How should I improve my website?

on 27th May 2015

(Last updated 11th March 2016)

Websites are ever-changing, evolving pieces of art. Unlike print, they have the ability to be constantly updated and evolve over time. Be it a usability enhancement, a user experience tweak or a simple bug fix - a website is never finished.  How is it best to approach these improvements?

For small, simple improvements this is a straightforward task, you make the change, test and launch. But what about a project that has a large lead time, spanning a couple of months?

Imagine this scenario:

You run a large eCommerce site where user testing has identified a large flaw in your checkout process that, due its technical nature, will need 3 months to update. There's a temporary fix that will take just a couple of weeks to do but only resolves 70% of the problem and delays the delivery of the full project. What do you do?

"If a job is worth doing, it’s worth doing right first time"

The push back we receive from clients in this situation is a hesitation about only doing ‘half a job’. Why spend time filling the cracks when you’re about to pull the whole wall down and rebuild? Whilst it is rewarding to launch a big update in one go that makes a fundamental change, it’s often not the best way forward.

Going back to our scenario above, lets look again at the options in a different way. Do we

  • Leave our website 100% wrong for 3 months before releasing the final update

  • Leave our website 100% wrong for 2 weeks, make it 70% right for 3 months before releasing the final update.

Surely 3 months of 70% right is better than 3 months of 100% wrong?

Eat, Sleep, Test, Repeat

Releasing incremental improvements also has added benefits. By working in a more agile way it provides the opportunity to test your proposed solution as you go rather than right at the end. By then you have already invested all your time and resources and it’s too late to make any changes as you’ve already finished! The additional learnings you receive lets you make small tweaks as you go and gives you a more complete final product than you may have done otherwise.

One of the secrets to success of this approach, as with any agile project, is a strong project management team. Once the approach of evolution rather than revolution has been agreed, it is easy for teams to reach the 70% stage and get distracted by a different issue that is shouting for attention. This then fulfils the prophecy of just filling in the cracks and not solving anything fully. You start to build your website on shaky foundations as a temporary fix is layered on top of another and another. You need to ensure that this is seen as one complete project with various milestones along the way, rather than two.


The digital industry has picked up a number of legacy phrases and perceptions from the print world, however the one that causes the biggest roadblock is the expectation that everything must be perfect before it is released. Whilst in print this is still true, you can’t update a brochure once it has been sent out, it doesn’t apply to digital. A website is ever evolving. Not only are incremental increases possible, but with the ability for testing and instant feedback it should be actively encouraged. It is a great way to ensure your website is always moving forward and responding to the changing needs of the user.

Just don’t forget about that last 30%.