The objection to buying capacity instead of projects is always the same, and it is a fair one: how do I know what I am getting? A block of credits sounds like a gift card until somebody shows you the conversion.
So here is the conversion, in the open.
The unit
One credit is one specialist day — one person, one working day, on your work. It is deliberately role-neutral, because the whole point is that you do not have to decide in advance whether next month needs a designer or a developer.
Two adjustments keep it honest:
- Development days count as 1.4 credits. A developer day genuinely costs more to supply than a content day. Without a multiplier, every partner would spend the entire pool on the most expensive role, and the model would collapse in a quarter.
- Nothing under half a credit is logged. A twenty-minute fix is half a credit. This is not greed — without a floor, a pool disappears into a hundred micro-requests that never turn into anything a client can see.
What typical work converts to
- A landing page, designed and built, with QA — 6 to 8 credits
- A month of SEO for one client: technical work, on-page, content briefs, reporting — 4 to 6 credits
- A month of paid media management for one client — 3 to 5 credits
- A long-form article, researched and written to brief — 1 to 1.5 credits
- A full-stack account: SEO, ads, content and strategy for one client — 12 to 16 credits
These are ranges rather than fixed prices for a reason. A landing page for a client with a design system and approved copy lands at six. The same page with three stakeholders and no brand guidelines lands at eight. The range is where the honesty lives.
If a supplier gives you a single unqualified number for a landing page, they have either padded it or they are about to send you a change order.
The rules that keep it fair in both directions
- Rollover is capped at 20%. A slow month is not a write-off. But without a cap, a partner banks three months of credits and drops them all in December, which is not capacity planning — it is a hostage situation.
- Overflow is available at plan rate plus 15%. Going over should be possible and slightly uncomfortable, so that persistent overflow becomes a conversation about the next tier rather than a habit.
- Revisions and QA are inside the credit. No hourly billing and no change orders. The estimate includes the rounds that real work actually takes.
- Three-month minimum, thirty days notice. Delivery teams are built ahead of the work. A one-month commitment guarantees you get whoever happens to be free.
Sizing your first month
Do not start from a budget. Start from what you have already sold:
- List every client deliverable you owe in the next 30 days.
- Convert each one using the table above, taking the top of each range.
- Add 15% for the work you have not thought of yet. There is always some.
- Subtract whatever your in-house team will genuinely deliver — not what you hope they will.
The remainder is your capacity number. Most agencies of eight to twenty-five people land between 20 and 40 credits a month once they stop being optimistic about step four.
If the number you calculate is under ten credits, you do not have a capacity problem — you have an occasional overflow problem. Buy a one-off project instead and revisit in a quarter.
The point of publishing the conversion is not transparency for its own sake. It is that a retainer both sides can predict survives longer than one that only works when nobody looks closely at it.
