skip to Main Content
A Complete Redesign Vs. Incremental Testing: When To Use Them

A Complete Redesign vs. Incremental Testing: When to Use Them

Redesigning your website is an exciting but often risky project. The temptation to start from scratch often leads to making decisions based on a lot of assumptions, and many underestimate the value of testing those assumptions.

To make educated, data-driven design decisions, testing some elements first might be a great way to start. In doing so, you might find that smaller changes over time will give you a better return.

There are actually cases to be made for both a radical redesign and incremental testing. In this post, we’re going to look at the costs and benefits of each, and how they can complement each other.

Here are some questions that might help you to make a decision:

  • Are you working with a website that is severely underperforming in multiple areas?
  • Are there some areas you know you need to improve? Are you following all the current web design and user experience best practices?
  • Is your website working reasonably well? Are you trying to fix something that is wrong, or just improve on something that?s already good?

Incremental testing is good for keeping your website running smoothly and your conversion channels optimized, while a redesign is something you might be tempted to consider if your website is dated or severely underperforming.

However, before you consider either, you need the analytics from your website to; a) gauge how specific elements and the site in general are doing, and b) have a benchmark for measuring improvement, justifying the investments in testing or redesigning your website, and demonstrating ROI to any stakeholders.

If you do decide to redesign your website, A/B testing can help you determine which direction to take on individual or combinations of elements. You can see if your changes are sufficient or dramatic enough to get the results you wanted, and determine how effective the changes were after the fact. In this way, a redesign and testing can work hand-in-hand.

Let?s take a closer look.

Download our quick comparison of incremental vs. radical A/B testing here

Incremental testing

The main benefits

The main function of incremental testing is to gather high-integrity data in order to make optimal, data-driven decisions. With incremental testing you test option A versus option B to determine whether you have a clear winner. Sometimes you won?t have an obvious answer, but this is one reason why incremental testing is something that should be looked at over the long-term.

The main values of incremental testing are that you can

  • Gather a lot of data on independent elements and their combinations
  • Test at a pace that suits your company, without making a radical change
  • Get specific performance metrics on parts of your sales funnel
  • Isolate areas that need to be optimized

Incremental testing is suitable for most websites, particularly if you don?t have obvious, severe issues for which there are straightforward solutions. It makes sense to test out those smaller changes over time, rather than make radical changes all in one hit, if possible.

Here are some reasons you might choose to test incrementally:

  • You identified a few areas of high traffic with low conversion rates on your site. Overall, you get sales, but you?d like to improve these problem areas.
  • While your website generally does well, you have certain products or desired conversion goals that are underperforming. You?d like to see those underperforming areas pick up.
  • You might have common areas that receive user feedback or cause friction. You?d like to improve those areas first.

Redesign vs. incremental testing

Things to consider

As with anything you invest in, there will be some costs, prerequisites, or even some risks that you should consider first. Here are some that are associated with testing:


Prior to making a decision, do you have enough traffic to be able to conduct tests? Without a decent number of monthly visitors, it is very difficult to gain reliable data to make decisions. The data you do collect can be corrupted for a number of reasons. We wrote about this recently, and talked about what you can do if you are too small to test.


The upfront costs of incremental testing tend to be much less than a complete redesign. This makes sense because you can test smaller things out at a more sedate pace, and be more conservative with your spending at the same time.

Simultaneously, this approach allows you to make any changes to your website incrementally, allowing you to optimize sales, which will help to cover your costs.

Opportunity Costs

One thing to consider carefully is the overall state of your website. The process to incrementally test areas of your website over time, and make changes based on those outcomes, may be a longer one in terms of getting results versus the process of a redesign.

Let?s say your site is severely underperforming overall, but the guidance of an experienced web designer could turn that around for you with a redesign. There would be an opportunity cost with choosing not to go for the redesign and to test incrementally instead. If your website was really in a terrible state, there?s a chance that you could lift sales more quickly with a redesign.


The main risk of incremental testing is that you might not get significant results. We find that just three out of 10 A/B tests are clear winners, so you can expect that you?ll need to keep implementing tests to find something that will help you to optimize.

Never underestimate the skill and experience required to do A/B testing well. While a redesign definitely requires a huge amount of knowledge, if you don?t have a good grasp of A/B testing, you can find that a steep learning curve, too. You might run tests that a more experienced eye could tell you will be worthless, for example.

Redesign vs. incremental testing


A complete redesign is where you make the decision to go big. You do a complete website (or page) overhaul in one hit so that it is radically different than the previous version. This sounds risky, right? So, why would you choose to do this?

