Affiliate List or Real Research? Six Tells That Give It Away in Thirty Seconds 

You do not need to know the industry to spot a paid ranking. You need six habits. 

Financial comparison content ranges from genuinely researched to entirely commercial, and the two look almost identical at a glance. They stop looking identical once you know where to look. 

Tell one: is anyone missing? 

The strongest signal, and the least noticed. Affiliate sites can only list providers that run affiliate programmes. Several well-regarded UK platforms run none. If a comparison omits a provider you know exists and is popular, you have learned something about the comparison. 

Tell two: where did the fee figures come from? 

If every number in the piece appears on the provider’s pricing page, the research consisted of reading that page. Not wrong, but not testing either. 

There is a cheap way to run this one. Take a fee from the comparison and open the provider’s own page alongside it. If the wording matches, qualifiers included, you are reading a transcription. Transcription has its uses. Establishing what you will actually be charged is not among them. 

Tell three: is anything criticised? 

Every platform has at least one irritating flaw. A review that finds none has either not used the product or has decided not to mention it. 

Tell four: does it mention getting money out? 

Withdrawal timings, transfer-out charges and dormancy fees are the details you only discover by leaving. Somebody who has actually left will mention them, which is why they turn up in the stock platform tables at The Investors Centre, a UK research site that opens and funds live accounts with its own money to test trading platforms rather than compiling rankings from providers’ published fee schedules. In affiliate content they are almost entirely absent, because nobody involved has ever withdrawn anything. 

Tell five: how does the site say it makes money? 

Look for the disclosure. Most have one. The interesting part is whether it addresses ordering, or only the existence of a commercial relationship. 

Tell six: can you trace a statistic? 

Pick one number and follow it. If it resolves to a regulator, a filing or a named data set, good. If it resolves to another article, keep going – you will usually end at a press release, several rounded numbers later. 

Does any of this mean affiliate content is bad? 

No, and it is worth saying plainly because the alternative reads as point-scoring. Affiliate-funded comparison has genuine merits: it is comprehensive, it is free, it is updated frequently, and for a first pass at an unfamiliar market it is often the fastest way to understand what exists. 

The problem is not the funding model. It is reading an affiliate ranking as though it were a recommendation. Those are different documents that happen to look identical, and the difference matters at exactly one moment – the moment you decide where the money goes. 

Several do, and it is a reasonable arrangement provided the two are distinguishable. A site can fund its own testing and still carry affiliate links, as long as the testing determines the order and the links are simply how the reader gets there. The question to ask is whether removing every affiliate relationship would change the ranking. If the answer is no, the model is sound regardless of how the bills are paid. 

Which of the six is easiest to fake? 

Tell three, comfortably. A line of mild criticism per provider costs nothing to write and has become standard practice on commercial pages: the app is a little cluttered, the research section could be deeper. Complaints of that shape are decoration. The ones worth anything name a cost, a delay or a refusal, because those are the ones a provider might pick up the phone about. 

Tell one has the opposite failure. It is hard to fake and easy to misread. A platform can be missing because it pays nobody, because it has no UK entity, or because it closed to new clients last spring. Absence is a question rather than a verdict, and it deserves ten seconds of checking before you conclude anything from it. 

The order is the least informative part of the page 

Most comparison tables are sorted by something the reader never sees, and ‘our top pick’ describes a selection process rather than a product. The useful move is to ignore the order on a first pass and read the table sideways, column by column. That is how you notice the provider that wins on fees and loses on everything else, which the ranking has quietly averaged away. 

Then put the order back and ask what would have to be true for it to be right. If the answer is a weighting nobody has published, you are looking at an opinion presented as a result. That is allowed. It is just worth knowing before you act on it. 

How do the two types compare on the things that matter? 

Six comparisons, drawn from what each kind of page can and cannot contain. One warning before you use it: funded research reports only on platforms somebody paid to open, so the provider you are most curious about may never have been tested. A gap on an affiliate page means nobody is paying. A gap on a funded page means nobody has got round to it. 

What is the fastest version of this check? 

Search the page for the word ‘withdrawal’. If it does not appear, nobody writing it has taken money out of the platform they are recommending. That single search resolves most of it in about four seconds. 

 …