Offer vs Trust vs Scenario: Three Variants of One Landing Page — a Car Rental Case

What the "three page variants" method looks like on a real business — a car rental company with 250+ vehicles. Audit score 96/100, three hypotheses, three complete 5-page sites, and a one-month traffic test plan.

The clearest way to explain a method is to run it on a real business. We took an operating car rental company — i-CarRent in Irkutsk, Russia, 250+ vehicles, 24/7 pickup — ran the homepage through the analyzer, and built three complete sites for three different hypotheses. Below is what came out and why. All three are live: case page with links to every variant.

Step 1. The audit: 96/100 — and what it means

The analyzer scored the page across 20 factors in five groups. Result: 96/100 — speed, mobile, SEO basics and trust all maxed out. Remarks: a 7-field booking form and an overlong title tag.

The key takeaway for an owner of a strong site: there's nothing left to fix. The classic advice — "speed up the site", "add reviews" — has already been done. The only way to grow is to test *a different presentation of the offer*. That's exactly what three variants are for.

Step 2. Three hypotheses

We don't argue about which is best. We formulate three different answers to "why does the visitor leave?" and let traffic decide.

Variant A — "Price and booking in 2 clicks"

Hypothesis: the visitor is comparing 3–4 rental companies in adjacent tabs and picks the one where the class price is visible instantly and booking takes a minute, not seven form fields.

On the page: headline "Cars for rent from 2,100 ₽/day — 24/7 pickup", a quick search widget (class + dates) right under the hero, four popular cars with prices, steps "book → confirm → keys", a two-field form.

Variant B — "Network #1: trust and fleet"

Hypothesis: in car rental the biggest fear is getting burned — hidden fees, a deposit that never comes back, a worn-out car. The winner is whoever removes the fear with scale and transparency.

On the page: hero over a row of cars, a numbers strip (250+ vehicles, 24/7, 5.0 rating from 27 reviews, 3 cities), reviews, a "Transparent terms" block (documents, deposit, insurance, refund), three vehicle classes.

Variant C — "Scenario: a trip to Lake Baikal"

Hypothesis: a significant share of demand in Irkutsk is tourists flying in for Baikal. A page built for *their scenario* — airport pickup, all-wheel drive, routes — converts that segment better than a general homepage.

On the page: dark theme, an SUV against the lake, headline "A crossover for your Baikal trip — from the airport to Olkhon Island", four routes with distances and drivetrain recommendations, only 4WD cars in the selection, "ready for the road" (out-of-city travel allowed in the contract, airport pickup, on-road support).

Step 3. Not landing pages — sites

Each variant is not one screen but five pages: home, fleet with filters (class, drivetrain, road type), a price table, terms with FAQ, contacts with a map. The data is shared and real — fleet models, prices from the rental's own site, phone, address, company details — only the presentation strategy changes. That keeps the test honest: we compare approaches, not "who gave more information".

Technically it's one generator and three themes: change a price or a phone number, and all 15 pages rebuild in a second.

Step 4. The one-month test plan

1. Variants live on separate URLs. The main site isn't touched — zero risk to SEO or current sales.
2. Ad traffic splits three ways. Search ads point to A, B and C. The "car rental for Baikal" campaign goes only to C — that's its native page.
3. A smart algorithm shifts impressions. The variant that brings more bookings gradually gets more traffic — leads aren't wasted on weak versions.
4. After a month — report and winner. The winner becomes the new baseline, and three new challengers are built from it.

What this case teaches

• A strong site is a reason to test, not to stop. i-CarRent scores 96/100, and that's precisely why experiments with presentation are the only path up.
• Three variants = three reasons to leave. Price, trust, scenario. One usually wins by a wide margin — and it's almost never the one the owner bet on.
• Variants must differ in strategy, not details. That's what makes the difference visible on hundreds of visits instead of tens of thousands.
• Real data, different presentation. Otherwise the test compares completeness of information, not strategies.

Want the same teardown for your page? Start with the free audit — in a minute you'll see the score and three variant concepts for your site.

FAQ