somebody who knows the work standing in the room.

Three. This should be where experts go to do that

If the second bet is right, the shortage is not models or engineers. It is experts willing to go and stand in the room. So the goal is a company where an expert goes into an organisation, builds, works alongside their people, and leaves it better than they found it. Forward deployed, whatever the field turns out to be.

Today that is one field and one person. Proposals, and me. The proposal department is the first instance of the idea rather than the whole of it, and saying otherwise this early would be the kind of claim this page exists to avoid.

It is written down now because it explains choices that look strange otherwise. Why we sell the handoff rather than the dependency. Why the knowledge base belongs to you and leaves with you. Why we say plainly that when your own team can run this, you should stop paying us. A company that intends to leave things better has to be built differently from one that intends to stay.

What happens if they are wrong

If the first bet is wrong, and the process really is the thing that wins, then a large consultancy is a better purchase than this and you should buy one. If the second is wrong, and a general tool built by people outside the domain turns out to be good enough after all, then this is a worse purchase than a subscription and you should buy the subscription. The third has no evidence behind it yet at all.

You are entitled to know which of those you are buying into. That is most of why the page is here.

The work these bets produce.

Four real cases with the awards as they landed, and the background behind the person doing it.

Start with one bid · $7,500