How to choose a web development company
The questions that separate a good fit from an expensive mistake, and the answers that should worry you. Written by an agency, which you should factor in.
We are a web development company, so treat this as a set of questions worth asking rather than a neutral guide. We have tried to write it so that we fail some of them — because a checklist that only ever points at its author is not a checklist.
Work out which kind of buyer you are first
Most bad hires are a category error rather than a bad supplier.
Defined and small. You need a website that says what you do, shows proof, gives prices and takes enquiries. You know roughly what it should contain. A fixed-price service fits, and an agency will overcharge you for process you do not need.
Defined and large. A platform, a shop, an integration with something you already run. The scope is knowable but substantial. You want a firm that does discovery properly and prices the risk.
Undefined. You know the problem, not the product. What you need is a paid discovery engagement ending in a scope and an estimate — not a build quote from someone guessing.
Buying the wrong category is how people end up paying agency rates for a brochure site, or handing a genuinely complex product to whoever was cheapest.
The questions
What is not included? The single most useful question. Good answers are specific and slightly uncomfortable. Vague answers become invoices.
Who will actually do the work? Names and roles. It is normal for the people who pitch not to be the people who build. It is not normal to be evasive about it.
Can I see something you shipped in the last six months? Not the best case study — the most recent. Portfolios age, teams change.
What happens if it runs over? On a fixed price, that is the supplier's problem. If it becomes yours, the price was an estimate wearing a costume.
Do I own the code, the domain and the accounts? Yes, from the start, is the only good answer. Anything complicated here is a reason to walk.
What does a change cost after launch? Ask about a realistic one: new prices, new hours, a new page. If the answer is hourly, assume you will stop asking, and assume the site will drift.
What do you need from me, and when? A supplier who has thought about this will have a list. One who has not will blame you for the delay later.
Who is my point of contact when something breaks? And what is the response time. Get it in writing.
Answers that should worry you
- "It depends, let's get on a call" — when you have already described the project. Sometimes genuine, often a technique.
- "We'll handle the domain for you" with no mention of whose name it goes in.
- A quote with no exclusions. Everything is excluded from something.
- "We can do all of that" to a list of unrelated capabilities. Nobody is good at everything.
- Pressure to sign quickly. Discounts that expire on Friday are a sales mechanism, not a price.
- A portfolio you cannot verify. Live links, or ask why not.
Where we fail this checklist
Because we should say so.
We do not do discovery. We sell one defined thing — a small business website at $499 plus $49 a month — and if your project does not fit that shape, we are the wrong supplier. We do not build online shops or booking systems at that price. We build mobile apps, but not at a fixed price.
If you are the second or third kind of buyer above, you want a firm that does the thing we deliberately do not: sit down with you, work out what should be built, and price it properly. Ask them these questions instead.
The short version
Work out which kind of buyer you are. Shortlist three suppliers in that category, not three of different sizes. Ask all of them the same eight questions. Compare the exclusions, not the totals.
The cheapest quote and the most expensive quote are usually for different projects. That is the thing to find out before you sign, not after.