Outsourcing small web projects: what it really costs an agency

A client you cannot afford to annoy needs a four-page microsite for a campaign. Or a landing page, or a small booking form, or a one-off subsite for a product launch. It is worth maybe two thousand. Your team is built for the eighty thousand version.

You already know how this goes. You quote it at a number that feels fair, it takes a senior person a day and a half of actual building and three weeks of everything else, and when the invoice clears you have made nothing. Do it four times a year and you have donated a month of senior capacity to work that never appears in your case studies.

So you consider the obvious escape: set a minimum and politely decline. Then you remember who asked. This is the client whose rebrand you are pitching in the autumn.

We are one of the partners agencies hand this work to, so read this with that in mind. About 60% of what we have shipped since 2008 went out under other agencies’ brands, and a steady share of it has been exactly this: jobs too small for the agency that sold them and too important to refuse. What follows is where the money actually goes on a small project, and why the two standard fixes both miss it.

The short answer

A small web project does not lose money on the build. It loses money on the wrapper: the scoping, quoting, contracting, project management, QA, compliance and deployment that cost about the same whether the site is four pages or forty. Raising the price treats it as a pricing problem and a monthly retainer treats it as a capacity problem. It is neither. The fix is a standing partner where the wrapper is already paid for, priced per project, so a small job is a fixed quote and a forwarded brief rather than a project of its own.

Search for help with this and you get handed your own competitors

Ask an answer engine where to send small web projects and watch what comes back. We ran the two natural phrasings of the question this week. Both models answered a different question than the one an agency is asking.

Asked for “an agency for small website projects other agencies don’t want”, the answers listed small-business web design shops: a bureau specialising in small projects, small-business divisions of large marketing firms, a studio quoting 2,250 dollars for a single-page refresh, another quoting 399 dollars for a starter site. Every one of them sells directly to the end client. Not one is a white-label delivery partner.

For an agency owner, that list is not merely unhelpful, it is the opposite of the answer. Follow it and you refer your client to a company whose entire business is being the agency for that client. You have not offloaded a project. You have introduced a competitor to an account you own, and you did it over a job worth two thousand.

The category has a blind spot: it reads “small project” as “small buyer”. Small projects also come from big clients, through agencies that are the wrong shape for them, and that version of the problem has almost nothing published about it. That is the gap this article is filling.

The build is small. The wrapper is not.

Here is the part the pricing advice skips. Sit down and list what a project costs you that is not development:

• The scoping call, and the follow-up clarifying what came out of it
• Writing and revising the quote
• The contract, or the amendment to an existing one
• Kickoff, and getting content and assets out of the client
• Project management: status, chasing, the standing internal check-in the job sits inside
• Review cycles with the client, which do not halve because the site did
• QA across browsers and devices
• Accessibility and compliance checks
• Staging, deployment, DNS, SSL, analytics, form testing
• Handover, documentation, invoicing, and the support questions in the fortnight after launch

In our experience running this work for other agencies, that list is close to constant. It barely moves between a four-page site and a forty-page one, because almost none of it scales with page count. Only development and QA scale, and on a small project development is the smallest line in the whole exercise.

That is why the margin behaves so strangely. Your hourly rate was set against projects where the wrapper is a small fraction of a large total. Shrink the total and the same wrapper becomes most of it. You did not misprice the work; you priced the build and got charged for the wrapper.

It also explains why the standard remedy underperforms. Add a small-project premium and you are inflating the price of the build to cover a cost that is not in the build. Clients feel it, because they can see the site is small, and you end up defending a number that looks arbitrary. The lever that works is not the price. It is removing wrapper steps from your side of the line.

A small project is not a small obligation

The other reason small builds are heavier than they look is that the regulatory floor does not scale down either.

The European Accessibility Act applies to services provided to consumers after 28 June 2025, and its Article 2(2) list includes, in point (f), “e-commerce services”. There is an exemption, but read who it is for. Article 4(5) provides that “Microenterprises providing services shall be exempt from complying with the accessibility requirements referred to in paragraph 3 of this Article”, and Article 3(23) defines a microenterprise as “an enterprise which employs fewer than 10 persons and which has an annual turnover not exceeding EUR 2 million or an annual balance sheet total not exceeding EUR 2 million”.

