How to choose a web developer
Most people choose a web developer on price and impression, because there is no obvious way to judge anything else. This is an attempt to give you that way, including the reasons you might skip me.
Commissioning a website is an awkward purchase: you pay up front for something you cannot evaluate until it exists, and if it turns out badly you often cannot tell whether the problem was the execution or the fact that you asked for the wrong thing. This covers the decisions you make before signing anything: who you are actually looking for, how to read a portfolio, what to ask, and who ends up holding the keys.
Agency, freelancer or a website builder
Builders like Wix or Squarespace make sense more often than developers admit. If you need a presence with a contact, a description of your service and a few photographs, and you are willing to assemble it yourself over a weekend, you will pay around ten euros a month and be finished. The limits arrive later: speed is harder to optimise, leaving the platform is harder, and anything the tool did not anticipate is difficult to build.
An agency makes sense when a project needs several people: a designer, a developer, somebody for the copy and somebody running it. You are also paying for that coordination, which is justified on a large project and expensive on a small one. A freelancer sits between: you work directly with the person doing the work, with no margin on somebody else, but that person is also the bottleneck when they are busy or away. None of the three is universally right, and if somebody tells you theirs is, that is the first thing you have learned about them.
What to look for in a portfolio, and what to ignore
Portfolios are built to impress, so it pays to look past the aesthetics. Open two or three of the sites on a phone using mobile data. Time the wait. Check whether the menu works, whether you can reach a contact in one move, and whether the text reads like somebody wrote it for that specific business.
Ignore the project count and the list of technologies. Thirty sites that are all the same template tell you less than three that solve three different problems. It is also worth checking whether those sites are still running: a portfolio full of dead links means the clients did not stay, and that is information.
Questions that reveal the level
You do not need to understand the technology to recognise a serious answer. Ask who will change the content after handover, and listen for whether they ask you questions back. Somebody who proposes a solution immediately, without asking anything about your business, is selling a package rather than a solution.
Other good ones: what happens if I need five more pages in a year, how do you measure whether the site is working, and what would you do differently on your last project. That last one is the most useful, because somebody with no answer either has not done enough projects or does not think about them after handover.
Red flags
A guarantee of first place on Google is the clearest of them. Nobody controls somebody else’s algorithm, so it is a promise that cannot be kept. A price quoted immediately, without a single question about scope, is close behind: it means you are getting a pre-assembled package rather than an assessment of your project.
The rest is subtler. A quote that never mentions what happens after handover. Reluctance to say which system will be used. A timeline that sounds too short for what you described. And most importantly, anybody who does not want the domain and hosting in your name.
Who holds the domain, hosting and code
This is the part that gets asked least and hurts most later. The domain should be registered to you or your company, not to the developer. The same goes for hosting. If both are in somebody else’s name, changing supplier stops being a business decision and becomes a negotiation.
The same applies to access. You should have an administrator account on your own site, access to the hosting, and, if the site was custom-built, the code somewhere you can reach it. None of that is controversial to anyone serious, because they were not planning to hold you hostage. If it is controversial, you have your answer before signing anything.
A short list before you sign
- Open two portfolio sites on a phone and time how long they take to load
- Ask who will change the content, and through what system
- Require that the domain and hosting are in your name
- Ask what is billed after handover, and at what rate
- Check that the price came after questions about scope, not before them
The decision at the end
If everything passes this list, what remains is a decision more about the person than the technology: will you be able to talk to them when something goes wrong. A website is not a one-off purchase but a relationship that lasts as long as the site does, so somebody who honestly tells you that you do not need their service is worth more than somebody who agrees to everything. And if you are still unclear about what exactly you need, that is normal, and it is the question a conversation should start from rather than one you have to solve before it.