How to Evaluate if a Product Requires Redesign
These days, there are tons of amazing concepts on social media — stuff that grabs your attention right away. A lot of designers love to redesign products to show off their creativity and vision. But is that always a good idea? Do all these products need a redesign without thinking about how users interact with them? How do you figure out if a product really needs major changes or not? In this article, I’ll walk you through the method I use to make that decision.
A good designer always prioritizes improving the product. Keep in mind, redesigning the product should be your last resort. To ensure you’re making the best choice, it’s crucial to carefully review both customer behavior data and business goals to gauge whether the existing product structure adequately serves both users and the company’s objectives. For instance, if we’re contemplating a redesign of the purchasing process, what specific data should we analyze? What’s our main goal? Let’s say the goal is to boost sales. The initial step is to establish the business’s sales targets. What strategies has the marketing team tried that did not succeed? Afterward, we assess the available data, analyze user behavior, and conduct interviews or usability testing sessions with users.
However, the data from interviews and tests is quite restricted. To ensure a well-informed decision based on big data, it’s essential to employ specific methods such as the System Usability Scale (SUS) questionnaire. This method will offer us valuable and transparent feedback not only on the product but also on the overall service experience we deliver to our customers.
To conduct a more precise evaluation, it’s better to administer this questionnaire separately with each demographic category. For example, administer it separately to males and females, or separately to new customers and loyal customers. This divided method gives a more precise insight into the persona types that may pose challenges in attracting.
Once the persona categories are identified, we proceed to design the questionnaire. This questionnaire model consists of 10 questions:
- I think that I would like to use this system frequently.
- I found the system unnecessarily complex.
- I thought the system was easy to use.
- I think that I would need the support of a technical person to be able to use this system.
- I found the various functions in this system were well integrated.
- I thought there was too much inconsistency in this system.
- I would imagine that most people would learn to use this system very quickly.
- I found the system very cumbersome to use.
- I felt very confident using the system.
- I needed to learn a lot of things before I could get going with this system.
Each question next to the Likert scale will appear as follows:
The scoring method is as follows.:
Subtract 1 from the score for each odd-numbered question. For each even-numbered question, subtract its value from 5. Total these adjusted scores and multiply the result by 2.5.
After completing these complex calculations, you now have your score out of 100. This score is not a percentage but provides a clear indication of your performance.
Here’s a breakdown of what your scores mean:
- 80.3 or higher: A — Users love your site and are likely to recommend it.
- Around 68: C — You’re doing alright, but there’s room for improvement.
- 51 or lower: F — Focus on usability improvements immediately to turn things around quickly.
Make sure to stick to the questions in the right order to avoid messing up the scoring.
Just so you know, we’re not judging based on the answers to these questions; it’s all about the scores we get.
Certainly, the score obtained should be evaluated relative to benchmarks such as competitor scores or industry standards. For instance, a specialized product like Excel would receive different ratings compared to a widely used product like Google’s search engine. Excel might not be expected to achieve a score as high as 90, whereas even a score below 70 might be considered unacceptable for Google’s search engine.
Here is a report on a product I’ve been working on. What other methods do you know? If you know any other practical methods, please mention them in the comments so we can discuss.
