Hiring an agency for a website rebuild: why start with discovery?
By Nicola Sirchia. Published . 3 minute read.
The short answer
Don't buy a whole website rebuild on the strength of a pitch. Buy a short, paid discovery first. In a few weeks it gives you customer interviews, an honest look at your data, first requirements and a roadmap, and it shows you how the agency really works before you sign the big contract. If the discovery is weak, you have lost weeks instead of a year.
What goes wrong with the usual pitch?
The agency shows concepts before it knows your customers. You choose the prettiest slide, and the next twelve months are spent building on guesses. Then the people who pitched hand over to a team you never met.
What should a discovery produce?
Five things, and I would not start the build without them.
- What customers and stakeholders said, in their own words, from interviews with both.
- What your data can and cannot tell you today, and what is missing.
- First requirements, agreed with the people who will use the site every day.
- A roadmap for content, site structure, search, technology and experience, with what to migrate, rewrite or retire.
- A creative direction and first concepts, built on what was learned, not before it.
Who has to be in the room?
More people than you think. Leadership sets the direction. Marketing and communication bring the brand and the stories. Digital and product know what the platform must do. And the business lines, the people who sell and serve, are the ones most often forgotten, and the ones who decide whether the site is used after launch.
In the middle sit three things every group has to share: the brand guidelines, what customers actually told you, and how success will be measured. If any of the three is missing, each group fills the gap with its own opinion.
How do you judge an agency in a pitch?
These are the questions I listen for.
- Did they talk to real customers before the meeting? A pitch built on real customer interviews is worth more than any concept board.
- Is the team in the room the team that will do the work? Ask to meet the people who will run it day to day.
- Do they talk about the people who will publish on the site? A site that only the agency can change is a dependency, not an asset.
- Do they bring measurement up without being asked?
- Do they propose a rhythm? Daily with the working team, weekly with the whole team, quarterly with leadership. If there is no rhythm, there will be surprises.
How would I write the contract?
Discovery as its own phase, with a fixed price and a clear exit. If it delivers, the build runs in sprints with a demo every few weeks. If it does not, you stop, keep everything that was produced, and choose again with far better information.
Where would I start?
- Write one page on what the site must do for the business and for customers.
- Invite three agencies, and ask each to talk to two of your customers before they pitch.
- Buy discovery from the best one, with a fixed price and an exit.
- Decide on the build only when you have read the interviews yourself.
Sources
This piece comes from my own work selecting and briefing agencies for website rebuilds. It uses no outside data.