Why choose to redesign?

You could look at the difference between incremental testing and a redesign like this; the incremental testing is excellent for getting your website from good to optimal, while a redesign might be in order to get your website from terrible to good.

Incremental tests optimize a good website; a redesign takes a website from terrible to good Click To Tweet

Here are some common reasons for a redesign:

Your design is terrible

Your website is a dinosaur. The design is so 2003 that it?s difficult to look at and certainly doesn?t encourage people to hang around, let alone follow the CTAs through the sales flow to a conversion. Perhaps the design is particularly unfriendly to the user, such as this:

Redesign vs. incremental testing

Source: Topdesignmag

You?re missing ?best practices?

Making incremental changes might be a waste of time and money if the site is outdated. A web designer might look at your site and point out some well-documented ?best practices? for website design that your website is missing. You might start making more sales just by implementing the radical redesign first and taking those best practices into account.

To add here, if you happen to have some benchmarking data that shows your site is far inferior to your competitor?s sites, then a redesign might be a good choice. While you don?t simply want to copy them with a ?grass is greener? outlook, you should respond to significant differences in the experience and level of service provided.

Your website has high bounce rates in all areas

If your website has a high bounce rate everywhere and you?re just not making the sales that you should be, this can be a case for a redesign. Perhaps the bounce rate is associated with an outdated design as above, or, it?s possible you don?t have clear CTAs in place.

In situations like this, you might find that gains from incremental changes are tiny, or even non-existent. You need the big change to radically turn around your results.

Your website isn?t mobile responsive

This is similar to the design ?dinosaur? category as above, except it is an issue with the technology behind the website. If significant proportions of your traffic come from mobile, you have a clear problem to fix.

Things to consider

Here are some considerations for a complete redesign:


A complete redesign can be jarring to users. Humans tend to be averse to change and might balk at a disruption to the user experience, which is risky for your business. You can test radically different design concepts at once, but you can be affected by visitor bias, and they may react more negatively than they otherwise would to smaller changes.

On the other hand, if you have significant evidence to suggest your design is hindering the experience, the change may be welcomed!

Be aware that sometimes minimal changes are really all that are necessary. An incremental approach may give you more significant data and the opportunity to ease users into the changes.

A complete redesign might also require a huge amount of coordination and resources, so all of your time isn’t sucked into the redesign, rather than spread incrementally. For example, in your company, how much time might need to be spent convincing stakeholders of a radical change?


A complete redesign will typically cost more, at least upfront, than incremental testing. You invest a large outlay in both time and money, rather than being able to spread it more conservatively over incremental changes.

Redesign vs. incremental testing

How testing can support your redesign

First of all, are you redesigning for the right reasons? Sometimes website owners fall into the trap of placing too much significance on what their competitors are doing, assuming it must be better. Another issue that crops up that is a poor reason for redesigning is that the website owner is bored with the current design.

To avoid these issues, make data-driven decisions. You can test ahead of a redesign to get benchmarking data and see if there?s anything you?d like to keep. A redesign doesn?t have to be the difference between night and day.

Also bear in mind that testing and a redesign really can work hand-in-hand. Doing some tests during the redesign process are a great idea to get that baseline data, rather than simply starting from scratch.

You want to identify clear ?winners? of tests and better understand user preference for your website. For example, you might have the aim of eventually having a completely redesigned site, but there is the opportunity to test key junctions of the sales flow throughout the process. You can make sure that your changes are based on true data, rather than opinions or best practices that you have read about.

Following a redesign, test the efficacy of your new site against the benchmarking data you gathered. This will help to justify the changes to any stakeholders. You can now look at some A/B testing to optimize website elements and further boost your results.

What are the pros and cons of incremental vs. radical tests? Get our quick comparison here

Final thoughts

In most cases, incremental testing is the more effective route to take. Consider the cost involved of a complete redesign, along with the logistics of getting stakeholders in the organization to agree to such radical changes. Projects often drag out due to their size and are a large commitment for the company.

Incremental testing gives you the opportunity to demonstrate significant results, with lower overall risk to the business. You can set smaller goals and deliver results more frequently over time.

However, if your site is an overall non-performer, this is a good case for the complete redesign. It is important to have clear, measurable objectives, which you can support through testing.

After a redesign, look at what you can test incrementally and implement a program to do so. This helps you to build a clearer picture of what actually makes a difference when it comes to achieving your site objectives.

Back To Top