We follow one problem through your website, working process and tools. You get a recommendation grounded in the evidence, then choose whether to commission the Build.
Your goals, the problems you feel, and whatever evidence you have: your site, how customers reach you, how the work runs today.
Diagnose delivers a decision pack: what we found, the evidence behind it, the important unknowns and our recommended next step. Up to three improvements, with owners, dependencies and a way to check the result. A separate Build scope and price where the evidence supports one.
Anything that sends, spends, publishes or can’t be undone. Nothing reaches your customers until you approve it.
It’s used only to do your work. The AI services Eren uses process it to produce your results. It’s never sold or shared for marketing.
A fictional service-business example. It illustrates how the decision is reached, not a client result.
A service-business owner asks for a new website because the team keeps chasing customers for missing job details.
The objective is a usable handover to the team. A website rebuild is one possible response, not an agreed diagnosis.
Compare the customer’s request, the office record and the handover. Ask what was missing, who added it and whether existing software could carry it through.
The evidence might support a clearer service page, a revised intake step or a connection between existing tools. Scope only the change the diagnosis supports.
A first draft says the missing details are costing jobs. Nothing yet shows that, so the claim comes out: record the observed gap, mark the impact as uncertain and agree how to measure it.
Ben reviews the findings with the owner. The owner chooses whether to commission a Build, with its price and acceptance checks agreed in advance.
Findings are checked against their supporting evidence and reviewed by Ben. When available, a separate model adds another check. Agreement between models is not proof, and unconfirmed findings stay labeled as such.
You approve the scope, price and checks before a Build. Customer messages, spending and publication need approval before they happen.
We review decisions and available outcomes to improve the method. A lesson from one business is not automatically a rule for another; changes to lasting knowledge require review and approval.
One problem, examined from the relevant angles. Eren uses focused AI methods for websites, search, marketing, sales and product decisions. Their findings feed one recommendation that Ben reviews with you.
A free First Look establishes fit. The paid Diagnose investigates the problem; a Build implements only what you agree to.