NordWest Taxi — the entire digital presence for a brand-new taxi company

The problem: a taxi company nobody is searching for yet

A brand-new taxi company does not have a brand problem. It has a being-found problem.

Nobody types your company name into the search box the first time. They type "taxi Halmstad", "airport taxi Landvetter", "taxi to the train" — or they tap the first number that appears. The competitors in town have been there for years and have hundreds of Google reviews. That kind of head start is not something you close on reviews any time soon.

What you can do is be present for everything the older sites never bothered to answer. What does a trip to Copenhagen airport cost? Will the golf bag fit? Do you run wheelchair taxis? Can a company be invoiced? Those are real questions people type every day, and they lead straight to a booking — not to a brand somebody is meant to remember next time.

The brief was the entire digital presence: name and visual identity, website, content, booking flow and local visibility. The company had a phone number, a licence and cars on the road. We built the rest.

A note on geography: the client is in Halmstad, on Sweden's west coast, and most of the site is in Swedish. The work is included here because the problem is not Swedish at all — it is what every local service business faces at launch.

Four things we said no to

Scope is always the first real decision. Here it came to four.

• No link to the car and no online payment. We built no dispatch system and no meter integration. A booking on the site is a request that reaches the company, which then confirms it. That removes the hardware dependency entirely, and it means the site can never promise a car it does not know exists.
• No price list. The site publishes no fixed prices, and it writes out why: an honest price depends on the trip — distance, time and vehicle — so instead of a list full of caveats, the customer gets a clear answer before the trip, every time. It costs a search. It is worth it.
• No visitor tracking. No analytics scripts, no pixels, no tracking cookies. The privacy policy says outright that visits are not followed, and the Google map on the contact page only loads when you click it.
• No borrowed stars. The reviews page could have sat empty, or been filled with kindly-worded quotes. Instead it says: honestly, we are brand new, so this is empty for now. No invented reviews, no borrowed stars.

How the site is built

The site is 47 pages — 41 in Swedish and 6 in English. It was built from scratch rather than dropped into a template, and the structure is one move used three times: a hub page that distributes to pages which each answer a search of their own.

• Services that branch into airports. Seven services: airport taxi, courier transport, medical transport, wheelchair taxi, large-vehicle and minibus, corporate accounts, and taxi to the train. Airport taxi branches again into six pages of its own — Landvetter, Copenhagen, Ängelholm–Helsingborg, Halmstad City, Malmö and Växjö — each with the distance and journey time written out.
• Areas. Eight pages for the towns and districts around Halmstad. They are written about the places, not about the template — the Tylösand page is about the beach, the hotel, the golf club and summer traffic; the Getinge page is about sitting on the motorway between two towns.
• Guides. Seven longer pieces answering what people actually wonder before booking: how to book, what a comparison price means, taxis with children and car seats, getting to the region's big retail destination.

The English is a deliberate selection of six pages — home, booking, airport taxi, about, contact and the name change — not a mirror of the whole site. Those are the pages a visitor from abroad actually needs, and doing six properly is more honest than doing all forty-seven halfway.

The booking: a request that reaches a human

The booking form on the site is Manecab — our own product, in production at a paying client.

It is worth saying plainly what that means. A taxi company this size normally gets to choose between a phone number and an enterprise project. Here a real booking flow was in place from day one, because we had already built it for ourselves. The Online booking module in the Manecab case study is the one sitting on the booking page.

The flow is deliberately short: name, phone, pickup, destination, date, time and number of passengers. Email is optional — fill it in and you get a confirmation and a link to your booking. A message field takes whatever does not fit a field: child seat, luggage, wheelchair. The button says "Send booking request", not "Book", because that is what happens. Bookings are accepted up to 90 days ahead.

The form exists in Swedish and English and follows the language of the page it sits on. Spam is stopped in two stages without a real customer noticing anything.

The phone is still the fastest route, and the site says so itself rather than hiding it. "Call now" and "Book a taxi" sit side by side in the header, in every section, and in a fixed bar at the bottom on mobile.

