How to recognize a good B2B web design agency
Key takeaways
- Rankings where the author is in first place are not a selection aid. They are advertising.
- Twelve criteria are enough to make providers comparable. You can check six of them before the first conversation.
- The most revealing question is not what someone can do, but what they turn down.
- A quote that asks more questions than it answers is almost always the better one.
What you can check before the first conversation
Six things can be assessed in twenty minutes and without contact.
Their own website. It is the only work sample that is completely under the agency's control. Does it load fast? Does it work on a phone? Does it say something concrete or only adjectives?
Case studies with substance. Not logo walls, but projects with a starting point, approach and result. If nowhere does it say what actually happened in a project, there is usually a reason.
Industry proximity. Not necessarily the same industry, but comparable complexity. Someone who has only done retail so far doesn't automatically understand a buying process with three decision-makers.
Reviews with a source. Google, Sortlist, Provenexpert: verifiable and dated. Quotes without a last name and company say nothing.
Stability. How long has the agency been around? How big is the team? With a single person, an absence is a real risk that you can take on consciously, but consciously.
Subject-matter traces. Does the agency write about its field? Anyone who regularly explains how they work usually has a method.
The question that reveals the most
Six things can be assessed in twenty minutes and without contact.
Their own website. It is the only work sample that is completely under the agency's control. Does it load fast? Does it work on a phone? Does it say something concrete or only adjectives?
Case studies with substance. Not logo walls, but projects with a starting point, approach and result. If nowhere does it say what actually happened in a project, there is usually a reason.
Industry proximity. Not necessarily the same industry, but comparable complexity. Someone who has only done retail so far doesn't automatically understand a buying process with three decision-makers.
Reviews with a source. Google, Sortlist, Provenexpert: verifiable and dated. Quotes without a last name and company say nothing.
Stability. How long has the agency been around? How big is the team? With a single person, an absence is a real risk that you can take on consciously, but consciously.
Subject-matter traces. Does the agency write about its field? Anyone who regularly explains how they work usually has a method.
The six questions for the initial consultation
1. What would you change first on our current website? A good answer is specific and names a priority. A bad one praises first and then dodges.
2. How many page types are you calculating, and why these? If they have no answer to this, they haven't thought the project through.
3. Who writes the copy? The answer decides a considerable part of the price and the result.
4. What happens to our existing URLs? If nothing about redirects comes up here, a central building block is missing.
5. What does operation cost in the first year after launch? The answer shows whether someone is thinking about the time after the project.
6. Which projects do you turn down? The most revealing question of all. Someone who does everything has no stance. A clear answer shows specialization and self-awareness.
The warning signs
Six patterns where caution is advised.
A price in the initial consultation, without follow-up questions. Anyone who names a number without knowing the scope, content and migration names a number that will go up later.
Ranking guarantees. Nobody can guarantee placements. Anyone who does is either selling nonsense or means terms nobody searches for.
Not a word about goals. If nobody asks during the entire conversation what the website should achieve, the result will look like it.
Only design in the portfolio. Beautiful images without saying what the project achieved. That is not a knockout criterion, but it shifts the risk to you.
Contract terms without an exit. Especially with "website for 99 euros a month" models. Calculate the full term once and check who owns the site at the end.
No direct contact with the people doing the work. If you only ever talk to sales, you don't know who is actually working on it.
The size question: freelancer, small agency, large agency
All three work, for different tasks.
Freelancer. Best value for money, direct line, high flexibility. Prerequisite: you take over project management and can live with the risk of one person dropping out. Good for clearly defined projects with a clear briefing.
Small to medium-sized agency. Several disciplines under one roof, project management included, absences covered. The usual path for mid-market projects between 15,000 and 50,000 euros.
Large agency or network. Strategic depth, experience with complex organizations, robust processes. In return, higher costs, longer channels and often changing contact persons. Makes sense for project sizes where coordination becomes a task in itself.
The decision shouldn't be based on likability, but on the question: How much project management can and do you want to take on yourself?
What the contract needs to cover
Four points that should be settled in writing before the project starts, and that make the difference in a dispute.
Usage rights. Who owns the design, code and content after the project ends? The answer should be "you", without restriction and without time limit.
Access. Who has access to the domain, hosting and CMS? You should always have your own accounts with full rights, not only once there is a dispute.
Revision rounds. How many are included, what does another one cost? This is the most common supplementary item of all.
Acceptance and warranty. When is the project considered accepted, and for how long are bugs fixed free of charge? Six to twelve months for defects in the build is common.
None of these points is unusual. Anyone who dodges all four is telling you something.
How to decide in the end
If two or three providers are comparable in terms of expertise, the criteria list alone won't help anymore. Then two things decide.
Does the agency disagree with you? A provider who likes every idea won't suddenly start steering in the project. Someone who says in the initial consultation "I would do that differently, and here is why" will take work off your hands later.
Do you understand what they are saying? If explanations sound like jargon and you couldn't repeat after the conversation what was proposed, the collaboration will be exhausting. Good providers explain their field in a way that is understandable outside of it.
Both sound soft and are, in practice, the best indicator of how the project will go.
More articles
