Mitt Romney’s get-out-the-vote tool performed poorly on election day. Politicians are pure marketers. What can we learn from the experience?
The tool’s failure was not just a technical problem: It was an interaction design problem.
Interaction design is an often overlooked aspect of “experience design” – making the quality of a person’s entire experience the project goal.
Ad agencies are good at the emotion end of experience design; engineers are good at the functional end. But marketers (and even technology companies) struggle mightily with making it come together – the interaction design.
We’ve put up with it from the technology world for decades. Think of all the frustrating experiences you have with your DVR, your smartphone, your computer. All designed by people whose livelihoods depend on getting it right.
If they can’t get it right, what chance do we marketers have?
The good news is that it is fairly well-understood how to get it right. There is the stuff that we know from experience, that expertise can help you avoid. And there is the stuff that can only be fixed with process.
It’s the process part that gets skipped most often.
Consider the Republican get-out-the-vote tools that performed so poorly on election day. Its problems were identified in a rant by a frustrated Romney volunteer:
- Critical information was emailed in a 60-page PDF. PDFs are a relic of the print era and are meant for printing. They are not meant for digital consumption. (Indeed, the author ran into trouble because his printer did not work right.) Most low-cost printers that people have in their homes are adequate for printing a page or two at a time; printing 60 pages was asking for trouble.
This is well known among experienced interaction designers.
- The checklist the tool provided was missing a crucial item: the pollwatcher certificate. It included two entries reminding people to take a chair. Proofreading is proofreading, but a usability walk-through would likely have exposed the problem.
- Three contact avenues were provided: a help line, a legal line and an email address. The author got no response from any. Again, a well-designed usability evaluation would have exposed this problem.
- The tool was billed as an “app”, which caused people to look for it in the big app stores run by Apple and Google. It was not a standalone app, though, but a mobile-optimized Web site that operated like an app. The naming created confusion. Language and labeling choices are critical to success with interaction.
- The Web site was at a secure address, so you needed to enter https:// in the address line, not the much more common http:// . This can be accommodated by a very simple technical trick on the server side, but that was not in place. While a person with a technical mindset would not regard this as a mistake – or worse, would attribute it to “user error”, a usability professional would identify it as a problem that must be addressed.
As you can see from this mix, it’s a bunch of different kinds of problems: proofreading, labeling, hardware, technical.
Interaction designers can help. They are not just concerned with the shape of a button on a Web form; they are concerned with creating an experience that is easy and intuitive in real-world conditions.
That’s understanding the way people use platforms (the PDF issue), ensuring comprehensiveness (the bad checklist), emphasizing the forgiveness of the system (the inadequate support), and knowing the way people actually act as opposed to the way they “should” (instantly assuming that “app” means a Web store product, not entering the necessary ‘s’ in the URL).
A lot of these problems can be avoided by having the right expertise and experience on the team. But that’s not nearly enough.
Interaction designers have known for decades that one tactic far outperforms any other: prototyping and iteration.
There is nothing more effective than watching someone trying to do what your tools are supposed to enable them to do. Good interaction designers get you a big head start, but there is no substitute – and there is science behind this – for watching ordinary people trying to use something that someone else has designed.
This isn’t a beta test. Beta testers are evaluating against a plan. They know, or are told, how something works. They are trying to find programming mistakes. (For a long analysis about how the Democrats tested their own systems, see “When the Nerds Go Marching In” in the Atlantic.)
Usability testing identifies design mistakes.
This is, and always has been, a tough sell. It’s only with the ascension of Apple that ease of use has been understood as a differentiator.
In the advertising world, we spend weeks looking at and refining a digital comp that is flashed up on a big screen. But we are reluctant to spend any time and money prototyping and evaluating our digital designs.
We largely assume that our designs work. In the winner-take-all-world of politics, that assumption can be deadly. Here’s that Romney volunteer on the consequences:
“So, the end result was that 30,000+ of the most active and fired-up volunteers were wandering around confused and frustrated when they could have been doing anything else to help. Like driving people to the polls, phone-banking, walking door-to-door, etc. We lost by fairly small margins in Florida, Virginia, Ohio and Colorado. If this had worked could it have closed the gap? I sure hope not for my sanity’s sake.”
What’s happening with your digital experiences? Do you even know?