Every independent consultant eventually receives the advice that retainers are the answer. Predictable revenue, less selling, compounding client knowledge, a valuation multiple if you ever sell. It’s presented as an obvious upgrade from project work, and consultants who hear it enough will convert a happy project client to a retainer and then spend eighteen months wondering why the relationship soured.
The advice isn’t wrong. It’s incomplete. Retainers and project fees are not two prices for the same work — they’re two different commercial products with different risk profiles, different scope dynamics, and different failure modes. A retainer applied to work that is structurally project-shaped produces a client who feels overcharged and a consultant who feels overworked, which is an impressive achievement in a single arrangement.
This is about which structure fits which work, and how to tell before you sign.
What Each Model Actually Sells
A project sells a defined outcome. A retainer sells access and continuity.
That’s the whole distinction, and most pricing disputes come from ignoring it.
In a project, the client is buying a result: the market analysis, the migrated system, the new pricing model. They can describe what “done” looks like. The risk of it taking longer than expected sits with you — that’s what you’re compensated for in the margin.
In a retainer, the client is buying your ongoing availability, your accumulated context, and a claim on your attention. The deliverable is not a thing; it’s a relationship with a defined shape. The risk that this month has more work than last month is shared, and it’s smoothed across the term.
The test: can you write the acceptance criteria?
If you can write down what “finished” means in a sentence the client would sign, it’s a project. Price it as one.
If the honest answer to “when is this done” is “it isn’t — the situation keeps evolving and I keep responding”, it’s a retainer. Price it as one.
The trouble starts when consultants use a retainer to price something that has clear acceptance criteria, usually because retainers sound better. The client then measures the retainer against the deliverable — “we’re paying €3,000 a month, what did we get this month?” — and you’re in a value-justification conversation every thirty days, forever. That is the worst of both models.
The reverse mistake is also common.
Pricing genuinely continuous work as a series of projects generates enormous sales friction. Every month becomes a scoping conversation. You spend a quarter of your billable capacity writing proposals for work you’re already effectively doing. This is the state most consultants are actually in when they’re told retainers are the answer — and in that case, they are.
The Cash Flow Maths
Project cash flow is spiky and the spikes lie to you.
A project practice at €120,000 a year does not receive €10,000 a month. It receives €35,000 in March, nothing in April, €18,000 in May. The average is fine and the variance is what determines whether you can pay yourself in April.
Three structural problems with project cash flow that are worth naming:
- The selling gap. You cannot sell effectively while delivering at capacity, so a full pipeline in Q1 produces an empty one in Q2. The sawtooth is not bad luck; it’s the natural consequence of one person doing both jobs.
- Payment timing. A project invoiced on completion and paid at 30 days means cash arrives 4-6 weeks after the work ends and 3-4 months after it started. You have financed the client for a quarter.
- The forecast is fiction beyond 60 days. You cannot plan hiring, tooling, or your own life against it.
Fix what’s fixable within the model before concluding you need retainers. Milestone billing — 40% on signature, 30% at a defined midpoint, 30% on delivery — solves most of the payment timing problem and costs you nothing but the nerve to ask for it. A deposit is standard in every other professional service and consultants are strangely reluctant to require one.
Retainer cash flow is smooth and it hides utilisation.
A retainer at €3,000/month paid in advance is genuinely better cash: predictable, financed by the client rather than by you, and it lets you plan.
What it hides is whether you’re making money on it. In a project you know the fee and you can measure the hours. In a retainer, the hours drift upward slowly — a call here, a quick question there — and the effective hourly rate erodes over eighteen months without a single moment where anyone decides anything.
The only defence is measuring it. Track hours against every retainer, monthly, even though nobody’s asking you to. When the effective rate has dropped 30% below your project rate, that’s the conversation to have — and it’s a conversation you can only have if you have the number.
The real financial argument for retainers is not the smoothing.
It’s the sales cost. A project practice spends 20–30% of capacity on selling. A retainer practice spends that capacity on delivery, because the revenue renews rather than being re-won. At the same headline rate, the retainer practice is meaningfully more profitable — not because the rate is better, but because more of the time is billable.
That’s the actual case. It’s a good one. It just isn’t the one usually made.
Scope Control — Where Both Models Break
Project scope creep is visible. Retainer scope creep isn’t.
In a project, scope creep has an event. The client asks for something outside the statement of work, and there is a specific moment where you either say “that’s a change order” or you don’t. You may handle it badly, but you can see it.
Retainer scope creep has no event. Month one: two strategy calls and a document. Month six: two strategy calls, a document, a standing Tuesday check-in, ad-hoc Slack access, and you’re reviewing their supplier contracts because you were there. No individual addition warranted a conversation. Cumulatively you’re delivering double.
Define the retainer as a container, not as a fee.
The retainers that hold up define the shape explicitly:
- Hours or capacity — “up to 20 hours per month” is honest but turns you into a timesheet vendor, and clients start counting.
