Some years ago I had a restaurant that was the kind of establishment that made its way into food guides. For a few years we even enjoyed a rosette from the Michelin. But as time went by I began to wonder about those food guides. Who wrote them and why? What criteria did they use for ranking?

There are two distinct species of guide - ones that restaurants pay to get into and ones that they don't. Now you don't have to be genius to spot the flaw in the first kind. For some guides a restaurant doesn't even need to make good food, it just has to pay for its entry. Some of them even offer to let you write your own review, just as long as you pay. What use these guides can be to people seeking objective information is hard to fathom. Some can be remarkably persistent; I still receive an annual application form from the Belgian Cyclists' Guide looking for 200 quid so that they'll list me. Thankfully for the would-be diner this kind of guide is easy to spot and avoid; it's full of advertisements.

The other sort, the ones where you don't buy a listing, are obviously the most likely to supply some real information. But still the choice is huge: Egon Ronay, the Michelin, the Gault-Millau, the Good Food Guide and the AA guide to name a few. Cross-checking these guides makes an interesting pastime. What may rank as great value in one is unedifyingly plain in another. One guide may give its highest accolade to a restaurant that isn't even listed in another.

Even in the group of serious food-guides there is diversity in how they approach the inspection. Some, like the Michelin, have full-time salaried inspectors whose job is to eat lunch and dinner in restaurants. Others have no salaried inspectors, but simply rely on readers to write in with their comments, good or bad.

As a restaurateur I did wonder occasionally about the qualifications of the inspectors. In most fields of human endeavour we expect expertise when we turn to experts. We want our experts to know more about the subject than we do, otherwise we wouldn't have consulted them in the first place. The problem with anonymous informants is just that; we know nothing about them or their individual likes and dislikes. But what of the salaried inspectors upon whose opinion we might rely? The Michelin Guide will only employ inspectors who have spent at least three years working in hotels or restaurants, so they have a good basis on which to make their assessments - or at least so I've been told. You won't find this kind of information in the guides themselves, and more's the pity. It would go a long way toward helping the readers know precisely what kind of guide they're using.

At the launch a couple of years ago for the Routier guide to Britain and Ireland I talked to their spokesperson Judith Burges. Routier is a membership organisation, she told me, which means that only members are listed. This explains why Patrick Guilbaud's or Ballymaloe House weren't in it at all. If you don't buy the membership you don't get listed. When I asked her how they did their inspections she explained that the inspecting and the ranking was done by their sales staff - but not in their own area, to 'preserve objectivity'.

Only a fool would lose sight of the fact that the principal reason for publishing a guide is to make money for the publisher. Which brings me to the nub of the matter. When a guide defines itself to me as a marketing aid to a restaurant, then by defintion it has compromised itself as an independent purveyor of information to the potential customer. You can't expect a marketing person to tell you the unvarnished truth about their product. What it comes down to is this: does a guide make its money from sales to the public, or from advertisements and membership charges?

As the buyer of a guide, it would be nice to know precisely how the selection was made and by whom. Without that information the buyer has no means to assess what sort of listing he is being offered. With this information to hand you could make a assessment as to whether a guide was put together and marketed for the restaurant or for the consumer. Perhaps as there are now so many guides on offer, there is room for one more: a guide to food-guides. Maybe the time has come to rank the rankers.