Your website has one job. Most websites don't know what it is.

A page can be beautiful and still be broken.

Here is the uncomfortable part: you probably cannot tell which one yours is.

Unbounce analysed 41,000 landing pages carrying 464 million visitors between them. The median page converts at 6.6%. The top tenth clear roughly double that.

Which means if your page sits at 2%, you are not slightly behind. You are handing back two thirds of everyone who already found you, already clicked, and already decided you were worth a look. You paid for those people. They left anyway.

Nobody sells you that problem, because the person who designed your site was never measured on it. They were measured on whether you liked the mockup.

I am not a designer who learned to sell. I am a seller who can build.

That order matters more than it sounds.

Kamil Dabek · Barcelona

01

I close deals for a living, today

I am an Account Executive at MongoDB carrying roughly 100 accounts and $4.3M of ARR. I run discovery calls every week and hear the exact objections that kill deals. Your page gets built around those objections, because I have already heard them out loud.

02

I spent three years at AWS on monitoring

I was a software engineer on CloudWatch, which is the service the rest of Amazon uses to find out whether anything is working. I instrument what I build. You will know what your page does, not just how it looks.

03

I build it myself

No handoff between the person who wrote the copy, the person who drew the page, and the person who wired the form. One person holds the whole thing, which is why it ships in days rather than quarters.

Every page gets one question before anything is drawn.

What is the single thing a visitor must do here? Everything else on the page either serves that or gets cut.

What most briefs ask for

A homepage, an about page, a services page, a contact form. A sitemap, essentially. It describes the company. It is organised around your org chart rather than around the decision your visitor is trying to make.

What I ask for instead

Who is landing here, what are they afraid of, and what is the one click that makes you money. Then the page becomes a sequence: earn attention, name the problem, prove you have solved it before, remove the risk, ask. In that order, every time.

Three builds, and the conversion decision inside each.

These are working builds you can click. I am showing you the mechanic rather than a lift percentage, because I would rather you judge the thinking than take my word for a number.

Investor mortgage lending

Ledgerock Deal Desk

Brokers in this market lose deals because the borrower cannot tell whether they qualify until someone calls them back. That delay is where the lead goes cold.

See it live

The conversion decision: Put a live DSCR calculator on the page instead of a contact form. The visitor gets a real answer in fifteen seconds, and the act of entering their numbers is the qualification.
Why it works: Value first, ask second. By the time the form appears they have already invested effort, and effort already spent is the strongest predictor that someone finishes.

Commercial solar

Kestrel Energy Systems

High-ticket, long cycle, and a sales team whose time is the scarce resource. Volume of leads was never the problem. Wasted meetings were.

See it live

The conversion decision: A staged assessment rather than a single form. Each answer narrows the recommendation, so the visitor feels progressively better understood while the business gets a qualified brief before the first call.
Why it works: Multi-step forms carry roughly triple the completion of the single-page equivalent with the same fields. Breaking the ask up is free conversion.

Legal services

Ashley River Family Law

Someone searching for a family lawyer at 11pm is frightened, not shopping. Standard firm websites lead with partner bios and practice areas, which answers none of that.

See it live

The conversion decision: Lead with the fear and the next concrete step, not with credentials. Contact is available from every scroll position, because high-intent visitors convert on impulse and impulse does not scroll back up.
Why it works: Matching the emotional state of the search beats matching the brand guidelines. The page is built as a template, so it re-skins for any local high-intent practice.

Live in ten days.

Fixed scope, fixed timeline. You are never wondering what week three looks like, because there isn't one.

Days 1 to 2

Positioning and the core action

We agree the one thing a visitor must do, who they are, and what is currently stopping them. I write the message before anyone opens a design tool, because the message is the thing that converts.

Days 3 to 6

The page and the funnel

Built, not templated. A page that earns the click and a multi-step funnel behind it that qualifies the lead before it reaches you, so the meetings you take are with people worth meeting.

Days 7 to 8

Wiring and instrumentation

Leads into your CRM, bookings onto your calendar, analytics on every step of the funnel so you can see exactly where people leave. This is the part most freelancers skip and the part that tells you whether any of it worked.

Days 9 to 10

Launch and handover

Live, on your domain, with a walkthrough of how to change copy yourself. You own everything. There is no platform you are locked into and no retainer you have to keep paying to keep the lights on.

Next 30 days

Iteration on real traffic

A launch is a hypothesis. For the month after go-live I keep reading the funnel data and making changes, because the first version is never the best version and pretending otherwise is how agencies get away with it.

The things you are actually thinking.

You have a full-time job. Am I getting your leftovers?

Fair question, and yes I do, at MongoDB. I take a small number of builds at a time for exactly that reason, and the ten-day scope is fixed so you are never waiting on me indefinitely. If I cannot commit to your timeline I will tell you on the call rather than take the deposit and find out later.

Can you actually promise my conversion rate will go up?

No, and anyone who does is guessing. What I can promise is that you will be able to see your conversion rate, which most people cannot, and that the page will be built on mechanics with evidence behind them rather than on taste. Then we iterate on what the data says for thirty days.

I already have a website. Does this replace it?

Usually not. Most of the time the right move is one purpose-built page for the traffic you are actually buying, sitting alongside the site you have. Cheaper, faster, and it isolates the variable so you learn something.

What do you need from me?

About ninety minutes total. One call at the start, one round of feedback on the draft, and whatever you already have: existing copy, analytics access if it exists, and your calendar or CRM login so I can wire the plumbing. I write the copy.

Where do the numbers on this page come from?

The benchmark is Unbounce's Q4 2024 study of 41,000 landing pages. The multi-step form figures are Formstack's. The funnel mechanics come from Cialdini on commitment, Nunes and Drèze on endowed progress, and Rackham's SPIN research across 35,000 sales calls. Sources are listed at the foot of this page. My own figures are my real numbers at MongoDB and AWS, and they are not client results, because I am not going to dress up my day job as a case study.

Find out what your page is costing you.

Eight questions, about ninety seconds. At the end you get the arithmetic on the gap between your page and a median one, using your numbers. Then you can decide whether I am worth a quarter of an hour.

Start the teardown