Join us

How to Conduct a Major Redesign of a Website and App. The Head of Product Design at Parimatch Tech Reveals All


Redesign always entails risks. Paradoxically,  the launch of an improved version is always accompanied by a decrease in metrics because users don’t usually want to change anything. In cases like these, no single scenario can guarantee success. However, several similar cases can teach us a thing or two.

In the run-up to the CASES online conference on digital products, Aleksey Stiques, Head of Product Design at Parimatch Tech, told Vector about the various stages of the redesign process and the best way to approach them.

Product interface development

This process is usually iterative. As the product develops, new functions, content blocks, and other elements are added to the interface. Iteration is standard practice when the product is evolving as it allows users to suggest modifications based on their needs.

There’s always the hidden danger of creating an interface with so many elements that it confuses the users. Excessive features can accumulate due to lack of oversight, leading to functions that either do nothing or are only meant for particular user segments.

At some stage, the number of unrelated elements reaches a saturation point. By this time, cosmetic changes are out of the question, and the product has to be wholly rebuilt so that it looks attractive and functions properly. First and foremost, this information is useful for the old-timers who’ve been in the market for more than five years.

The beginning of the redesign process

The redesign process definitely shouldn’t begin with references—these are the last things your designer wants to see. After all, the look and functionality of a competitor’s products are usually the results of their experience and the mistakes they made in the process.

It is much more important to make sure that everyone involved in the process understands the purpose of the redesign and knows what they want to achieve.

In terms of the visuals, there are always two options:

• be like all the others;

• be different.

Both options are pretty good. But it is crucial to understand:

• what makes the products look similar;

• what is the pattern for the industry and the client;

• what to “break” to be different;

• how not to damage the overall UX.

Once you’ve answered the questions above, you can start collecting your ref-lists.

What are the indicators of success?

The most demonstrative indicator is the conversion into the required actions. You need to understand your funnel and track the critical points of business action—at what stage do your users leave: registration, login, or transaction? Then you need to modify elements that can bring good results with little effort.

It is also important to remember that there are a number of elements that may not directly affect conversions but, at the same time, can damage the overall perception of the product and brand.

Users don’t need to be qualified designers to notice a product that doesn’t work as it should or looks unattractive.

It’s therefore essential to maintain a basic level of product hygiene—to ensure the compatibility of solutions and provide resources to support visual design.

Measuring redesign efficiency

If a specific metric is introduced as a key feature of the redesign to increase the speed index or reduce the error rate, for example, then the efficiency should be measured by this metric.

But the redesign is not always about numbers. The idea of updating the product for a more representative visualization of the brand’s emotion may lie at the root of the process. This is a matter of your feelings.

You might engage focus groups to measure the effectiveness of the redesign or you may have someone on your team who’s an expert in the field. In that case, they would validate and identify whether or not the emotion is appropriately communicated. But even with a person like this in your team, you still need to track the funnel.

If drastic changes to the product make users feel rejected, the metrics decrease. Although they generally tend to level out after a couple of weeks and may even increase, it’s important to double-check.

Try migrating your audience to the new interface in parts—one after another. In this way, after migrating one group of users to the new design, you can identify what they don’t like and fix it before migrating to the next group. As a result, the migration will be smooth for the audience.

The case of Parimatch

When migrating the Parimatch website to a new platform, the main goal was the migration itself, not the redesign. Nevertheless, since we were writing a new front, we also decided to improve the product for our pro audience.

We divided the audience into several groups: by experience, profitability, and the platform they play on. Going forward, we planned out when each group would migrate to the new platform and measured the activity of each group after the migration.

Because the migration was gradual, the number of mobile phones increased. There was a drawdown with desktop users—so, we paused the migration and delved deeper into the study of this particular group. We found and fixed all the elements that were inconvenient, complicated, and time-consuming for these users.

This approach has allowed us to increase all the indicators of the previous version: the number of conversions to the key action and the speed index. And the new platform has increased the download speed and the ability to withstand peak loads for simultaneous bets per minute.

Please rate the post
1 2 3 4 5
Thanks for your rating!