Information Architecting The Coffee Shop
As a user experience architect I often get asked what it is exactly that I do. A learned friend once explained to me that if you can’t explain your job to your mum then there’s something wrong. I never understood whether this meant there was something wrong with my mum, my job per se or just the title but either way it’s time to address that very issue.
I think I feel that my job is more of a re-labelled Information Architecture (IA) role than anything else but that’s even more ambiguous. Plenty of people have tried to define IA, notably Rosenfeld & Morville (“Information Architecture for the World Wide Web”), Wikipedia and recently, Tim, Tom and Chris Farnum. Having recently discovered Indexed I’m in a bit of a ven diagram mood so I’ll borrow from that and the Design Critique podcast to visualise IA as follows (left). I think this encapsulates what we as IAs try and do, to balance the three (oft-competing) requirements of users, the business and the content we have to organise.
You’ll notice that this definition isn’t tied explicitly to the web. It’s a model that covers all kinds of information interaction and whilst I could sit here on a blustery January day and bang on about web-based IA I think it’s time to talk coffee. It’s quite a good analogy I think for real-world IA, where non-web people can get a feel for what practitioners like myself do.
I occassionaly find myself in need of a new place to go for lunch or for coffee. In Norwich this isn’t easy to do, it’s fair to say I’ve tried most of the sandwich bars and decent coffee shops and what strikes me in the same way it does in towns up and down the country, is the organisation and presentation of the menu board. In London these places tend to be quite busy at lunch with a healthy queue in most places giving you plenty of time to look up and read the menu while you wait. In Norwich there’s not so much time, you’re asked for your order within a minute of entering the shop and if you’re new to the place that’s not enough time. Part of the problem is the bewildering array of choice. Lame stand-ups (and my Dad) regularly joke about the ‘I just want a coffee’ situation when walking into a Starbucks but, coffee aside, sandwich shops like O’Briens have a range of sandwiches, toasties, panninis, ciabattas countless hot and cold fillings, salads and soups. I like variety in my life (!) so I want to try something different and will scan the menu boards looking for combinations, unusual fillings, relishes and breads. This takes time and it’s not helped by poor information architecture.
If I was re-designing a menu board to improve customer journeys in a sandwich shop or coffee bar this is what I’d do:
:: Prioritise. Order the most commonly requested drinks/sandwiches in a section of perhaps 10 items on the far left of the board. Most people scan left-to right and would hit this section first. I would use till data to define the top requested items and order them accordingly. The only risk here is that this is self perpetuating, new people come in an order the same things everyone else does but the trade off is that they’re likely to come back and their natural boredom threshold will encourage them to explore other options later, particularly if the alternatives are promoted successfully…
:: Structure. Group similar items, all the hot food items in one group, cold in another. Then sub-divide, meats, fish, vegetarian. In each subcategory, order these by price, most expensive to least. If there’s a consistent pricing structure, promote it: e.g. bacon and cheese sandwich £2.50, toasted £2.90, think customer journey. People come in and decide first what type they want: sandwich (brown, white, granary), baguette, panini, soup or ciabatta (olive, sundried tomato) first. Then how they want it: hot, toasted, plain then filling. The menu should reflect that from left to right.
:: Display. Use a clear and consistent typeface. Have bold headings, use colour sparingly and intelligently (‘Hot Food’ in red for example), don’t add illustrations. Most people know what a cup of coffee looks like, or what a baguette is. Position the board high and use as much space for it as possible. If your bar has two entrances or queues, have two boards. Consider having a second smaller version close to the till. All too often brand typefaces are used that are fine for the shop logo, napkins or coffee cups but don’t read well from 5m away and 3m below. Use space, less is more. Listing every possible combination is pointless, especially if all your fillings are displayed individually, make it clear they can build their own and promote a few suggestions.
Change it. Be prepared to change the board regularly. When you stop selling lines or start selling new ones, adding them with stickers, hand written notes or another board breaks the structure. It’s like a library getting a new book and sticking it on the end of the shelf. It might get found but it’s more likely to be if it’s categorised and filed appropriately. Be prepared to listen to customers and analyse their habits. Are people saying and buying the same things, can they find your new caramel frappuccino or is it that they just don’t want one? Do you only seem to sell salads to vegetarians, do people know they can have their soup in an oversized takeaway cup?
Ok, so I don’t work in the restaurant trade but to me, that’s a good thing. As a customer I know about the problems I have and as an IA I know of some of the ways these can be solved. I’d love to know if Starbucks uses IAs to define its menu boards before the designers are let loose with their crayons and I’d bet that two thirds of small-town sandwich bars write their menu boards by consulting a long list of what they plan to sell.
