Software Quality Assurance & Testing Stack Exchange is a question and answer site for software quality control experts, automation engineers, and software testers. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

We have now been conducting browser testing on our websites for over 4 years and are conducting a review on how we work. We split browser testing down in the following areas.

Functional - All keys processes should be tested to ensure they continue to work. Automated

Rendering - All sites should render in the same way on each browser

Usability - All sites should maintain the same standard of usability across each browser.

We find the quarterly checks on all sites to be time consuming, although have proven useful in maintain the standards of the site.

What does everybody else do? How important do you see browser testing?

share|improve this question
There's going to be a lot of opinions on this question, but I think it would be a good candidate for a general reference question – Kate Paulk Jul 21 '14 at 10:54
up vote 1 down vote accepted

Some things I've consider from a strategy point of view in various situations:

Could you move to a 'test based on change' approach? Could you monitor the industry for software upgrades (of for example browsers) and new screens (with different dimensions and characteristics) and thus test in response to change as opposed to periodic testing? Clearly you do monitor change to a degree because you have to upgrade tests to reflect these changes.

How closely do you monitor the system to try and discover issues as they occur. Are all web server and application logs automatically monitored? You can see indications of issues, users taking too long to perform a task, transactions abandoned, etc.

Could you move the testing requirement on to your clients? How easy could you make it for your clients to tell you about issues and problems? And how do you thank them? Entry into competition or a credit perhaps?

Do you actively try to measure the impact the issues you have discovered in the past have had? For example characterising issues by:

  • percentage clients impacted
  • length of time existed
  • loss of revenue
  • estimated type of issue: outage, limited access methods, loss of functionality, lack of usability, etc.

It can be interesting and knowing the estimated cost of issues might be useful as part of your review.

Clearly every situation is unique and I know nothing of what you do so just some thoughts that I hope may stimulate further thoughts.

share|improve this answer
Some interesting points. We do no monitor the what changes are being made to browsers which is why we only conduct site reviews every 3 months, although we struggle with a change and risk based approach as we find not all of the release note are accurate (Especially with IE). I like the idea of having continuous testing through the end users (Clients in your email) that could work. – Tim Mottram Jul 21 '14 at 12:31

Where I work, there are three web applications that need to be maintained, with a fourth (which is intended to unify all three existing applications) in development. Each has a distinct purpose:

  • Application 1 is an enterprise-level employee and payroll management web application used by both internal payroll specialists and externally by customers.
  • Application 2 is a much smaller site allowing individuals to view and print their paystubs.
  • Application 3 is an internal-facing portal for managing customers, schedules, and the like.
  • The new application is starting with App 2's functionality and will gradually integrate with, then replace App 1 and 3.

I'm the one and only test specialist here, responsible for all testing. My team is 10 people: me, 7 developers, and the leads.

My plan is:

  • Functional - for the new app, build automated regression around key functionality as it is released and stabilized. For the existing applications, only App 1 gets manual regression testing, and then only for a small subset of key functions (due to time constraints). I plan to automate this as time permits.
  • Appearance/Rendering - this is not tested in a systematic fashion. I check all new features in the existing applications for consistency with the rest of the site (using all the supported browsers), then hope nothing changes it.
  • Usability - apart from testing that the shortcuts the applications are supposed to allow continue to function, this is not formally tested. The internal users of the applications are quick to notify the team when usability is an issue.

In general, I think you'll find that most places will automate functional testing as much as possible, and manually test appearance/rendering in different browsers. Usability testing will vary depending on whether the users are expected to be trained users or general public - for applications where the user base is the general public the specialist testers can often provide enough information about usability. With a user base of trained power users, it's often better to have them perform the usability tests.

share|improve this answer
Do you test the browsers in beta or once they are released? – Tim Mottram Jul 21 '14 at 12:35
Generally before release, although if there are major look/feel changes they'll get a quick look. – Kate Paulk Jul 21 '14 at 16:25

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.