Wrinkly ladies nationwide have coordinated a stampede for Boots No 7 Protect & Perfect face cream, after it was endorsed by the BBC's Horizon: so let's do face creams. Basically they're all the same. They all moisturise, like vaseline, but without the greasiness. And most also contain mashed up protein chains: these are long and mobile when they're soggy, but curl up and contract when they dry, for instant tightening gratification. Companies give these proteins French sciencey names such as Tenseur Peptidique Vegetal.

Lastly, they may contain active ingredients: a double-edged sword. So vitamin A seems to be helpful on skin, but its active form, tretinoin, also causes burning, flaking and redness, so it's prescription only. You can give weaker forms, with names such as "pro-retinol", but they're converted to tretinoin slowly, so they don't work well: you could give lots but if it works, it might have tretinoin side effects too.

Now some have criticised research on the cream purely because Boots funded it. But what did the researchers do? They took nine volunteers with aged skin, and put some cream (moisturiser, weak cream, strong cream, and tretinoin) on to a row of 6mm patches on their arms, regularly, for 12 days. They put tretinoin on only for the last few days, because of side effects, but they used it because we know it works, so that helps you see if your outcome measurements are picking up true improvements.

Then they took biopsies, sliced them up, and compared them under a microscope, looking at the "extracellular matrix", the supporting architecture around the skin cells, and particularly at fibrillin microfibriles. These are damaged by exposure to sunlight, and tretinoin partially repairs that network. They assessed how good the fibrillin looked, scoring from zero to four. Nobody involved knew which cream was which, until the end, and the paper will be published shortly. The results? The No 7 Protect & Perfect cream scored better than basic moisturiser, but not as good as tretinoin (and both caused some reddening). You may now stampede.

So what's bad about this study? Well, it measures a "surrogate outcome", a lab finding, that is one step removed from the real world, and one that may flatter the cream. And it compares No 7 Protect & Perfect against placebo (a simple moisturiser) instead of other competing creams on the market.

But this is the astonishing thing. Other manufacturers don't submit their creams to a university dermatology lab, or publish academic papers: so these aren't so much the best results, they're the only ones. Why? The tests cost just £15,000. Cellular Radiance Cream by La Prairie, for example, costs £340 a pot. Fifty pots would fund one study.

But those tests could bite any company right back: because if they show no effect, then your business is trash; but if they do show an effect, then busybodies wade in to regulate your pharmaceutically-active product. If it was my cosmetics company, I'd stick with the sciencey diagrams and hope for the best.

· Please send your bad science to bad.science@theguardian.com