5 min read

Why Designing for Data Matters And How To Go About It

Data driven design is a known and commonly applied concept in UX design. Data driven design refers to design that is backed by findings from data collected by users. 

But data driven design can often miss out on collecting information about the context, motivations and intent behind the decisions users make. In order to collect better data, we need to change the way we collect it. And we can do this by designing to collect better data. 

What is data-driven design?

Creating a user-centred product involves doing research about what the users’ needs and expectations are. To minimize the chances that your product will fail, the more research that you do, the better. Many products fail because designers project their own beliefs and assumptions about what the users want. To avoid such errors, they first collect data to back up the choices they make in design.

Data-driven design has to be incorporated at every stage of the design process and is iterative, meaning it can change on the basis of feedback or new learnings. Both qualitative and quantitative data is collected through various means when researching the users through different tools like A/B testing, surveys, user flow analysis etc. The aim is for designers to make the best decision they can based on as much insightful data as possible. 

Why is data-driven design not enough?

why is data driven design not enough

While data-driven design can tell us who our users are and what their expectations from the product are, it can often miss out on reporting the nuances that can help inform the experience design. Collected data is often missing context and motivations of the users. A good product not only brings together data and design, it also blends together empathy and deep insights about user behaviour. Adding these elements can push UX design beyond the ordinary and into the extraordinary and really set a product apart. It can help to create a truly compelling and intuitive experience for the user.

Adding context to personalized recommendations

Data collected through simple algorithms, like the kind of data collected by Netflix or Spotify based on your usage history, lacks one very important factor. This data is completely without context. This means that it paints an incomplete and disjointed picture about your true experience with the service and makes assumptions that might fail to engage you the next time it gives you a suggestion.

For example, say you played dance music for your friends because you were throwing a party. Spotify might assume that this is the kind of music you like to listen to regularly when you are, say at work, and might suggest playlists of dance music to you instead of the more ambient music you usually prefer. Its suggestions will most probably be a miss for you and you will fail to engage with the app and press play. 

Similarly on YouTube or Netflix, you might start watching a video because it was recommended to you by a friend, but decide for a few minutes that it is not for you. The algorithm though does not understand this context and will keep recommending the show and other shows similar to it for you to watch. 

What is missing in the above common experiences? Nuance. Most data collected through user interactions is missing any context and nuance. There needs to be a better design that can collect more correct data based on true user interactions.

Designing for useful data insights

Design today needs to capture data not only about the “what”, but also the “why”. The idea that the more data there is, the better the design will be is not true anymore. Without nuance, a lot of data will be useless when designing iterations for your product. It is also important during design to know what data is important and what data should be ignored.

Rather than making data-driven design decisions, designers need to make data-informed decisions. This means that along with data points, they need to evaluate user motivations, expectations and emotions when they are interacting with their product. 

This data has to go beyond time and space and ask questions like “what is the user motivated by? Competition and rewards? Or messages from friends and family? Do they respond to messages during work or at night after their day is done? Do they travel a lot or mostly stay in the same place? Do they prefer to use a service on their desktop, phone or wearable?”

Proactive design to capture user behaviour 

While data-driven design can be reactive, designing for data is proactive.

How do you usually collect user information for research? Chances are you collect data through surveys, heat maps and analytics. These collect data about users while or after they have interacted with your product. 

Designing for data takes the approach of collecting data about user behaviours and actions that are guided by the design of the product itself. This will capture data with intent and lets you assess the answers to focused questions like how a user will change his or her behaviour depending on changes to the design of the app.

The following steps will help you to design for data at your own organisation

Step 1: Collect intent-driven data

As explained above, the data collected about the user’s behaviours and experiences has to be more nuanced and insightful than what is usually collected when doing research. It needs to capture the user’s motivations, emotions and expectations along with needs and pain points. 

Step 2: Make decisions based on this data

make decisions based on data

The next step is to use this data to inform the design decisions that have to be made. Only data that can be used is useful data. Use tools to analyse all the collected data to generate insights and then use the insights that will help you to solve problems and design products that will make your customers have a better experience.

Step 3: Use insights to inform design

use insights to inform design

The next step involves using any insights gathered so far to design the product in such a way that it can collect even more data to add more detail and nuance to the insight. For example, if you have learned that most users are not finding the FAQ section of your website useful, then design the FAQ section in such a way that collects information about why they are finding it difficult to get answers. This design will be data-driven and will also be designing for data. The product will benefit greatly from more insight and intuitive designing and this in turn will set you apart from your competitors.

Step 4: Repeat this design loop

Conclusion

While the data-driven design is important, it is more important to design with useful data that captures the context and emotions along with the user behaviours. The reasons behind the choices are as important as the choice itself. Incorporating these insights into the data-driven design will take it to the next level and the best way to capture this additional data is through design itself! In this way, we can design for data and the design loop will continue.

Published by Rashika Gautam

Digital Marketing Manager

Rashika Gautam Digital Marketing Manager