Infomongo

Essays

The Other Kind Of Research

by John Phillips, July 24, 2017

The first step in most UX work is research. You need to understand who is going to use the product and what they want to accomplish. You need to understand their backgrounds. What applications they use and like. User research is foundational.

But there is another kind of research that needs be done early in a project. You need to understand the business goals: how the organization will make money.

An Illustrative Story

I used to work for Local Matters, a company that built web sites for Yellow Page (directory) companies. We had some state-side customers, but most of clients were in Europe.

(I got to go all kinds of places I never thought I’d visit, like Helsinki, Finland. Which was fun, and I totally recommend it. But that is a different story…)

Very early days, before I joined the company, Local Matters had done a bunch of user research. They understood that users of these sites skewed younger. Older people would use the printed directories but everyone else would use the web. And for most types of searches, all things being equal, the best match would be a nearby business.

For example, if you were trying to find a shoe store, a hardware store or the like, generally you wanted one close to home or work. If you were looking for a plumber or a locksmith, then location didn’t matter as much.

Local Matters had also done a lot of competitive research on the existing directory sites and found that often the results weren’t surfacing nearby business. It seemed like we could do better. We could offer clients local search sites that were better tuned to the needs of their users. The company had a lot great technology and smart people. It seemed like a win-win situation.

When I joined the company, we were in the thick of our first customer deployment and things were getting real, even a little testy. As it turns out, the company’s plans were rubbing up against the business model of the yellow pages.

How Yellow Pages Make Money

The directory companies make money by charging small and medium sized businesses (SMBs) to advertise in the books. The books list businesses in alphabetical order by category. They list every business, and offer business various kinds of ads. Some are simple and cheap, like bolding the business name and phone number. Or a business could add a logo or border to its listing. And of course, businesses could buy larger and more expensive ads.

The directory companies had large sales staffs. They had sold a lot of ads to a lot of business. Since the books existed before the internet, the directories tended to sell online advertising as add-ons to existing ad packages in the books.

I joined the company in In 2005, and at that time the internet was a force. It was reshaping businesses and turning over a lot of apple carts, but the directories were still making a lot more money from their printed books.

Top advertisers in the books would have those large quarter page ads shown at the beginning of a category in the books. In fact, for popular categories, the first several pages would all be ads. To the directory companies, it made sense to show their advertisers first in the search in the search results. They wanted to sort the results by advertising tier, platinum listings first, then gold, then silver and so on.

This is in direct conflict with what their users wanted. Remember, users typically want the closest business offering what they are shopping for.

The sheer number of advertisers was also a hurdle. When using the books, you had to view quite a number of pages. You’d page thru the book to find the right category, then page thru that section looking for a store close to you. That meant there were lots of opportunities to display adds and the typical user saw a lot of them.

But when using search site, you don’t expect to view many pages. You’d enter your search term (the what) and a location to search (the where). People would expect that best results would be listed first, that nearby businesses would be given priority. This makes it impossible to show the same number of ads on the web.

If we showed just advertisers, sorted by tier, it could look like there were no businesses nearby, if none of the top advertisers were close. But site visitors knew better, so the results would just look wrong. Plus, people only view the second page of results if the first page was helpful. If the first page seems off, people tend to get frustrated and leave. Again, limiting the number of ads users would see.

So Where Did That Leave Us?

UX work in this space meant explaining these differences to clients and advocating on behalf of users. To effectively represent users, you had to be able to explain the differences between paging thru the book and searching online. To guide clients to good choices, you had to understand their ad-driven business model.

In truth it was a mixed bag. Some of the directories did not embrace the differences. That first client, who offered bold names and phone numbers as an ad product in their books, they carried that over to their site. With some names bold and others plain, it looked kinda like a bug when the page rendered.

Other directories took better approaches. With many, we sorted the results by distance, with advertisers shown before non-advertisers when the distance was even. Advertisers would get a slight bump, but wouldn’t dominate the first page of results.

Wrapping Up

To be an effective UX designer at the company, you needed to understand the directory business. You had to learn what was possible and how to sell your ideas to the stakeholders. You couldn’t be an absolutist and argue only from the the user’s point of view. You needed to understand what users wanted, and also the relationship between the directories and their advertisers. A lot of the job was saying no to things, trying to talk the directories out of bad ideas. And to do this effectively, you had to understand the business realities of an advertising-driven industry.

Need help with your UX? Learn more about John Phillips and Moonsault UX