If you sell something technical, you have probably watched a good conversation stall after the engineer on the call is satisfied and the person signing the cheque has gone quiet. Your website sits between those two readers, and most technical sites are written for the engineer, because an engineer wrote them. There is a way to fix that without dumbing anything down.
Two people are reading, and only one of them is technical
In deep tech and specialist platforms, the buying group is almost always split. There's an evaluator who understands the method and the accreditation, and there's an approver, often a commercial director, who has to justify the spend to a board and will never open your technical appendix. That second person forms a view in three sentences.
We saw that shape working with XOCEAN, who collect ocean data using uncrewed surface vessels. The engineering audience wanted survey specifications, sensor payloads and endurance figures. The commercial audience, offshore wind developers and their project managers, wanted to know whether this was a real operational capability, and that question decides whether the first gets asked.
The usual response is to pick a side. Either the site becomes a specification sheet that reads as impenetrable to anyone outside the field, or it gets flattened into vague claims about innovation that make the evaluator suspicious. The flattened version does more damage, because engineers can tell within a paragraph whether you do the thing.
Ordering the detail rather than removing it
There's a distinction worth holding onto here. Simplifying means taking detail out, while making something legible means ordering it so a reader can find their own level and stop there, with everything below still available. Those two instincts produce very different websites, and the second is harder to sell internally.
A legible technical page usually works in three layers. The first is a plain statement of what the thing does and what it replaces, written so a smart outsider follows it without a glossary. The second is operating detail: how it works, where it has been used, what the constraints are. The third is reference material almost nobody opens and that must be there.
The layers let you write honestly at every level, because a plain language door in front of the technical claim doesn't weaken it. When Dave works on this with clients, the hardest conversation is the first layer, since the person who knows the technology best finds that painful. It reads as lost precision, when it gets the second reader in far enough to care.
Write the sentence before you touch the site
Here is something you can do this afternoon without hiring anybody. Write one sentence saying what you do, what it replaces and who it's for, with no adjectives and no category claims in it. Then send it to a customer, to someone in your industry who isn't a customer, and to a complete outsider.
Ask each of them what they think you sell. If the outsider gets it roughly right and the first two don't wince, you have something usable. If your customer says it undersells you, ask what specifically is missing, then decide whether that belongs in the sentence or in the second layer.
Most of the positioning work we do starts here, and it's the part clients can do themselves before spending anything. One founder arrived with the sentence already tested, which took two weeks out of the schedule. On another project it took four sessions, because the company was doing three things and hadn't decided which one it led with.
Credibility signals that carry weight with a non-technical reader
The approver can't assess your method, so they assess proxies instead: who else uses this, how long you have been operating, whether the language sounds like people who do the work or people who market it, and whether anyone has touched the site recently. They're crude instruments, and without technical judgement they're what your reader has.
Nordic Aviation Capital sits at the far end of this. Aircraft leasing carries an enormous amount of underlying detail, and the audience includes airlines, banks and institutional investors who each need a different slice. That work was largely about structure, giving each audience a legible route through the business. We do a lot of it through Brainstorm, our Irish practice, established in 2010.
The practical version of this is fairly unglamorous. Name your customers if your contracts allow it, date your content, put a photograph of a real deployment on the page instead of a stock render, and say how long you have been doing the work. If you hold a certification that costs money to maintain, name it.
Where the engineers get lost
The failure running in the other direction is real, and we've caused it ourselves on early projects. You write the plain language layer well, the commercial reader gets it, and then the technical evaluator arrives, finds nothing to chew on, and concludes you are a reseller. Where that evaluator holds a veto, the conversation ends quietly.
Two things prevent it, and the first is keeping the deep material actually deep, written by or with the people who do the work rather than smoothed over by a copywriter. Specification tables, method notes and field data all count. The second is linking to it visibly from the plain language layer, so a technical reader sees the door.
The tone of that deeper layer should be flatter than your marketing pages, with fewer claims and more description. Engineers tend to read confidence as a mild warning sign and specificity as reassurance, which is roughly the reverse of how the commercial layer works. Writing both layers in the same voice is the most common error.
What this costs and how long it takes
For a technical company with a defined audience and existing material, a positioning and website project here typically runs $20,000 to $35,000 over twelve to sixteen weeks. That covers the positioning work, the structure, writing the plain language layers with your team, then design and build. Branding alongside it pushes the timeline out to around twenty weeks.
The variable that moves the number most isn't design, it's how much of your technical content already exists in usable form. If your engineers have written internal documentation or method descriptions, we work from that and the writing phase is short. If it's all in people's heads we are running interviews, and that time with busy staff is the commonest cause of a slipping schedule. There is more at the Irish studio's content strategy page.
Quite often a full project is more than the situation needs. If your positioning is sound and the problem is that the site buries it, a restructure and rewrite on your existing build can land between $6,000 and $12,000 in four to six weeks. We'd rather say that than sell a rebuild on unresolved positioning.
A short list of things to check on your own site
Open your homepage and read the first two sentences aloud to someone outside your industry, then ask them what you sell. After that, look at whether any customer name appears above the fold, whether there is a route from that opening to technical detail within one click, and whether the newest dated item is from this year.
Then check the reverse direction, and find your deepest technical page. Ask whether someone in your field would find anything there they couldn't have guessed, because if the answer is no, that page is decoration rather than evidence, and it is the page your evaluator will judge you on.
Most of what goes wrong on technical sites is structural rather than aesthetic, so the checks above are worth doing before you talk to anybody about design. If you work through them and the site still isn't carrying both readers, that's the point where a conversation about a project makes sense. Dave takes those calls himself, and a fair number end with us saying the problem is elsewhere.