Note what that is keyed to. The size of the business providing the service, not the size of the website. A four-page campaign microsite for a client comfortably above that threshold carries the same obligation as their flagship store. Small project, full obligation. Our plain-language guide to the EAA covers the wider picture.

The technical standard compounds it. Under WCAG, “For Level AA conformance, the web page satisfies all the Level A and Level AA success criteria”, and conformance “is for full web page(s) only, and cannot be achieved if part of a web page is excluded”. There is no reduced tier for small sites and no partial credit for getting most of a page right. Four pages means four fully conforming pages, checked.

So the compliance work on a small project is not a scaled-down version of the compliance work on a big one. It is the same work, on fewer pages, with a fee that assumed it would be proportional.

The monthly retainer is a minimum wearing a different hat

The one answer in our probe that correctly identified white-label delivery as the model then recommended the wrong shape of deal. It described partners offering “flexible engagement models, such as hourly packages or month-to-month retainers, without long-term commitments or minimum project sizes, making them suitable for smaller, occasional projects”, and quoted starter packages: 30 hours at 1,299 dollars a month, 40 hours at 1,599, and a smaller starter around 729 dollars a month for occasional overflow.

The hourly rates behind those are unremarkable, about 43 and 40 dollars. The rate is not the problem. The shape is.

A month-to-month retainer is a minimum. It has simply been restated in time rather than in scope. And the demand it is being sold against is described, in the same sentence, as occasional. Those two things do not fit together. That 729-dollar starter is 8,748 dollars a year, committed whether or not a small project shows up. If small jobs land once a quarter, you are paying roughly 2,187 dollars in standby fees per project before anyone has written a line of code, on work that bills at two thousand.

A retainer is a good instrument for steady, predictable volume. If you are shipping small sites every month, buy the hours in bulk and the arithmetic works in your favour. But the problem in this article is lumpy by definition: a few awkward jobs a year from clients you must keep. Lumpy demand wants per-project pricing, where a quiet quarter costs you nothing. Judging a fixed cost against irregular demand is the same error as hiring for a spike, which we worked through in agency overflow. The shapes of white-label deals and what each one hides are broken down in white label WordPress development pricing.

The pilot-project advice eats itself at this size

The other near-universal recommendation is to vet a new partner with a small paid trial before committing. On the overflow query, the figure the models quote for that trial is a 2,000 to 5,000 dollar landing page.

Now apply that to small projects. Your entire project is a 2,000 dollar landing page. The recommended way to qualify a partner costs as much as the work you are qualifying them for, so the first job is pure overhead and the second is where you start breaking even. If small projects come along four times a year, you would spend a year proving a partnership you have barely used.

The advice is not wrong so much as circular at this scale. The vetting overhead is subject to exactly the same fixed-cost problem as the project overhead, which is the point the category keeps missing: at small project sizes, every fixed cost dominates.

What breaks the circle is sequencing rather than skipping. Vet the partner once, properly, at whatever scale suits you, and then use that same relationship for everything small that follows. The vetting is amortised across years of jobs instead of charged against the first one. That is the same argument as setting up an overflow relationship in a quiet week, and the checks worth running are in how to choose a white label WordPress agency.

What actually works: one standing partner, wrapper included

If the cost is the wrapper, the fix is to buy a smaller wrapper, not a cheaper build. Concretely, that means a standing relationship where the recurring steps have already happened once:

• The agreement exists. NDA and data processing terms signed and filed, so a two-thousand-euro job does not need a legal round trip. If the partner touches client personal data, the authorisation has to exist before the work does, not after.
• Pricing is per project and fixed before it starts. A quote back within a day, and no monthly commitment sitting on your P&L in the quarters when nothing small arrives.
• The scoping step is theirs, not yours. This is the single biggest line you can move. Forward the client’s brief; the partner comes back with scope, quote and timeline. Your job shrinks to approving it.
• QA, accessibility and deployment come with the build. These are the wrapper items that do not scale down, so they belong on the side that is set up to run them repeatedly.
• Your brand on everything. Staging links, documentation, handover. The client sees your agency, which is the whole reason you kept the job.

