Retainer or project: which one fits

Almost every awkward conversation in a long-running client relationship traces back to a single unexamined decision: whether the arrangement was a project or a retainer, and whether both parties understood it the same way.

The distinction is not about duration or size. It is about what is being bought. A project buys a defined outcome — this exists at the end, and it did not before. A retainer buys availability: a certain amount of capacity, reserved, applied to whatever turns out to matter.

When each one fits

The pattern in the right column: projects suit work whose shape is known in advance, retainers suit work whose volume is known but whose shape is not.

The retainer that is really a discount

This is the most common failure mode in the category. A client and a supplier agree a monthly fee without defining what it buys, on the shared assumption that it will roughly work out. It roughly works out for a few months and then does not.

It fails in one of two directions and both are bad. Either the client uses very little and starts to feel the fee is unearned, or the client uses a great deal and the supplier starts to feel the fee is inadequate. Neither party did anything wrong; the agreement simply had no shape.

The unused capacity question

This deserves its own treatment because instinct points the wrong way. The intuitive answer is that unused hours should roll over, since the client paid for them. In practice rolling hours reliably breaks retainers.

What happens is that a balance accumulates over several quiet months, then gets spent in one demanding month — at which point the supplier is delivering three months of work in one while receiving one month's fee, and has no capacity left for anyone else. The arrangement that was meant to smooth demand has amplified it instead.

The workable positions are: hours do not roll over, and the fee is understood as buying reserved availability rather than a quantity of output; or hours roll over with a hard cap and an expiry, which limits the accumulation. What does not work is unlimited rollover, however reasonable it sounds when the agreement is signed.

What each side is actually getting

It helps to name the trade honestly, because the value is real on both sides and is often left implicit.

• The client gets a supplier who already knows the system, responds faster because there is no requoting cycle, and can be asked small questions without a commercial conversation attached to each one.
• The supplier gets predictable income, which lowers the cost of sale substantially and shows up as a better rate for the same work.
• Both get accumulated context, which is genuinely valuable and never appears on an invoice — the second year of any relationship is more efficient than the first.

That second point is why retainers can be priced better than the equivalent project work without either side losing. Predictability is worth real money to a supplier, and some of that saving can reasonably be passed on.

The simple test

Ask what happens if nothing needs doing next month. If the honest answer is that the fee is still fair because availability was the product, a retainer is right. If the answer is that it would feel like paying for nothing, the work is project-shaped and should be bought that way, however continuous the relationship feels.