One business, two entrances
No client commissioned this one. Czech surveying runs on developer contracts, and when construction froze across Europe that flow stopped. So I set the brief myself: open a second audience without losing the first. Geo Vzor is the firm I invented to build it for, and the site is live.
What this was
and what I did on it
- Project
- Self-initiated — there is no client. Geo Vzor is an invented Prague surveying firm; vzor is Czech for “specimen”, and the name says so on purpose
- Audience
- Two at once: developers and infrastructure projects (B2B), and private owners dividing or registering a plot (B2C)
- My role
- Strategy, UX, interface, motion, front-end, copy (CZ), SEO, analytics, deploy — one person
- Scope
- An eleven-page site: two audience entrances, six SEO service pages, a price estimator, lead capture
- Built with
- Plain HTML, CSS and JavaScript — no builder, no framework, no build step
The trade speaks to one customer
and that customer stopped buying
Czech surveying firms have run for decades on large B2B work — developers and infrastructure. When rates and inflation froze construction across Europe, the usual flow of orders went with it.
A private owner dividing a plot needs exactly the same survey, pays 9–15k CZK for it, and rarely finds these firms: their sites are written for procurement departments and say nothing a homeowner could act on. That gap is the brief I set myself: open the door without making the B2B side look smaller.
Five calls that shaped it
and what each one cost
-
01
Two entrances instead of one homepage
The first screen forks: one route for developers and infrastructure, one for private owners. Each leads into its own language, its own examples and its own proof.
Trade-off. A fork on the first screen is a risk — it asks the visitor to choose before they know anything. It was worth it because the two audiences share nothing: not the vocabulary, not the objections, not the price bracket. -
02
A site of pages, not a landing page
Six separate service pages, each answering one commercial query in the words people actually search. That is the part that keeps earning after the ad budget stops.
Trade-off. Six pages take far longer to write and structure than one long scroller, and they only start paying back after Google indexes and ranks them. A landing page would have looked finished in a week and brought nothing from search. -
03
Motion used as evidence, not decoration
Surveying is a precision trade, so the animations do precision: instruments, blueprints and scan lines drawn stroke by stroke rather than faded in.
Trade-off. Hand-drawn motion is slow to build and easy to overdo. The rule was that every animation had to demonstrate something about the work — anything purely decorative was cut. -
04
An estimate before the phone call
A calculator turns “how much does it cost to divide a plot” into a number on the page, then into a request with the parameters already filled in.
Trade-off. It commits a firm to a public price range in a trade that prefers to quote privately. That transparency is exactly what the private client was missing, so the discomfort sits on the right side. -
05
One person across every stage
Strategy, interface, copy in Czech, front-end, SEO, analytics and deploy came from the same hands, so nothing was lost in a handover.
Trade-off. It makes the schedule serial rather than parallel — no two stages run at once. In exchange, what shipped is what was designed, down to the focus ring.
Built to be looked at
not just described
Numbers I can stand behind
Lighthouse 12, mobile profile, measured on 2 August 2026. There is no traffic to quote and there never will be — no business runs on this site. These are build numbers; whether the SEO pages would rank is untested, and I am not going to pretend otherwise.