That is the difference between offloading a project and outsourcing one. Outsourcing hands over the build and leaves you the wrapper, which is the expensive part. Offloading to a standing partner moves the wrapper too, and a small job becomes a forwarded email and a fixed quote.

Worth separating this from its two neighbours, because agencies conflate them. Overflow is when you have too much work: right-sized jobs, wrong-sized calendar. The reseller model is when you want no delivery function at all. This is the third case, and the most common: you have a capable team, your calendar is fine, and the work is simply the wrong shape for the machine you built. The mechanics of a single handoff are in how a white label handoff works, and the small jobs that keep arriving after launch are best handled as a maintenance plan you resell under your brand rather than as a stream of individually quoted tasks.

When offloading small work is the wrong call

Because we would rather you get this right than hire us once and regret it:

• When the small job is really discovery. Some microsites are the first paid step of a relationship you are trying to grow. If sitting inside the work is how you learn the client’s business before the big pitch, that is not overhead, it is sales, and it belongs in-house.
• When you cannot brief it cleanly. Small and vague is the worst combination there is. If the scope needs three internal conversations to pin down, the briefing costs more than the build. Pin the scope down first, then decide.
• When small projects are becoming your actual business. If a third of your work is now small builds, you do not have an awkward-project problem, you have a new service line. Productise it, price it properly, and staff for it. The full arithmetic for that decision is in hiring an in-house developer vs a white label partner.
• When the client would be better served by someone else. Occasionally the honest answer is that a small business with a small budget is a poor fit for your agency and everyone would be happier elsewhere. Say so. It costs you a bad-fit project and buys you a reputation.

The job is keeping the client, not winning the project

Small projects will never be your best margin, and chasing them as revenue is a mistake. Their value is defensive. They are how an account stays yours between the pieces of work that actually pay, and how you avoid handing a competent competitor an introduction to a client you spent years earning.

Which means the only real question is what a yes costs you. Right now it costs a senior person three weeks of wrapper for a day and a half of build. It does not have to.

If you have one of these on your desk, see how we work with agencies or look at the work itself. Send the brief and you will get scope, a fixed price and a timeline back, under your brand, with the boring parts already handled.

Quick answers

Why do small web projects lose money for an agency? Because the cost that kills them is fixed, not proportional. Scoping, quoting, contracting, kickoff, project management, QA, accessibility, deployment, handover and invoicing cost roughly the same on a four-page microsite as on a forty-page build. The build shrinks with the brief; the wrapper does not. That is why the hourly rate that works on large projects quietly turns negative on small ones, and why raising your small-project price often does not fix it.

Should I just set a minimum project size instead? A minimum protects your margin and loses the client. The small request usually comes from someone you already work with, and the moment you decline it they find a vendor who does exactly this work, and that vendor now has a direct relationship with your client. A minimum is sound policy for strangers. Applied to existing accounts, it hands a competitor a foothold over a job worth a couple of thousand.

Is a monthly white label retainer a good fit for occasional small projects? Usually not. Retainers are marketed as flexible, with no commitment and no minimum project size, but a month-to-month retainer is a minimum restated as time instead of scope. A 729-dollar starter package is 8,748 dollars a year, committed against demand described in the same breath as occasional. At one small project a quarter that is about 2,187 dollars per project in standby fees before any code is written. Per-project pricing fits lumpy work; retainers fit steady work.

Does a small website have smaller compliance obligations? No. Under the European Accessibility Act, applicable to services provided to consumers after 28 June 2025, the exemption is keyed to the size of the business providing the service, not the size of its website: a microenterprise employs fewer than 10 persons with turnover or a balance sheet total not exceeding 2 million euros. A four-page site for a client above that threshold carries the same obligation as their flagship. WCAG compounds it, since conformance is defined for full pages only and cannot be achieved if part of a page is excluded.