Lessons from Celebrity Chefs: heuristic inspection or user-centred design?

A plate of fish and chips dreamed up by Heston Blumenthal for Little ChefA couple of nights ago, I was watching Heston Blumenthal cooking his ‘Roman Feast’. Blumenthal is a proponent of ‘molecular gastronomy’. His Fat Duck restaurant has been described as the best restaurant in the world, and is famous for its bacon-and-egg ice cream and snail porridge. Roman Feast and others in the same series have been pure culinary fantasy. “Don’t try this at home,” he warns viewers before plunging into techniques that used anything from dry ice to the skills of a taxidermist.

What has that got to do with our field? Well, nothing really. But it got me thinking back to his previous programme, Big Chef Takes on Little Chef, and comparing that with another mega-celebrity’s programme: Gordon Ramsay’s Kitchen Nightmares.

The basic recipe

At first glance, the two shows followed much the same format. A restaurant is in trouble. The celeb chef arrives, finds out what the problems are (with much swearing, in the case of Ramsay), tells them what to do and departs. If the restaurant’s owner and staff follow the advice, all is well. If not, disaster – and plenty of screen footage of Ramsay telling us, the viewers, what idiots the restaurant people are for failing to follow his instructions.

In some ways, that’s not completely unlike what we do in usability: arrive, dive in, find problems, tell them what to do. If our client does what we say, the results are often pretty good. If not, that’s the client’s fault, right? Well no, not really. Real clients are more complex than that. They often have good reasons for not doing exactly what we tell them to do. Which led me to thinking about the differences between the two programmes and what we might learn from them.

The heuristic inspection – Ramsay style

I realised that Ramsay is really doing heuristic inspections. He has a list of specific things that a good restaurant should do, starting with basic hygiene (he’s very keen on not killing the diners). He looks at the quality of service, level of organisation in the kitchen, portion control and profitability. What he ignores – or at least, has ignored in the programmes I have seen recently – are the aspirations of the restaurateur and the style of the restaurant. You want to be a tapas restaurant in Sheffield? Hard luck! Ramsay’s heuristic says “You must cook simple food based on local ingredients according to my recipes”. Out go the tapas, in come big pies from local meat.

Now, the culinary pedants amongst us (including me) would argue that proper Spanish tapas would indeed be simple food based on local ingredients – and there’s no reason why you couldn’t do a Yorkshire equivalent, given the wide range of produce available there. No, Ramsay knows best, forget your tapas ambition and follow his recipes or be in for some pretty unpleasant television mauling.

The tapas restaurant in Sheffield did what they were told, re-launched themselves as Silversmiths, and are doing very nicely thank you. Others have not been so lucky: the Ramsay recipe is strict, and doesn’t always fit so well with the capabilities of the restaurant or its ambitions. It’s hard to find out the truth: a recent News of the World article claimed that most of the restaurants in a recent series have shut down, but (as the commentators on the article point out) it’s difficult to tell whether the problem rests with Ramsay’s advice, the restaurateurs, or somewhere else.

I think this is very much like a heuristic inspection. It would be hard to argue that a restaurant should ignore its hygiene or profitability. Similarly, many of our heuristics are just as basic. For example, one of Nielsen’s Heuristics is: “Help Users Recognise, Diagnose, and Recover from Errors”. That’s just hygiene. If we look at the opposite, which would be “don’t bother telling users about errors – let them struggle”, we can see how vital it is. But heuristic inspection is known to be a very unreliable technique – as I’ve pointed out previously. It takes little account of what the business wants to achieve, what changes it is capable of making, what the users really want, or actual user behaviour.

User-centred design – Blumenthal style

In Big Chef Takes on Little Chef, four programmes in January took us through an apparently similar process. Little Chef is an iconic chain of British roadside restaurants, perhaps similar to Denny’s in the USA. Strong on the all-day breakfast, weak on gastronomic excellence or innovation. Blumenthal was brought in to try to revitalise the menu of the ailing chain. The managing director kept asking him for ‘Blue sky thinking’.

At first, the series followed the typical Ramsay-style heuristic approach. We saw Blumenthal and his top chefs being appalled by the menu and food in a typical restaurant. We saw the shock as they realised that the kitchen lacked any saucepans. We saw anger as the managing director refused to release the profit figures.

We saw a new menu being produced, complete with fancy touches such as the oyster in the beef stew and an ‘orange aroma’ to be poured around chocolate ice-cream. So far, so Ramsay. But here’s where the user-centred bit came in. They organised a head-to-head of the trial menu in a Little Chef, with the ordinary menu alongside. The ordinary menu won, hands down. The diners didn’t want fancy, they wanted familiar and predictable.

Blumenthal watched the videos of the user reactions and totally changed his approach (now, doesn’t that sound familiar?). He started to think about the users, and not just the diners who already chose Little Chef. He included the restaurant staff: what could they cook, and what skills did they have?

He went looking for the target audience: people who do not currently go to Little Chef, but might do. We saw his team questioning people in a big railway station. And we saw multiple iterations as the team tried to develop a menu that would meet their own standards for taste, the need for profitability, the cooking skills (or lack of them) of the staff, and the desires of the restaurant diners.

There were some further battles along the way, including scenes where a Little Chef cook was specifically chosen to be trained on a new, more sensible proposed menu because of her lack of culinary skills – but then felt belittled for the very quality that had caused her selection. Overall, though, what we saw was user-centred design that responded to the realistic business ambitions of the client. Design that responded to a full range of users: staff, existing customers, and new customers. Isn’t that inspiring?

Improving your advice

Try to learn as much as you can about the business that you are advising, what drives it, and the changes that it is capable of making. Be user-centred, in the widest sense: the users who will use the product, the staff who will help them to do so, and the client who is commissioning all of it. Involve users as much as you possibly can. If you’re forced to do an expert review, at least try to do a ‘persona-led heuristic inspection’ to bring some users into it.

This article first appeared in Usability News, 6 April 2009

Picture of Little Chef – Heston Blumenthal Popham, by madrudge creative commons