Does this sound familiar?
There are just two simple things that separate UX design from graphic, web and other sorts of design.
Find out what they are now...Read More
Derek Halpern posted a video this week: "Why I hate split testing"
What a headline! He certainly got me watching. I was shaking my virtual fists at him: "what are you TALKING about?!"
But what is he really saying?Read More
I wrote a review of Visual Website Optimiser for TrustRadius. Find out who I recommend it for, what it does well and what it could do better right here:
Last month, one of my split tests resulted in about a £500,000 saving. If I were getting a modest 5% revenue share, that would mean £25,000 for one month’s work.
Sounds pretty good, right? And there are even bigger results coming soon.
But I'll never do revenue sharing. Here's why:Read More
By studying the opposite of what you want, you can identify important elements that aren't immediately obvious.
To work out what a terrible user experience for a web-based social network would be like, I brainstormed a load of nasty things that could happen. They fell into a few major categories:Read More
Changing your mindset changes what you notice and opens your eyes to pick different signals out of the noise.
Open your mind to all the lenses through which you can view your websites and you will find new and surprising opportunities to improve them.Read More
The fact is that many websites and web businesses cause this feeling in their users.
And I can tell you exactly why and exactly how to fix it.Read More
I was talking with a fellow designer this morning. He's just discovered the wonderful world of Steve Krug and "Don't Make Me Think" and mentioned breadcrumbs.
"Steve Krug said about.com do it best but I can't actually see them anywhere on the site."
He was absolutely right. They're gone. Things have changed since 2005.
I think breadcrumbs have lost most of their power since the Internet moved on to new paradigms of search and specificity.
And most sites that had them are now so enormous and multi-faceted that they're just impossible to implement. As soon as you're basing your architecture on tags or semantics instead of a fixed tree structure, breadcrumbs become crumble. When you can access any single piece of information through multiple routes, you can no longer assign it a single fixed home in a tree structure.
What do you think? Are breadcrumbs dead? And what are we using to orient our users instead?
We moved to run away from the crap. But we didn't need to move to anywhere in particular. And that made deciding where to go surprisingly difficult. Merely running away from the things we hate can lead in any direction. Anywhere except here really doesn't narrow it down.
But moving towards our values gives us a clear and definite heading to follow. It's that way, cap'n. 5 degrees either side won't do.
It was too hard getting in and out with the kids. When we did get out, there was nothing we wanted to visit within a 20 minute walk. The garden was too hard to get to (and too full of fox poo.) Moan, whinge!
Top priorities for the new place are often the things you hated most about the old place, be it layout, location, or water pressure for a decent shower.
The risk is that you throw the baby out with the bathwater and forget to look for the things you found loveliest about the old place. Too late you realise that your new place lacks character, the new bath is uncomfortably small (though at least you can have a decent shower) and then you discover that the electrical wiring was installed by Mr Bean. Moan, whinge!
A great home inspires the behaviours you value. And unless you've got megabucks, you can't afford to tick every single box, so you have to prioritise the behaviours you value most.
- Love being right in the middle of the action? Get a little bolt-hole in the centre of the city. Just accept that you won't have much space and forget about a garden.
- Determined to involve the whole family in preparing and eating meals together? Go for a big open plan kitchen/diner with room for all the clan and more besides. Just accept that you'll compromise on location.
A great website inspires the behaviours you value. And unless you've got megabucks, you can't afford to tick every single box, so you have to prioritise the behaviours you value most.
Ever started a website redesign project like this: "We hate the blue in the colour scheme. We're bored of these images. We don't like having the menu down the left. It's too hard to find our white papers?"
Don't worry: these feelings are valid. Your website is a very important and visible element of your brand. It's a sorry state of affairs if you feel so embarrassed by it that you apologise every time you share the link. A shonky website is a warning flag to all your potential customers, like showing up to a business meeting wearing a torn and dirty t-shirt with the word, "MUPPET" on.
But unlike your home, your website isn't actually for you. Your prospects and customers are the ones that live there. Your site must inspire them to behave in ways that create value – both for them and for you.
Know what your visitors are doing
When an architect is commissioned to redesign a public space, the first thing they do is sit in the space and observe current behaviour. They (or an intern) will sit with a clicker, pad and pencil and trace out how people move around the building.
You need to understand how people use your site as deeply as you understand how you use your own home. Only then can you compare current behaviours with desired behaviours.
How much easier it is for us web architects, with tools like Google Analytics, KISSMetrics etc. to watch people for us. The problem is not too little data – it's too much. Instead of getting bogged down in the morass, try being an architect and focus on drawing out your visitors' routes:
- The main thoroughfares through the website
- The most heavily used and lucrative routes (NB. do valuable customers take different paths from visitors that don't buy?)
- Areas that appear to be inaccessible to traffic
- Areas that really interest people – and areas that really don't
When I do this with my clients it's always a real eye opener.
Find out why
Architects typically interview a sample of visitors, discovering their reasons for being there, how often they use it and how they feel about the space. The architect wants to design the space to enable valued behaviours – layout and style will follow purpose.
Interview your customers and visitors. Don't be fancy or formal: just call them up for a chat. And sit with people while they use your site for some guerrilla user testing.
Do ask thoughtful questions. People are terrible at predicting their future behaviour, so avoid asking "would you buy X?" or "would you like to see Y?" Try instead asking things like:
- What was your main reason for visiting the site today? (and were you successful?)
- What were the three biggest reasons that you chose to buy our ?
- Can you tell me about a time when you
- What was the main reason that you didn't take us up on today?
- If we could wave a magic wand and solve your biggest challenge for you, what would you have us solve?
This research will give you a picture of what your users are running away from – but crucially, you'll also have a bead on the values they're moving towards.
Redesign to inspire your visitors to valued behaviours
Let your research surprise you and keep an open mind! You'll probably find that your users value or hate different things from you. Remember that the website is for your customers, so their opinions must supersede yours. It really doesn't matter whether your boss likes the colour scheme if she's not part of the website's audience. If you and your bosses/shareholders focus on making the space right for your visitors, you'll reap the rewards.
The best bit is that you'll probably find that you don't actually need to redesign your whole site. Just knock down a wall here, add an extension there, have an interior designer come up with a more inviting colour scheme and, of course, install a power shower…
So make the alterations. Then keep watching your visitors to make sure it's worked.