This note explains one tier of our work in enough detail that you can decide against it. It is written the way I would want a service explained to me: what it does, what it cannot do, where its numbers come from, and when not to buy it.
What it is. A scan of a business-for-sale listing, run by our own algorithm. You send up to three links; each one comes back as its own report, in minutes rather than days. The three reports are never compared or ranked against each other — that choice belongs to you and to nobody else.
The part that prints every time
Here is the design decision that matters most, and it came from a measurement that went badly for us. We had assumed the core of the tier would be identifying the business behind an anonymous advertisement. Then we measured it. We took twenty listings, chosen so that the mix of professional sellers and private ones matched the mix in the real market rather than a mix that would flatter us, and tried to name the business behind each. It worked in three. Building a paid tier on something that happens one time in seven would mean two buyers paying the same €99 and receiving documents of very different weight.
So the core of the report is what can be produced for any listing we are able to open. Seven measures, and each one is printed together with the spread of results it was measured against, so you can see where this listing falls inside it:
- How long it has really been for sale. Not the date the portal shows. Half of the listings we counted had been up longer than 146 days; 28 % longer than a year; the oldest we found had been up for 3,918 days.
- Whether "refreshed today" is the real date. 96 % of that section had been refreshed today or yesterday. That is why almost everything looks new, and why a fourteen-month-old listing can appear on your screen as three hours old.
- Where the asking price sits among comparable listings in its sector, as a position rather than a verdict.
- Price per square metre, on the same basis.
- The transfer price expressed in years of rent. The tightest measure we found, because rent is the price of the location: the middle half of the market pays between 3 and 8 years, with 5 in the middle.
- Who is really selling. 81 % of that section came from professional sellers; ten sellers held 43 % of it; the largest single seller ran 105 listings alone.
- What the listing does not say — each silence printed next to the share of listings that break it, so you can tell a normal omission from an unusual one.
That last one is the least obvious and, I think, the most useful. A missing figure means nothing until you know how often other sellers state it. Revenue appears in 10 % of listings, profit in 1 %, and the number of employees in none of the 140 we read line by line. So silence about revenue is simply how this market behaves, not a signal about your particular seller — while a missing rent, which 41 % do state, is genuinely unusual and worth a question.
One hole in the first measure, and you would find it eventually anyway. We read how long the advertisement has been up. If a seller deletes his advertisement and posts it again as a fresh one, that clock restarts at zero, and we will report the listing as new — because as far as the portal is concerned, it is new. Nothing in public material tells us this is the same business coming round for the third time. Where the business can be identified we sometimes catch it; where it cannot, we do not. That is a limit of the measure, not a detail, and it is why the report prints the market's own numbers next to the age instead of leaving the age to speak alone.
The part that prints sometimes
Where public material allows the business behind the advertisement to be identified, the checks that depend on knowing its name run as well. That is the one-in-seven part, and we report it as a measurement rather than sell it as a feature: three identified out of twenty, and two of those three confirmed a second time from a different source.
The reason it fails so often is the market, not the method. Eight of those twenty carried no shopfront in any photograph — agency logos, interiors, and images the portal itself captioned as generated by AI. When the business cannot be identified, the report says so and names precisely which checks that closes.
What ran, and what could not
The report prints which checks ran and what each returned — and then, in its own block, every check that could not run together with the reason.
Four checks are usually in that second block: the company register, insolvency, court records, and the trail a business leaves in public view — its own website, the map listing, the reviews. All four need a named company and a person to weigh up what comes back. That is the €550 Screen, not this one. A scan that quietly left them out would read as though everything had been examined. Naming them costs us the appearance of thoroughness and buys you an accurate picture of what you are actually holding.
Where the numbers come from
On 2 August 2026 we counted 344 live transfer listings in Spain, read 140 of them line by line, and traced the sellers behind 511. Every percentage in a report is printed with the number of listings it was calculated from. A percentage with no sample size attached is somebody's opinion in disguise, and you have no way of telling whether it came from a thousand listings or from four.
The base is narrow on purpose, and I would rather say so than have you assume otherwise: one portal, one country, one day, counted by us rather than quoted from an industry survey. A Portuguese listing is measured against that same Spanish base, and the report states this on the page instead of letting the numbers pass as local.
“This measures the advertisement, not the business”
That is the strongest objection to this tier, and it is correct. Everything above is measured from what the seller chose to publish, and from how his advertisement sits among several hundred others. Not one claim inside the listing is verified. If the advertisement says the bar takes €180,000 a year, we cannot tell you whether it does. We can tell you that only one listing in ten states a revenue figure at all, and that this particular one has been on sale for eleven months while stating it.
So what is the €99 buying? It answers a narrower question than is this business any good — it answers the question that comes first: is this one worth my next week? A listing that has been up for fourteen months, priced at nine years of its own rent, from a seller running a hundred other adverts, with no rent stated in a market where four sellers in ten do state it — none of that proves anything is wrong. It tells you where to point the first hour, and what to ask before you spend a day driving out to see it.
And a well-written dishonest advertisement can look perfectly ordinary on all seven measures. I would rather write that here than have you find it out afterwards. These measures catch the common failure — a business quietly unsellable for two years at a price nobody in its sector is paying. They do not catch a careful liar. Catching one of those takes documents, and documents are the tiers above this one.
The average listing runs to 798 characters and states two numbers out of the seven that matter. The value here is not in squeezing that text harder. It is in surrounding it with what is missing.
What you send, and what happens to your money
Three links. No address, no tax number, no documents. We open the pages before you pay anything, and payment opens only for the listings we could actually open. If a platform refuses to serve its page to an automated reader, we tell you which link it was and ask you to paste the text — and that link does not count against your three. If a link turns out not to be a business for sale at all, the money comes back in full. The €99 credits toward a Screen if you upgrade within 30 days.
Three situations where you should not buy this scan
- The seller has already given you documents. If you are holding accounts, the lease and the licence, your money buys more by having those read than by having the advertisement measured. That is the Screen.
- You have already decided to buy. At that point you need a contract and a lawyer, not an analysis. We are a filter before that step, not a substitute for it.
- What you actually want is a registry and insolvency search on a named company. This tier will list those checks as not run. Skip it and start where they are performed.
What the scan does not pretend to be. One automated pass over public material. It computes no money for any flag — statutes are cited, sums are not modelled. It does not read private seller documents, and does not analyse them even if they are attached. It is built to tell you whether a listing is worth the next step, and to be explicit about everything it left untouched.
Where this fits. Quick Scan (€99) reads the advertisement and places it against the market it came from — up to three listings, one report each, in minutes. Deal Screen (€550) is where a person reads the public record around a named business, weighs it and signs the memo.
Check my deal →Figures in this note come from our own count of the transfer section of one Spanish marketplace on 2 August 2026: 344 listings measured for age and price, 140 read line by line for disclosed fields, 511 traced to their sellers, and 20 taken in market proportions for the identification test. Sample sizes are stated with every share. The base is one portal, one country and one day, and reports say so where the listing comes from elsewhere. General information and commercial opinion, not legal, tax or financial advice.