- Outputs — “one strategy session, one written analysis, and unlimited email” is better. It defines the shape without metering the minutes.
- Response commitment — “questions answered within one business day” is often the thing the client actually values most, and it costs you the least. Sell it explicitly.
- Explicit exclusions — implementation, travel, work for other departments, anything requiring more than X hours in a week. Name them.
- The overflow rule — what happens when a month is genuinely bigger. Either it rolls, or it’s billed at a defined rate, or you flag it and agree. Decide before it happens, because deciding during is a negotiation you’ll lose.
Do not roll hours indefinitely.
Unused retainer hours that accumulate forever create a liability that eventually gets called in during your busiest month. Cap the rollover at one month, or don’t roll at all — the retainer buys availability, and availability you didn’t use was still provided. That’s a defensible position if you state it upfront and an indefensible one if you spring it in month nine.
Review the container quarterly.
Put a scheduled review in the contract. Not to raise the price — to re-examine what’s actually happening. This normalises the conversation about scope, so that when the work has grown 40% you’re having a routine quarterly discussion rather than an awkward confrontation. The review is the mechanism that keeps a retainer honest over years rather than months.
Choosing, Per Engagement
Price the project when:
- The client can describe the finished state
- The work has a natural end
- You’ve done something similar and can estimate within 25%
- The client is new and neither of you knows if this will work
- The value is concentrated in a deliverable rather than in your availability
Price the retainer when:
- The work is genuinely continuous and reactive
- The value is in your accumulated context — you know their business, and a new consultant would need three months to catch up
- The client’s need is unpredictable in timing but predictable in volume
- You’ve already worked together and both know the shape
- Speed of access is what they’re actually buying
The sequence that works: project first, retainer second.
Almost always. A first engagement should be a defined project, because it’s a bounded risk for both sides. You learn whether they’re a client you want. They learn whether you’re worth the money. Nobody is committed to twelve months of a relationship neither has tested.
Then, if the project reveals ongoing need — and it usually does — the retainer conversation happens from a position of demonstrated value rather than promised value. That conversation converts at a rate that cold retainer pitches never approach.
The hybrid that’s better than both.
For most independent consultants the strongest structure is a base retainer plus project work on top. The retainer covers the continuous layer — the availability, the standing review, the accumulated context — at a level small enough that the client never questions it. Discrete initiatives are priced as projects against that foundation.
This gives you a revenue floor that covers your fixed costs, plus upside that scales with client need, plus a scope boundary that’s naturally enforced because anything substantial is visibly a project. It also solves the “what did we get this month” problem, because the retainer isn’t pretending to be a deliverable.
The same logic that makes recurring revenue work for distributors applies here: the base isn’t the profit centre, it’s the thing that makes the relationship durable enough for the profit centre to exist.
Making the Move
Don’t convert an existing client at the same effective rate.
The most common conversion error: a client paying €30,000 a year in projects gets offered a €2,500/month retainer. Same money, and now they have unlimited access. You have just given away the option value for free.
A retainer should be priced on what continuity and access are worth, not on what last year’s project total was. If the retainer replaces project work, it should either cost more per unit of output or explicitly cover less.
Pitch it as risk transfer, because that’s what it is.
The honest client-side pitch is not “this saves you money.” It’s: “you get a defined response commitment, you don’t re-negotiate every time something comes up, and I’m not starting from zero context each time.” Those are real and they’re worth paying for. Pretending it’s cheaper is both false and unnecessary.
Start the term short.
Three months, then quarterly. A twelve-month first retainer is a large commitment from a client who has never bought this shape from you and it slows the close considerably. Short terms with easy renewal close faster and, counterintuitively, last longer — because the renewal is a moment where value gets re-confirmed rather than resented.
Watch the concentration.
Retainers create comfort and comfort creates concentration. A practice where one retainer is 50% of revenue is not a stable practice — it’s an employment relationship with worse benefits and no notice period. The predictability that made retainers attractive becomes the thing that stops you selling, and then the client restructures and you have a very bad quarter.
Keep any single client under roughly a third of revenue, and keep some project capacity open specifically so you’re still in the market. The operational discipline behind this is the same one that keeps a minimal tech stack minimal and a consultant’s LinkedIn presence active during good quarters — the work you do when you don’t need it is what stops you needing it.
The question is never “retainer or project” in the abstract. It’s “what is this specific client actually buying, and does the price structure match it.” A client buying an outcome should pay for an outcome. A client buying your availability should pay for availability. Getting that match right removes most of the friction that consultants attribute to difficult clients.
And if you’re currently all-project and exhausted by the selling: the fix probably is a retainer layer. Just build it on a client who has already paid you for a project, price it on what access is worth rather than on last year’s revenue, define the container in writing, and measure the hours from month one — so that in month eighteen you’re having a data conversation instead of a resentment one.
Sources: European Commission — late payment in commercial transactions · EUR-Lex — Late Payment Directive 2011/7/EU