The ultimate goal of web design is to change a certain behavior.
An important thing before we get started: What I’m talking about here is web design for commercial purposes, not for personal projects or pleasure. Otherwise the above quote won’t hold true. What I argue, however, is that when designing for the web for commercial purposes, like a project for a client or working on your own product/service, testing whether your design “works” needs to be an essential part of your to-do list.
The Definition of a Web Design That “Works”
In the general sense, when you say that something “works”, what you actually mean is that “thing” is helping you accomplish a particular goal. Examples:
- This dating technique “works” / This dating technique helped me accomplish the goal of getting a girl
- This study tip “works” / This study tip helped me accomplish the goal of passing my exam
And a special one for the ladies (not sure if they use this phrase often though):
- This type of makeup “works” / This type of makeup helped me accomplish the goal of getting more attention from men.
We use the “(something) works” phrase a lot in our real lives; yet, we don’t use it as often in the virtual world. So I’m assuming this title has given many people some puzzled looks and they’ve been wondering what it really means. When a business person says “this web design works”, what he really means is this:
“This web design helped me increase my leads/sales/brand awareness.”
In other words, this particular web design helped him get closer to accomplishing a business goal.
Why and How to Put Yourself in Other People’s (Business) Shoes
I have plenty of friends who are web designers and one thing I noticed about many of them is that they love what they do. They love to tinker with small details and get things just “right”. Even if that button is 1 pixel off, they’ll do everything to correct it.
One thing many (I’m not implying you’re one of them, just stating my experience) of them hate is business. “But I’m not a businessman” is one excuse designers often use. Well guess what: If you’re working on a project and getting a monetary compensation for it, yes, you ARE in the Web Design Business. And like with every business, the Basic Economics 101 principles apply:
- Price is dictated by supply/demand
- Buyers are trying to get the most for their money, while sellers are trying to do the least for the most money
- You need to satisfy your clients needs/wants in order to get repeated business
Think about the last part. How would you do that? How would you get more repeat customers? By making sure your web design project works and produces results.
Use Real, Objective Data to Prove Your Design Works
If you Are Redesigning a Website:
If you’re redesigning a website, it’s important to first take data from a sample of visitors who’ve seen the old design and then take the same type of data from visitors who were shown the new re-design. Your goal is to compare how users behave before and after the re-design.
- Write a set of questions related to the site (how visitors like the site, what do they think about the products and so on) and survey them using a tool like KISS Metrics or
- Ask 30+ user to fill out a survey for the old web design (while it’s still there)
- After the redesign is up and running, ask the same questions to further 30+ people
- Compare the answers
Make sure that the only variable that changes is the design itself. It’s not recommended to change the text of the site because a lot of results from A/B and multi-variable testing has shown that this can have a profound results without even changing the site design at all. So changing just the design and not the text is a good place to start.
- Familiarize yourself with the various usability tools which can help people interact better with your site. Like with the previous idea, run them on the old version of the site and then on the new version and compare the results.
Mashable has an amazing list of some of the best usability tools out there. I won’t go into many of them, but here are some which left a good impression:
- Usabilla.com – I’ve seen many tools similar to Usabilla.com but they all have crazy prices starting from $200. This tool, however, has an acceptable price and some pretty neat features that allow you to track your users behavior and ask for their opinion.
- FiveSecondTest.com – A great tool that you can use for free (you need to “do tests to earn tests” or pay a certain price for X responses/month). This is great for testing the first impression people have for your site. The way it works is that users view your site for 5 seconds and then they answer questions you’ve set for them.
Here is one good question to ask people on FiveSecondTest (or any similar usability first-impression-testing site):
What is this site about?
MarketingExperiments.com, which is probably the only company in the world that has conducted thousands of research articles on huge websites and published them for everyone to see. At a webinar the person who was in charge of the testing procedure/reviews said something I will never forget:
“Based on the thousands of tests we’ve done for our research partners, we’ve concluded that clarity trumps persuasion. If users arrived at your site and had no idea what it’s about, they’ll leave. Before pitching why customers should buy your product, answer their questions: “Where they are” and “What can they do here””.
- Google Page Speed – When redesigning, make sure your new design has better or at least the same load speed as the previous one. Why? Because page speed has been correlated with how many people will buy from your site, how long they stay and many other types of interactions.
If you are making a site from scratch:
The same principles apply here as to redesigning a site with one very important twist:
When you are designing a site from scratch, you don’t have an old web design to compare your new design to. Invent one. Make several variations of your new design and A/B split test them to see which one makes people interact more. In other words, if you have several drafts of the same design that look different, try them all and see what works. Make your hypothesis on what you think would be the best performer and put your assumptions to the test. You’ll often be very surprised.
How to Persuade Clients That Testing Works
One of the best ways to persuade people to do or use something is with anecdotes. Sites like this are full of examples where changes, even if they were really small, produced drastically different results. 300% more people filled a form, 500% increase in sales and so on from a really small change. There are some really impressive stories that will persuade clients that they should test.
One question that might pop immediately is:
Why bother persuading my clients to test at all?
Two words: REPEAT BUSINESS. If they see that with your web design they got 200% increase of sales, they’re more likely to hire you for their next project. It’s not just because they would be glad that you’ve done a great job, but also because they’d have more money to potentially invest into improving their designs.
What do you think?
I’d love to hear you opinion on the pros and cons of testing whether your designs “work” and how to improve them. Let me know what you think in the comments below!