You already won the hard part.

Let me be honest about what a website is for, because plenty of people will tell you otherwise. Nobody in Kenya buys a five-million-shilling car in a browser. Anyone selling you an online checkout is selling you a feature you'll pay for and nobody will use.

The website's job is narrower and far more valuable: have the first four conversations without a human, so your salesman only handles the fifth — the one that closes.

Every car on the site shows a monthly figure, not just a price. Autochek does this in Kenya and it is the single most effective thing on their listings, for a simple reason:

Your buyer does not shop by price. He shops by what he can carry each month. The calculator on the mockup's vehicle page is real — drag the deposit and the term and watch it move. It's a proper reducing-balance calculation, the way NCBA and Equity actually quote.

Two details that sound small and aren't. Every enquiry form is three fields — name, phone, when suits you. Industry testing consistently shows cutting a dealer form to three fields lifts lead volume by around a third, on the same traffic. And every vehicle has a WhatsApp button that opens a message with the car already written into it, so the buyer never has to explain which one he means and your salesman never has to ask.

This is what you asked for when you said the app should do more, and it's where the real value is. A website shows cars. This runs the dealership. It has two sides — what your customer signs into, and what Dennis and Edwin sign into — and they're the same system. Both are clickable in the mockup.

My Garage. Saved cars, compared side by side. A saved car is a named human with a phone number attached to a specific vehicle — intent that today exists only inside somebody's DMs, where it can't be counted or followed up.

Alerts. "You saved a Harrier. The price just dropped." "Three SUVs under five million landed this week." This is how you use thirty-three thousand people more than once. Instagram shows a post to a slice of them for two days; an alert list is yours, permanently, and reaches everyone who asked to hear from you.

Pre-qualification. Eight questions, documents uploaded once, an indicative ceiling before he's even chosen a car. This is where your deals die today — a buyer decides on something, then spends three weeks chasing payslips and a bank letter, and goes cold. Collect it early and calmly, and the paperwork stops being the reason he walks away.

Trade-in valuation. He photographs his old car and gets a firm number, valid seven days. Note what this really is: people who would never fill in a contact form will happily photograph their car to find out what it's worth — and that request tells you both what he owns and what he's ready to spend.

Import tracking — and I'd build this one first. A fresh import takes six to ten weeks. Right now every one of those weeks produces anxious phone calls a salesman has to answer. Instead: a page showing purchased → loaded → vessel → Mombasa → KEBS → duty → NTSA → handover, with photographs at each stage. It kills the "where is my car" call, and — this is the part I'd bet on — he screenshots it and sends it to his friends, because waiting becomes something to show off instead of something to worry about. That is free marketing generated by paperwork you already do.

After the sale. Service reminders, insurance renewal, a referral link that pays him. The cheapest sale a dealership ever makes is the second one to the same person, and right now nothing in the business remembers a customer once the car leaves the yard.

The inventory manager — the feature the whole project stands on. A salesman photographs a car in the yard on his phone, types the details, publishes. Live on the website in about three minutes. Sold? One tap, gone.

The lead inbox. Every enquiry from every channel — Instagram, TikTok, WhatsApp, the website, Google — in one list, assigned to a named person, with a stage. In the mockup, two leads are deliberately shown unassigned and ageing. That's the point of the screen: today those two are sitting unread in somebody's personal WhatsApp and nobody in the business knows they exist.

The finance pipeline. Which application is at which bank, at which stage, for how many days. You work with several banks and track all of it in somebody's head. A stalled application is a sale quietly expiring, and the only way to catch it is to be able to see it.

The import board. Dennis updates a consignment's stage once; the customer's tracking page updates itself. The customer-facing magic costs nothing extra because it's a by-product of work the yard already does.

The numbers. Which cars get views but no enquiries — those are priced wrong. Which get no views — those have bad photographs. How long each unit has been sitting, because stock ageing is the largest silent cost in any dealership. Which salesman actually closes.

An online checkout. Nobody buys a car this expensive in a browser here. Deposits online, yes, eventually. Full purchase, no.

An App Store app — not yet. The webapp works on every phone, needs no download, and updates the instant we change it. A native app is a second product with a second cost and a review queue. Revisit it when there's a reason, not because it sounds impressive.

An AI chatbot pretending to be a person. Your entire advantage is that a real person answers, and that the person knows cars. Automate the paperwork, not the relationship.

A blog nobody writes. You already produce more content daily than most agencies produce in a month. The site pulls your Instagram in — it doesn't demand new work from you.

What you've been clicking through isn't a picture of a website. It is the website, running. The filters filter. The calculator calculates properly. The webapp switches between your customer's view and your salesman's. Nothing here is a placeholder except the cars themselves and the finance figures — and those are placeholders on purpose, waiting for your real ones.

We don't start again. We finish. Your stock goes in, your terms go in, your domain points at it, and it's live.

Take your time with this. Send it to one or two people whose judgement you trust. When you're ready, the list above is where we start, and we can talk about what each phase costs when we speak.