A Bridge Fee Cannot Quietly Carry a Delivery Team

I spent an evening modelling three shapes for the same client relationship, and the useful output was not the model. It was noticing that I had been about to run the largest of the three on the fee attached to the smallest.

The three shapes: a full delivery pod with named people doing production work; a narrower enablement partnership where I advise and build the operating system but do not staff it; and an in-kind design-partner arrangement with commissioned work billed separately. Each has different deliverables, a different headcount, and a materially different cost.

What was actually on the table was a one-month bridge fee — a fraction of the pod's cost, agreed informally to keep the relationship warm while the bigger conversation happened.

And I was already standing up the pod.

How this happens, because it is not stupidity

Nobody decides to subsidise a client. It arrives one helpful act at a time.

The bridge fee was agreed for a scope roughly the size of the bridge fee. Then the kickoff went well, and the follow-through from a good kickoff is a training programme. The training programme needs materials, and the materials need validating, and validating them needs a named owner, and now there are assigned accounts and a cohort and a weekly cadence — all of which is delivery, none of which was priced, and every step of which was the obvious next thing to do given the step before it.

The scope did not creep. The scope was never bounded, so there was nothing for it to creep past. A bridge fee with no stated deliverable is not a small commitment; it is an unbounded commitment with a small number attached, which is considerably worse. The number tells the client what they are paying. Nothing tells either of us what they are buying.

The arithmetic I had to look at properly

The thing that made me stop was running my own runway alongside the pod cost. At my then-current balance against a minimum monthly burn, I had a little over two months. The pod, staffed as designed, would consume more than the bridge fee every month it ran. So the difference was not coming from the client. It was coming from the two months.

Put plainly: I was proposing to fund a client's delivery capacity out of my own survival window, and calling it a relationship investment.

That is a real strategy in some circumstances — you can buy a reference, a case study, or a seat at a table. But it is only a strategy if you have decided it, sized it, and named what you are buying. I had not. I had drifted into it because declining felt like the thing that would damage the relationship, and because the work was interesting.

What I am now writing down before any work starts

Four lines, and they fit in one message:

• The fee, and what period it covers.
• The payment date, agreed in advance rather than discovered at invoicing.
• One priority — the single thing this period is for. Not a list. If there are three priorities there is no priority and the scope is unbounded again.
• The delivery boundary — what is explicitly not included and would be a separate conversation.

The fourth line is the one that does the work, and it is the one I have historically left out because it reads as ungenerous. It is the opposite. A boundary is what makes the next conversation possible; without it, every expansion is either free or an awkward renegotiation of something the client reasonably thought they had already bought.

The related rule, from the same evening: present distinct commercial shapes separately, each with its own deliverables and its own funding. Blending a founder-advisory relationship with an execution pod produces a proposal where the advisory fee appears to include the execution. Three clean options invite a choice. One blended option invites the client to assume the generous reading, and they will, because I wrote it.

Where this goes wrong

Insisting on a written boundary before a small bridge can kill a relationship that was going to become a real engagement. There is a version of this discipline that reads as distrust at exactly the moment trust is being built, and with a long-standing client that cost is not theoretical. The mitigation I believe in: the boundary is one line in a message, not a contract amendment, and it is framed as clarity about what I will focus on rather than as a limit on what they can ask.

And a bounded scope can be the wrong instinct when you genuinely do not know what the work is yet. Early discovery sometimes has to be open. But "we don't know what this is yet" is itself a priority you can name and price — a scoping period with a date and an output. That is a different thing from staffing a pod on a handshake.

The honest counter-argument: I may be reaching for a process fix for what is actually a concentration problem. If one client were not most of my revenue, I would not be tempted to over-deliver into them, and no amount of boundary-writing changes the underlying incentive. That is probably right, and the written boundary is still the cheapest thing available this month while the harder fix — more clients — takes quarters.