What stock should I buy? Reading your own numbers

Your sold history already knows what works on your forecourt. How to read it, why volume and speed beat gut feeling, and the trap of buying what you personally like.

Updated 1 August 2026 · 5 min read · Written for UK independent dealers

Most independent dealers buy on a mixture of instinct, habit and whatever looked cheap on the day. Instinct is not worthless — it is compressed experience — but it is also where every expensive habit hides, because instinct has no memory for the cars that did not work.

Your own sold record does. It is the most valuable dataset you own and almost nobody reads it.

Start with what has actually sold for you

Pull your last thirty or forty sales and put them in a table: make, model, price band, days to sell, and what you retained. Then group by make and model.

Three columns tell you nearly everything.

Units. How many of that model have you sold? Volume proves there is a repeatable market for it in your patch, with your customers, at your prices.

Average days to sell. How fast does your money come back?

Average retained profit. What did you actually keep, after VAT and costs — not the sticker margin.

Rank by units first, then by speed. Profit per unit is the last of the three to look at, not the first, for two reasons: it is the number most likely to be recorded sloppily, and a fat margin on a car you sell twice a year is worth less than a thin one on a car you sell monthly.

Buy more of what turns

The segments to buy more of are the ones where you have sold at least three in ninety days, they turned inside your target window, and you currently hold fewer than you sold.

That last condition is the one people miss. If you sold four Fiestas last quarter and you have four on the forecourt now, you are already stocked. If you sold four and have one, there is a gap, and gaps are where the easy money is.

Ease off what ages

The mirror image: segments where you have two or more units past your urgent threshold, and where the ones that did sell took longer than your target.

The instinct here is usually to blame the individual cars — this one had the wrong colour, that one was priced badly. Sometimes true. But when three cars from the same segment all age, the segment is telling you something about your customers.

What people get wrong

Buying what they like. The most common and the most expensive. A dealer who loves fast German saloons will buy fast German saloons, will pay a bit more for them because they can see why they are good, and will be slower to discount them because they know what they are worth. Every one of those instincts costs money. Your taste is not your market.

Chasing a headline margin. A car that looks like it will make £2,500 is attractive right up until it takes five months. Two £900 cars that turn in a month each, twice over, beat it comfortably and leave you with your capital back.

Buying because it was cheap. A car is not a bargain because it went under the guide. It is a bargain if it sells. A cheap car in a glutted segment is cheap for a reason that will still be true when you try to retail it.

Drifting out of your band. Every dealer has a price band their customers occupy. Buying well outside it — a £22,000 car when you normally sell £8,000 cars — means a different buyer, different expectations, different marketing, and usually a much longer wait. Occasionally worth it. Rarely as often as it happens.

Widening the mix too far. There is real value in stocking makes you understand. You know what goes wrong, what the parts cost, what the buyers ask. Twelve makes across thirty cars means you know none of them well.

The checks before any purchase

Whatever the source, four questions before you commit:

1. What does it sell for, and how fast? Sold prices for the exact specification at that mileage, and the median days-to-sell.

2. What is the days' supply? Active listings divided by daily sold rate. Over 150 days is a glut — walk away regardless of price.

3. Has anything like it sold near me? If nothing comparable has sold within 25 miles in 90 days, there is no local buyer, whatever the national figures say.

4. What is my ceiling? Retail, minus prep, minus warranty, minus your grossed-up target, minus fees and transport. Written down before you bid.

A car that fails any one of these is a car you can afford to let go. There is always another auction.

Building a stock profile

Over time, the record produces a picture of what your forecourt actually is: a price band, a mileage ceiling, an age range, a set of makes, and the body styles your customers buy.

That profile is worth writing down, because it is a filter. When something comes up that falls outside it, the question is not "is this a good car" but "is this a good car for me" — and the answer is usually no, even when the deal looks strong.

The profile should be a bit wider than your current stock. Too tight and you never evolve; too loose and it is not a filter at all. A reasonable rule is to allow roughly 30% either side of your normal band, and to treat anything beyond that as an exception requiring a proper reason.

Reviewing it

Run the exercise quarterly. Markets move, your customers change, segments that worked last year get crowded.

The specific things to look for each time:

• Which segments have moved from fast to slow?
• Where has supply increased?
• What did you buy that you would not buy again — and why did you buy it?

That last question is the valuable one. Almost every dealer has a recurring mistake, and it is usually invisible until you have written down four instances of it in a row.