The key to successful marketing is advertising a solution, and selling the tool that provides it. Put another way, by looking at the jobs people are trying to accomplish, marketers can better interact with their prospects. Now imagine we think about UX in the same way: the key to a successful product is creating a solution (a better way to make a hole), and not just a tool (the drill).
“People don’t want to buy a quarter-inch drill. They want a quarter-inch hole!” – Theodore Levitt, Harvard Business School.
What Does Marketing Have to Do With Anything?
Sales and marketing teams often spend more in-person time with customers than the engineers or product developers do. As a result, the best information about what the user needs may be coming from the people who know those users best: the marketers.
Clayton Christensen, a Harvard Business School professor and author of many books on business and marketing, has studied many successful businesses, and the solutions that made their marketing successful. He theorizes that Theodore Levitt was right: people want to buy a solution.
Fedex and their “jobs-to-be-done” philosophy
Imagine how hard it was starting FedEx. The US Postal Service was the main competitor, millions of dollars of startup capital were required and, as with any new business, there was a pretty big risk that it would fail. Imagine yourself in this situation, starting a new company like FedEx with these huge risks. How would you discover what your customers really wanted?
Start by looking at things from the ‘jobs-to-be-done’ perspective. What are people who use postal service trying to get done?
One thing only: Get their package from one place to another as quickly as possible. The traditional mail service in 1973 was highly unreliable and many packages were lost or damaged in transit. The founder of FedEx provided customers with a solution: FedEx focused on getting a package from one place to another in days, instead of weeks or months.
This illustrates an essential point. What customers do is more important than who they are. Regardless of age, social status, or gender, when you buy a drill, what you want is a hole, and marketers know this.
What does this mean for UX?
Selling a solution works well in marketing, but user experience designers aren’t marketers or salespeople. We don’t want to sell a solution – we want to build solutions.
Eric Schmidt, the CEO of Google describes Google as more than just a tool. In his 2011 speech at D: All Things Digital, he says, “we’re trying to move from answers that are link-based to answers that are algorithmically based, where we can actually compute the right answer.”
A search engine is a tool that responds with links. Google is a solution – a way to provide people with answers. Back in 1999, the best way to do this was providing links to relevant sources, and Google did so. Now, as algorithms make it possible to provide answers directly, Google is moving toward that.
Try typing “London temperature” directly in Google. Before the search results, you’ll see a box where you’re shown the most recent temperature for the city and the temperature predictions for future dates.
Or type “Berlin flights.” Before any results show up, you are shown a summary of flights to Berlin. Better yet, if you turn on location, you’ll be shown flights from your location to that specific city.
In other words, you are being provided a direct answer to what you’re looking for. Google’s ultimate goal is to give you answers to even more complex queries, for example, “the best way to lose weight”. One thing they could possibly do is to intelligently compile the most popular advice on losing weight, compare that against scientific research on what works and display the summary at the top.
How is it that Google understands exactly what visitors are trying to get done, and provide such a successful product to solve visitor’s needs? It all comes down to user research.
Start with User Research
Earlier, I said that what customers do is more important than who they are. Ask yourself a few questions to see if it’s true in your case:
- Do you identify more according to your ancestry, or your occupation?
- How do you introduce yourself to others?
- Are you more likely to ask someone about their activities, or their looks?
User research is about how users spend their time, what they care about, and what actions they take. By learning about your prospective customers before beginning any product development, you can create a truly successful product.
Observe what your visitors already do
Use tools like CrazyEgg and ClickTale to get a good understanding of how people interact with your site or application. These are so-called “behavioral tools” which allow you to observe the actual way your users behave.
One particular feature that distinguishes them from Google Analytics is the visual overview they provide. For example, Google Analytics has a very primitive heatmap tool while ClickTale can go as deep as recording each individual visitor actions and mouse move heatmaps and Crazy Egg has a confetti report. I’m guessing that the reason why Google doesn’t still have such features is the processing power they take, their severs would probably collapse if they have given you an option to record each individual user session!
There are many successful stories of companies that used those tools, and one common theme in the testimonials is the unexpected results they’ve got which helped them make a better re-design of their overall sites/landing/sign-up pages and so on.
Before using these tools, write down your own predictions. How do you believe users interact with your site or application?
I cannot emphasize enough how important this is because of the handsight bias which is basically looking at past events and interpreting them as being predictable. Every time you tell yourself “I knew it all along” after an event occurs, you’re probably a “victim” of this bias. Here are some examples:
- After some unexpected event occurs (like the 9/11 attacks or the 1986 space shuttle disaster), there were many people that cited Nostradamus and some verse as a “proof” he predicted it was bound to happen.
- The “analysis” part on news networks after some event occurs. When the Madoff ponzi scheme was busted, you saw all sorts of experts saying “if only this and this was in place, this would not happen”.
- In one study, several students were asked if a particular person nominated for the Supreme court would be confirmed. 58% said yes. After he was confirmed, the same students were again asked if they thought he would be confirmed. This time, 78% said yes.
That’s why it’s important to try and write your predictions first to stop this bias from occurring and realize the unexpected interactions of your users with whatever it is you’re testing. Stop the “I knew it all along” thought before it’s too late.
Ask users what are they trying to do after you observe them
You can do this after observing what they do using various tools. Specifically, ask them what they are trying to get done by clicking on x and then on y and then on z. There are many usability testing and survey tools to accomplish this, but after all, they’re just tools. What you’re trying to get done here is ask the right questions and come up with answers that will improve your prospects understanding.
It’s simple: First observe what they do and then try to find why they do what they’re doing by asking questions. What were you trying to accomplish while clicking on x and then on y? Why did you click y after clicking on z? etc.
To give you a good idea about how to get started with this, I recommend you first:
a) Install behavior analysis tool(s) (Clicktale, CrazyEgg and KISSMetircs are all good for a start) and run it for at least 7 days and make predictions how they’re going to behave before you start running the heatmap tools, as mentioned above.
b) Get at least 10 users and ask them what they are trying to get done with your tool
If your prospects are close to you, you might consider making in-live interviews. If your prospects are all around the world, then get their contact info and conduct an online survey. I recommend you talk to them directly and not just sending them a generic survey via email.
c) After the heatmap study, see any unexpected occurrences and ask your users what they were trying to get done by clicking on y and then on z.
It’s simple to get started
User testing provides UX designers with the information they need to create successful products. Just as marketing teams learn from speaking with users, so can we. And in doing so, you can come away with new insights and lessons that will help you create the perfect solution.