A designer, a frontend and a backend developer. All three deliver what you asked for, the project still runs late. What that costs, and when one is cheaper.
Written by Aleksi Huusko
Published at 7/30/2026
Most founders end up with three contractors: a designer, a frontend developer, and someone who knows the backend. All three are good, and all three deliver what you asked them for. The project still runs late, and you cannot point at whose fault it is.
What happened is that you hired three specialists and accidentally hired yourself into a fourth role. Nobody quoted you for that one.
Write down what you actually did last week on the project. Not the strategy work, the mechanical parts.
You explained the same context twice, once to the designer and once to the developer. You decided whether a screen that looked fine in Figma but broke at 400 pixels was a design problem or a build problem. Somewhere in there you chased a file, and you answered a question about the API shape that the designer needed and the backend contractor already knew. Then you looked at a deployed page, felt that it was slightly wrong, and could not say which of the three to send it back to.
That is a job. On a normal build it is somewhere between a day and two days a week, and it lands on whoever cares most, which is you. If your time is worth anything to the business, the cheap version of the project was not cheap.
The most common one is that the design file and the code disagree and nobody decided which of them wins. Your designer updates the file. Your developer already built the old version. Six weeks later nobody can say whether the file or the site is the specification.
Then there are the states between the screens. Designers deliver the screen that looks good. Loading, empty, error, and the very long name that breaks the layout are usually not in the file, so the developer either invents them or ships without them, and both answers are wrong in a way you only notice after launch.
Responsiveness gets treated as a build detail. The designer drew two widths, real devices are a continuum, and everything in between becomes someone's improvisation.
The last one is the expensive one. Each contractor delivers their piece, and the place where the pieces meet is nobody's deliverable, so it becomes yours by default.
Genuinely often, and I would rather you get this right than hire me.
Several works when the work splits cleanly into pieces that barely touch. A logo, a copy pass, and a Webflow build can run in parallel with almost no interface between them. It also works when you need depth one person cannot cover, like real infrastructure work next to real brand design, or when the project runs long enough that a single person becomes a bottleneck or a risk.
And it works if you have someone in-house who is genuinely good at running contractors, has the time, and is not also the person selling. That role is the whole difference between a group of freelancers and a team.
One person is cheaper when the interface between design and implementation is the hard part rather than either side of it.
That covers most marketing sites and most rebuilds where the previous attempt stalled. It also covers product surfaces where the interaction detail matters. The hard question in those projects is rarely "what should this look like" or "how do we build it." It is "what happens when someone with a long company name signs up on a narrow phone." One person holding both sides answers that in the moment instead of scheduling a conversation about it.
Day rate on its own is the wrong number. Add the coordination hours you stop spending and compare that.
Worth doing whether or not you ever talk to me.
Name the source of truth before work starts and write it down. Either the design file is authoritative and the code follows it, or the deployed site is authoritative and the file is a sketch. Either convention works, as long as somebody says out loud which one you are using.
Give the interface an owner who is not you. Pick one contractor, and pay them for integration explicitly, as a line item. Unpaid ownership is not ownership.
Ask for the unglamorous states in writing: loading, empty, error, too much text, too little text. Put them in the brief, because otherwise they arrive as opinions during review.
Make one person responsible for the deployed result rather than for their own artifact. "My part works" is not a useful status when the page is broken.
And before you sign with any of them, ask the buying-side questions. I keep twelve of them in a checklist, starting with who actually does the work and whether you will ever talk to them.
I do design and implementation myself, so the handoff does not exist. I bill 2,240 euros plus VAT per week, and a typical marketing site is two to five weeks. Recent projects are here.
That rate is higher than a single freelancer's and lower than an agency's. The comparison worth making is against your three contractors plus the day or two a week you currently spend holding them together.
If your project is the kind that splits cleanly, hire the three. If the seams are the problem, that is the thing I am actually for.
There is no official recommended folder structure for the Next.js App Router. The docs lay out the options and leave the choice to you, which is honest and completely unhelpful when the repo is empty.
A professionally built B2B marketing site in Europe runs roughly 2,500 to 25,000 €. What moves a quote between those numbers, and what I charge per week.
Which platform fits your B2B marketing site in 2026? A decision tree from a design engineer who ships both Webflow and Next.js, with real running costs.