I understand that you should be thinking of your users all the time but given the current climate where every pound counts, is user testing just for the richer companies? Should you discount your services so that the smaller companies can benefit?

  • 2
    Yes, it's a necessity. And, sadly, all-to-often only a luxury.
    – DA01
    Apr 4, 2011 at 14:37
  • Is a healthy lifestyle a luxury or a necessity? Apr 6, 2011 at 5:23
  • 1
    Basically what you are saying is: "On time of crisis, forget the users, because money comes first!" - uau. :) Nice perspective. ;p
    – MEM
    Apr 11, 2011 at 11:39

14 Answers 14


User testing is a necessity.

Heuristic evaluation will help narrow the gap between user requirements and the finished product, as will good user research, and good integration of that research into the design/development process (e.g. via personas and scenarios). But there is no substitute for user testing as part of a rapid prototyping cycle for edging your product closer to usability nirvana.

I'm a great believer in low-cost, high-speed, high-impact guerilla user testing. In four hours it's possible to run 3 user tests and deliver a report to the project team, armed with nothing more than a laptop and a quiet desk. 5 users is better, 8 maybe slightly better still, but the drop-off in terms of the number and seriousness of the issues you find is so steep that a 3-user run will often be enough for one phase in a sequence of rapid prototypes.

Steve Krug describes this in 'Don't Make Me Think' (Amazon .com .co.uk).

I take your point about the current economic climate, but a morning's user testing need cost no more than 3 users' expenses, and half-a-day's labour from one or two usability people. There's really no excuse for not doing it!

  • If I were a developer company, I'd rather employ a usability expert to do all that for all my projects. Gives you the ability to do user research so requirements will be better from the start, and test more often and quickly in development. So maybe the question is: as a consultant agency, how do you justify using your service instead?
    – Lisa Daske
    Mar 25, 2010 at 10:38

i'm going to continue in this theme - not doing user research can easily and quickly cost you a lot more than doing user research, but at the same time there are plenty of ways to decrease the cost of research. Be creative, be lean, but don't stop doing research. (Perhaps stop calling it 'testing' tho'? that may be a whole other post!)


IMHO rather than undermine the real commercial needs of the user experience agency it's much better to offer a series of lower cost alternatives services.

Such as:

  • Remote users testing
  • Hallway user test

User testing doesn't have to be lab based to be effective.



Cost benefit often leaps when the solution to a tricky issue becomes blindingly obvious in a relatively quick and small scale user test.

That lightbulb moment with a user can save hours or days, sometimes days, of discussion both within internal dev/design teams and when communicating with clients.


I'm in agreement with Ali. The cost of not doing user testing is often much, much larger than the expense of running a small test.

Instead of dropping user testing, or dropping prices, change what you do.

Is lab based testing necessary? It depends. Personally I get much more value out of more informal hallway, guerilla style tests in many if not most situations.

Is writing up a long report necessary? It depends. Personally I find taking the conclusions directly to management and the developers and talking to them is much more effective than producing a large chunk of paperwork.

Do you have to do it all yourself? Train up other people to help out - get more people involved.




I agree with Matt.

Also according to Jakob Nielsen, getting in 1 person will reveal nearly a third of usability problems and only 5 users will reveal 75% - using your network of friends or family members to come in and perform tasks on your website or prototype will give you so much insight into the way the site works and reveal so many usability issues (of course,the more targeted, the better). All you need to offer is some small incentive.


Keeping the user in mind cannot be achieved by just thinking,you MUST need the objective,truely,real user data to input.UT is just one of those ways to help u. Even inside user is better than no user.


If your app is not ultra specialized, you can hire "mechanical turks" at Amazon.com. They are real humans who are happy to test drive your app and provide feedback, for not very much money.


User testing is not a luxury and it certainly doesn't have to cost a fortune either.

I've just completed a study using 30 people. It cost us nothing aside from my time and some book vouchers, but what we learned validated our approach helped us refine the content way beyond what we could have achieved without the testing.

Thankfully I'm not an agency trying to sell my testing services, my organization has invested in getting internal resources capable of doing this type of work. Design agencies that work along side their clients and share their expertise though stand a much better chance in the current climate. Skills transfer is a viable way to add value to any project.

  • And Nathan as touch an important thing here that the OP may not have considered. When we talk about receiving money we cannot squarely think: "If it's cheaper then, IT WILL BE cheaper". That second don't necessarily follow the first. Plus, on a time of real crisis, food is important, shelter is important, sites with usability or w/out usability neither are important. :) Anyone, Nathan gets a point here because: "Add value" to a project is indeed a great benefit of using usability tests, it will make a difference between a success thing and another not so good. The first may save more money.
    – MEM
    Apr 11, 2011 at 11:32

It depends how complicated the functionality of your website or software is.

The more complex or unfamiliar to users - the more you'll probably need to test it.

  • Indeed. Let's face it, a 6 pages website is not likely to need usability testing at all. Common. Some AI perhaps, but no more then that. On the other hand, however, it will not cost you that much, in a way that if you do you may be bankrupt.
    – MEM
    Apr 11, 2011 at 11:35

Go on, someone say testing isn't worth it...!

OK, I'll try. Formative testing where you do a few rapid iterations of test and redesign is always worth it, even if you are just re-using common design patterns. But most people aren't going to outsource this type of testing.

It gets a bit more difficult when you are trying to justify recruitment and facilitation costs for a larger summative evaluation with a glossy report. That's where remote unmoderated testing comes into it's own so I agree with Matt about low cost options.

But in the end I'd use my usability testing budget to get the best designer I could afford and skip formal testing altogether**. The best tests can tell you what users are struggling with and why, but you've still got to figure out the best way to fix it...


** Caveat: All our applications are for internal use so we have users on tap and can run guerilla tests very easily for the cost of a few donuts.


Fundamentally what is user testing and research there to do. Will it make your users like the product more...probably. Will it look like you care what your customers want...yes. But when it comes down to it any customer research potentially allows the company to not make stuff which no one wants and produce products which are needed.

User research is a 'soft' technique that enables 'hard' decisions.


The basic premise of usablity testing early on is that it helps define the product design process along the requirements of the user and potentially enhances its acceptance in the market.

Hence if you are looking at a large scale product where it would be potentially expensive to make drastic changes to a product based upon user reactions on releasing it to the market,you will need to ensure that you are well aware of the needs of your user and their reactions to your product before you release it to the world and that kind of information can only come from doing usability testing .

However if you are looking at small scale project such as small websites who can be quickly altered based upon the user reaction and behavior after going live, usability testing might not be a required option.However since first impressions of a site matter and if people are not favorably disposed towards a site,there is a possibility that the site might lose those visitors once and for all even if it cleans up its act later.

First Impressions Count in Website Design

That said and done,there are a lot of good resources on how companies can be encouraged and do usability testing on a budget :

Quick and Dirty Usability Testing: Step Away from the Book

Why You Only Need to Test with 5 Users

Streamlining Usability Testing by Avoiding the Lab

Usability Testing on a Budget

Quick and Dirty Remote User Testing


User testing needn't be feared or difficult. By establishing very clear goals, the who and how of the test become clear. I have done many tests where I could ask literally anyone and the duration was exactly 5 seconds - this is the '5 second test'. Doing this often throughout the process will produce many lightbulb moments which will mitigate a lot of the need for large and 'expensive' tests, and even these needn't be large or expensive. I have done fairly elaborate tests with a statistically significant number of representative users in a few hours for a few hundred dollars. The key is you - do the research and preparation and get started.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.