I think I feel that my job is more of a re-labelled Information Architecture (IA) role than anything else but that’s even more ambiguous. Plenty of people have tried to define IA, notably Rosenfeld & Morville (“Information Architecture for the World Wide Web”), Wikipedia and recently, Tim, Tom and Chris Farnum. Having recently discovered Indexed I’m in a bit of a ven diagram mood so I’ll borrow from that and the Design Critique podcast to visualise IA as follows (left). I think this encapsulates what we as IAs try and do, to balance the three (oft-competing) requirements of users, the business and the content we have to organise.
You’ll notice that this definition isn’t tied explicitly to the web. It’s a model that covers all kinds of information interaction and whilst I could sit here on a blustery January day and bang on about web-based IA I think it’s time to talk coffee. It’s quite a good analogy I think for real-world IA, where non-web people can get a feel for what practitioners like myself do.
I occassionaly find myself in need of a new place to go for lunch or for coffee. In Norwich this isn’t easy to do, it’s fair to say I’ve tried most of the sandwich bars and decent coffee shops and what strikes me in the same way it does in towns up and down the country, is the organisation and presentation of the menu board. In London these places tend to be quite busy at lunch with a healthy queue in most places giving you plenty of time to look up and read the menu while you wait. In Norwich there’s not so much time, you’re asked for your order within a minute of entering the shop and if you’re new to the place that’s not enough time. Part of the problem is the bewildering array of choice. Lame stand-ups (and my Dad) regularly joke about the ‘I just want a coffee’ situation when walking into a Starbucks but, coffee aside, sandwich shops like O’Briens have a range of sandwiches, toasties, panninis, ciabattas countless hot and cold fillings, salads and soups. I like variety in my life (!) so I want to try something different and will scan the menu boards looking for combinations, unusual fillings, relishes and breads. This takes time and it’s not helped by poor information architecture.
If I was re-designing a menu board to improve customer journeys in a sandwich shop or coffee bar this is what I’d do:
:: Prioritise. Order the most commonly requested drinks/sandwiches in a section of perhaps 10 items on the far left of the board. Most people scan left-to right and would hit this section first. I would use till data to define the top requested items and order them accordingly. The only risk here is that this is self perpetuating, new people come in an order the same things everyone else does but the trade off is that they’re likely to come back and their natural boredom threshold will encourage them to explore other options later, particularly if the alternatives are promoted successfully…
:: Structure. Group similar items, all the hot food items in one group, cold in another. Then sub-divide, meats, fish, vegetarian. In each subcategory, order these by price, most expensive to least. If there’s a consistent pricing structure, promote it: e.g. bacon and cheese sandwich £2.50, toasted £2.90, think customer journey. People come in and decide first what type they want: sandwich (brown, white, granary), baguette, panini, soup or ciabatta (olive, sundried tomato) first. Then how they want it: hot, toasted, plain then filling. The menu should reflect that from left to right.
:: Display. Use a clear and consistent typeface. Have bold headings, use colour sparingly and intelligently (‘Hot Food’ in red for example), don’t add illustrations. Most people know what a cup of coffee looks like, or what a baguette is. Position the board high and use as much space for it as possible. If your bar has two entrances or queues, have two boards. Consider having a second smaller version close to the till. All too often brand typefaces are used that are fine for the shop logo, napkins or coffee cups but don’t read well from 5m away and 3m below. Use space, less is more. Listing every possible combination is pointless, especially if all your fillings are displayed individually, make it clear they can build their own and promote a few suggestions.
Change it. Be prepared to change the board regularly. When you stop selling lines or start selling new ones, adding them with stickers, hand written notes or another board breaks the structure. It’s like a library getting a new book and sticking it on the end of the shelf. It might get found but it’s more likely to be if it’s categorised and filed appropriately. Be prepared to listen to customers and analyse their habits. Are people saying and buying the same things, can they find your new caramel frappuccino or is it that they just don’t want one? Do you only seem to sell salads to vegetarians, do people know they can have their soup in an oversized takeaway cup?
Ok, so I don’t work in the restaurant trade but to me, that’s a good thing. As a customer I know about the problems I have and as an IA I know of some of the ways these can be solved. I’d love to know if Starbucks uses IAs to define its menu boards before the designers are let loose with their crayons and I’d bet that two thirds of small-town sandwich bars write their menu boards by consulting a long list of what they plan to sell.
Take a look at my very quick sketch of how a well organised menu could be set out. I hope this helps to bring IA alive for some people, feel free to feedback either by commenting on this post or emailing me … john at smorgasbord hyphen design dot co dot uk ... P.S. No more posts about coffee, promise.
UPDATE 3 Feb 07: The guys at Adaptive Path have posted about trying to explain what user experience architects do using a few options. Worth a read...
1 comment:
nice post
Post a Comment