Synthetic Monitoring
Did you ever notice how many UX or User Experience conferences and written words there are about this topic?
That is all well and good but they all typically miss the ultimate point – performance. It does not matter if you have the best content, the best graphics, the best layout. You will let yourself down if you choose to ignore the response times that a typical end user experiences, wherever they are.
The rule is – not time to test, don’t bother to deploy !
Sounds harsh but do you really want to expose your brand to possible web related performance? Much better to de-risk and take the safe option.
Here’s my How(se);
- Understand the business objective of a service/application on your site
- Build a service/application to match that business objective
- Do basic low level capacity testing in the development process
- Perform scalability/Stability and spike traffic testing prior to deployment
- Make sure you have enough headroom in your infrastructure for growth in concurrent users
- After deployment, monitor your performance 24×7, get alerted if service degrades and ensure your monitoring tools provide you with information to help you fix problems
This is all Best Practice – is there a good reason why you don’t follow it?
So let’s agree, UX really does include performance to be meaningful
If you agree or disagree, Im willing to discuss.
Best Regards
Mike Howse, VP International
office: +44 (0)1628 440826