111: Speed UX

There are two speed metrics: 1) based on actual pageweight — you have offloaded some weight (sharing with Picassa); and 2) perceived load time (page rendering or recognition). As long as you are “perceived” as under 2 seconds you are fine.

Google owns Picassa. They are using good technology to speed up the image “delivery”. You benefit from their expertise and wonderfulness. This is a good example of benefiting from cloud services.

I classify speed as “obstruction” when it is bad. It’s weight is twice any other factor. You have zero UX after 9 seconds. It doesn’t matter how beautiful or desirable the content is. People will not tolerate a slow load. You’ve ruined the experience. Like kissing with bad breath, it’s memorable for the wrong reason. The first-impression is ruined. Obviously, instant loads of under 1 seconds are the best. Speed is then transparent.

Now try your tablet speed on a commute or remote location –not near a home wireless router. :) A real test of mobile user environment.

The 2-second threshold is not an “Internet” created parameter. It has been proven for decades a threshold of human tolerance for presentations. In other words, it’s more classic (hardwired) than fad (habit). People ARE becoming more impatient and intolerant but only when the 2-seconds is violated. Their annoyance is more pronounced after 2 second waits. Up until then everything was fine.

Like McDonald’s has shown American’s will wait 5 minutes for food. Then they get more and more agitated with the wait and will even leave or make threats. That is for food. It’s different for pages.

www.useit.com/papers/responsetime.html

If you liked this tutorial, please raise your hand. :)

110: Interview with Joe Harris

Can development teams do rapid user satisfaction testing themselves?
This would be called a focus group or simply emailing survey questions. You know how good those are and how long they take to assemble. They tell you what you want to hear. Teams sometimes try to do testing themselves by using in-house staff (hardly unbiased) but with interviews or paper and pencil or email –not electronic automation. It is more expensive to analyze –not to gather data. You end up with a pile of reports and video on your desk to try an intuit and sift what it means. They don’t need complexity. Overkill creates delays.

Simple and fast is better.
There are “survey and questionnaire” subcontractors –but the turnaround is slow because it doesn’t indicate what to do next. You still have to have a brain. The problem is the project is stalled or may be turned in the wrong direction while waiting. Paralysis by analysis. Nor do traditional methods necessarily produce a baseline for comparisons on subsequent iterations (two week sprints). The user experience is bad with questionnaires (no seamless ecosystem). I take “online tests” to evaluate competitor weaknesses. Our “voting ecosystem” is fast and painless by comparison for the tester and instant results to view by the design team.

How much are you really saving companies after you charge your fee?
They literally can double their output or half the cost, etc. It’s efficiency we sell. How much is that worth? Our potential fee is $50 per iteration (every 2 weeks) per project ($100 per month). That is peanuts compared to the potential gain. If it’s a big million dollar project (not uncommon), they will save a half million dollars. If it takes a year, that means we only charged them $1,200. It is disposable reporting. We are not talking speculation about profit or sales. We are talking budgetary improvement. “X” is alloted. They save 1/2 of that.

Disposable feedback doubles project efficiency. << Is this a new positioning statement?

Imagine building a product where you NEVER asked the customer what they thought about features until you spent 2 years and $250,000 dollars. Is that smart? No! But that is how websites are usually built. Then it’s all demolition, maintenance, and repair work after that. Very ineffective. If the customer/user is involved in tweaking and reviewing the design, they save 50% of the development cost. The shelf-life of a website is 3 years. It’s almost time to start again!

The best analog comparison is lean manufacturing which also requires lots of in-process feedback for good quality throughout the process –not just at the end. You build in the quality, not rebuild in the quality.

In your ROI calculations, are you also including the cash flow from sales generated from the time you have saved them.
Nope. Too fuzzy. Just budget and project improvement. No direct or potential sales income is calculated.

If they are 50% faster to market, they have the 50% time saved to sell their product.
That is true. But usually they already started pre-selling the product. That is the marketing guys have. Not the programmers.

Or am I off base?
Your questions are very valid. I’m not sure I’ve answered them well. I’m thinking about what you have asked. Some is immeasurable. Like asking “What is the benefit from wearing lipstick?” How much is self-esteem worth or a really good kiss?

Sometimes there is a psychological benefit from just knowing you are doing it right or installing more lights in production areas for good morale.