Being visible when somebody needs a car

This part is the actual business. A taxi ride is booked in the moment, by somebody standing somewhere with a phone in their hand, and whoever is not visible right then does not exist.

The work sits on three levels:

• Every page has a search job of its own. Page titles are written as searches, not as headlines — "Taxi Halmstad–Copenhagen Airport", "Wheelchair and accessible taxi in Halmstad", "Taxi Tylösand — the beach, the hotel and the golf". No title duplicates another.
• Structured data across the whole site. The company is described to search engines as a taxi service, with address, around-the-clock opening hours and the area it actually covers — sixteen towns and six airports. Every service carries its own description, every guide its own, and the booking page is marked up as exactly that: a booking function. Thirty of the pages carry FAQs tagged so the answers can appear directly in search results.
• Content that answers before anyone has to call. The questions on the pages are real ones: what a taxi from Tylösand to the centre costs, whether paying by card costs extra, what a comparison price means. The guide on choosing a taxi ranks for a question that invites boasting — and declines it: nobody gets to appoint themselves best.

The Google Business Profile is set up and awaiting verification. Until that lands, the site and its structured data are doing all of the local work.

The hard parts

The page count says what exists. What made it a real build sits underneath.

• A rebrand mid-launch. The company previously marketed itself under a different name and became NordWest Taxi on 4 September 2026. The old address forwards to the new one page by page, so no inbound link and no indexed page is lost. The site also carries its own page about the change — same company, same number, same drivers — with a blunt warning that the company never asks for new payment details by email. A rebrand is an opportunity for fraudsters, and it is cheaper to say so in advance than to clean up afterwards.
• Language markup only where it is true. Pages tell search engines they exist in another language — but only the six that actually have an English counterpart. The other thirty-five say nothing, because they have nothing to point at. This is the single most common fault in bilingual sites, and an unpleasant one to discover after the fact.
• Zero third-party scripts, genuinely. There is not one external script on any of the 47 pages. The map loads only on click. The fonts are self-hosted and measured against their fallbacks, so text does not jump as they load.
• The booking frame measures itself. The form sits in its own frame and has to grow and shrink without an inner scrollbar — including shrinking back down after a booking is sent. The height travels over a small dedicated channel between the frame and the page, with the sender verified at both ends.
• Motion that ends. The headline on the home page types out the place names one at a time. It stops on the last one rather than looping, and stands completely still for anyone who has turned motion off in their operating system. An animation that never ends is an accessibility fault, not a detail.

What a client engagement shows that our own product cannot

At this point in the Manecab case study there is an honest caveat: there, we were both the client and the supplier, which makes the product strong evidence of what we can build and weaker evidence of what it is like to be our customer.

This case is the inverse. Somebody else set the scope, the budget and the date. The rebrand arrived in the middle of it. The decisions above — no price list, no borrowed reviews, no tracking — are decisions we proposed and the client agreed to, not decisions we made for ourselves.

What you will not find here is figures: no booking volumes, no rankings, no traffic, nothing about the fleet or the staff. That is a deliberate choice, not a gap. The client's numbers are the client's, not our marketing copy. Everything else on this page you can check for yourself by opening the site.

Where the work stands today

The site is live and the engagement is ongoing: we run it and keep working on SEO and content. The domain and the accounts are in the client's name — if they change supplier, the site goes with them, intact.

Three things are started but not finished, and they belong here as upcoming work rather than as delivered:

• The Google Business Profile is set up and awaiting verification. It is the single biggest remaining lever — in map results, a verified profile with reviews outweighs everything on the site.
• Photography. The site has deliberate placeholders where the images will sit, including on the team. Pictures of real drivers and real cars do more for trust than another page of text.
• Analytics starts as a step of its own, so it becomes possible to see which pages lead to bookings without building a tracking mountain.

To see the site, it is at nordwesttaxi.se.