Monday, October 23, 2006

product innovation and user reality

As I pursue the complex and mysterious realm of API design, testing, and promotions, I have made some wonderful friends and received some excellant advice.

Without revealing the real inner secrets of API marketing to the developer community, for which my client is paying large sums of cash to me, I will share and discuss some of the core issues involved.

One of the helpful resources for this client project has been the Microsoft Developer Network (MSDN). Krzysztof Cwalina, and Steven Clarke, are active in software usability testing and analysis for Microsoft, and have been a big help, both via email and through online articles and videos.

You know it's obviously true: the best way to design a product is to find out what people are having trouble doing, and come up with a way to make it easier or more efficient to accomplish these tasks.

How do you find out what difficulties people are having? Not by asking them, that's very limited and prone to memory errors and ego defenses. Humans can be very inexact in their confessions and descriptions.

Here's what Steven Clarke has proposed, to get a better idea of what people need.

"Help us learn about how you work"


We want to make better products.

[VASPERS: Innovation and customer satisfaction originate with a strong desire to provide products that meet the real needs of real people, as they try to (1) solve a problem (2) fix a broken (3) enhance a lifestyle (4) enjoy an activity (5) pursue an interest.

This includes work, play, hobbies, and personal passions, like music or art.

A successful product is a product that makes a user successful at some task.]

One way of doing that would be to design a product that we would like to use and then hope that others outside of Microsoft would also like to use it. Sometimes that approach works, other times it doesn't.

[VASPERS: I think too many software products are made by advanced users for advanced users, forgetting that most users are not advanced.

Typical business users, along with the vast majority of non-technical consumer users, know only enough about computers to do what they like, or are required, to do: email, spreadsheets, VoIP, video chat, photo sharing, video uploading, music mp3 downloading, and other fun or work-related activities.]

Not everybody writes the same kind of software that we do and not everybody likes to work the same way that we do. So it's likely that not everybody will like tools that we design for ourselves.

But the only way to learn if we need to design products differently is to understand how people work and how they use the tools that they use.

One of the best ways to gain that understanding is to visit people at their work and observe what they do.

[VASPERS: The user observation test is the only way to know the true usability of a product. To actually witness the product in use, by a customer, in their attempt to solve a problem. This is also a good advertising strategy, to show the customer using the product to accomplish some task.

Not surveys, not focus groups, not questionnaires, not interviews, not intuitive hunches, not educated guesses. These can be helpful to a limited degree, but watching users is unquestionably the best strategy.]

Only then can we really understand what they do, how they do it and why they do it that way. Other approaches such as focus groups and lab studies are useful, but one thing that we've learned is that it's easier for somebody to really explain what they do by doing it, rather than just talk about it.

[VASPERS: When a person explains what they do, they forget many details, especially the preliminaries, the set up tasks, how they prepare.

People also tend to present themselves in a good light, and exaggerate their abilities, their performance speed, and their difficulties. Nobody wants to appear stupid, especially to the expert who is questioning them. More especially: if you both work for the same employer. Nuff said.]

So, we're trying to find people who are willing to have us come and visit them for a few hours to simply watch them work, and ask a few questions along the way. We're not asking people to take time out of their work. Instead, we'd like them to do whatever they would normally be doing, just with us watching them. If you'd be willing to participate, please let me know.

[VASPERS: This approach is superior to any other usability study. To quietly watch people at work or play, rarely interrupting them, taking notes, asking most of your questions when they're done, if possible.

In my MUNOT or Multiple User Non-intervention Observation Testing, I remain silent. This is essential.

When a test subject gets stuck, I merely offer sympathy, not solutions. You must allow test subjects to struggle along, with no coaching, no advice, no clues. Otherwise, it's an artificial situation, not reflecting the actual work situation. The web site or software has to be self-describing and user-guiding, as intuitive as possible, which means adhering to norms, except in cases where you have extremely good reasons to deviate from what most users expect.]

Obviously security concerns are an issue - we will not share any proprietary information that we learn about while visiting you.


To rule the world, one must first understand what's really going on within it. Steven Clarke is on the right track.

Now, what can you do, today, to learn more about your blog readers or customers?


No comments: