Our 5 biggest UX-design mistakes that you should avoid

Firstly, I have to admit that at the beginning of Stackfield none of our team members really had the experience of SaaS tools’ design and their usability design. For all of us, this was a completely new territory, but we pushed through continuously and learned a lot in the last 3.5 years. So that you do not make the same mistakes we did, you can find below a list with the 5 biggest design mistakes that we made:

1. Mistake: Too short or too long registration on the home page

The registration process for a SaaS platform is one of the toughest design challenges. The right placement of the registration form is critical for the user growth. There is a general consensus that the registration should request as little information as possible. We followed this exact advice from the very beginning; however, the conversion rate was not as we hoped for. After surveying several test users, the conclusion was clear: the mere request of the email address gave little confidence in the platform, as it did not transmit the clear message that an account will be directly created. The security of the encryption was also questioned in this context.

The request for less information gives also the impression that the data will be used for newsletters or events, which are considered a suspect/spam behavior. During our tests, we have tried various combinations that queried for additional information. From this exercise we have found, for example, that the request of a phone number reduces the conversion rate dramatically.

Therefore, you should test different options, but you should also keep in mind that the retrieval of information must always be proportionate to your platform. Additionally, the registration process will affectthe subsequent on-boarding processes, which will most likely be extended if little information is requested at the beginning. Approaching users in the follow-up emails will also be more difficult if you did not query the name of the user. Always keep these points in mind! Our various tests have led to deviations of the conversion rate of up to 52% with the best results obtained from querying the user’s name, email address and password. Therefore, it is important to test several options to see what works for your tool/site! Likewise, the registration form should be directly visible on the home page, rather than being hidden through a button. The direct display of the registration form on our home page improved the registration rate by 34%.

2. Mistake: Forcing users to learn

Any new platform requires a certain adaptation phase for all users. We knew this from the beginning, so we established a sequence of milestones based on the actions, which increased the probability for users to stay as long term users. In order to assist the users with getting started and moving faster through these learning milestones, we used a wizard, which permanently displayed certain objectives. The objectives included, among others, inviting a person to the platform, creating a new data room, etc.

After a short time, there has been a large amount of complaints about the lacking ability to remove the wizard, which led us to eliminate the feature. The user was supposed to go through a series of steps, however, we quickly realized that the wizard led to frustration and resentment, as it continuously displayed all the things a users hasn't done yet.

Similarly, we found that any tutorial, which could be perceived as intrusive, is ignored. Even pulsating circles that included one tool tip were simply clicked away and displayed only for 1.3 seconds in average. In conclusion, it is almost the most important thing to design the platform as intuitively as possible. Each user needs to know immediately how he must act in order to utilize the platform – without being prompted with intrusive tips.

3. Mistake: Too much flexibility

This point suggests that this is not a UX design mistake. But this is precisely the case. When we kicked off the beta phase of Stackfield, we hoped that a variety of features would bring a competitive advantage over our main competitors. Everything was supposed to be possible and completely flexible, so that any user could turn Stackfield into his own project management tool. After some time, it became clear that - although this all sounds good - a large share of our users could simply not find the right way to use Stackfield.

The solution for that problem is really simple, quite literally. No matter what type of platform it is, all functions are useless as long as the user is overwhelmed by their quantity, whereby he has no idea what to do. Thus, limit yourself to primarily designing the core functions that can be shown to the user on a clear path. All other functions, which are considered optional, can be integrated a click away. The user should have a clear understanding of all core functions from the beginning, therefore remove all unnecessary elements.

We also received an incredible number of requests for the development of certain features, which weren't available to that time. While we added some, we had to reject the most of the requests. Why? Because these features, although requested, were simply not used in practice – for example, sorting the workroom list by date and name, content hiding, personal spaces and many more. Although it sounds paradoxical, the majority of the users are not aware that these functions are not necessary, but they assume them to be.

4. Mistake: Different PNGs for each display resolution

The trend of high-resolution displays evolved along with the development of Stackfield. Since the first version of Stackfield was designed only for the normal resolution, the display on devices with Retina Display was of substandard quality. We initially tried to solve this issue through the use of PNGs in different sizes. This solution proved to very compley and highly unpractical over the long term, since all graphics were deposited in three different sizes so that they remain clear at higher resolutions, and quick when used on devices with low resolution screens.

It is advisable to use SVGs instead of PNGs. These vector-based graphics are displayed neatly on any device (provided that they have been optimized for the lowest resolution) and additionally keep you away from countless PNG files. SVGs are not compatible with any old browser (yes IE8, we are looking at you!), but our encryption is only supported by modern browsers, too. Therefore, the problem was insignificant in our case.

5. Mistake: Too much information

Besides the mistake of integrating too many features, we have also encountered another problem that persisted for a long time – the intention to present all the information at once. There are definitely advantages by displaying the texts in font size 12 as it permits the presentation of all a lot of information at a glance, but it also brings many disadvantages: Too much text makes it difficult to focus on relevant content; displaying all available filter option may add more content than the filtered data itself; not all the profile pictures of all users must be presented within each feature – it becomes annoying at some point.

Again: less is more. Use the entire screen, but display only the most relevant information. Are you considering font size 12 for a project? Everything under font size 14 is obsolete! (Sounds trivial, but it is frequently encountered …) Focus on the important information, and don't do it as we have done it: Four editable columns with content have simply led to confusion – we have now arrived at 2 columns! Try to find a balance between function and design – if only one of these is at its best, no user will be convinced!

We have certainly made plenty of other errors in the past and may even make some more in the future. However, these 5 mistakes have stuck with us over the years in regards to design mistakes. What mistakes have you made with your team so far? We welcome your contributions in the comments section!