The data of design division

Most design behavior is to solve the problem. How to determine whether the problem has been solved or not? If the impact of your design is quantifiable, then there is a relatively objective way of evaluating it: look at the data.

How to formulate data indicators

How to effectively use data validation design is an issue that has plagued me forever. My main doubts are:

• Before designing, how do you know what data you need to monitor to aid in verifying the design results?
• How to judge whether a design is good or bad through data?

An article by Google user experience researcher Kerry Rodden inspired me a lot. At Google, data metrics are determined in two steps: “user experience quality” and “product goal.” “User experience quality” simply means “what aspects of the design do you want to observe?” and “product target” is based on the aspects you want to observe. Data indicators are determined in the order of “target → sign → index”. . According to this method to formulate data indicators, the thinking is clear and the overall process will be such a matrix:

The left column of the table represents the user experience quality dimension to be considered. Kerry calls it the “HEART Framework”. It does not need to be fully complied with in practice, and can be focused or increased or decreased according to the needs of the project.

The three horizontal terms in the table are worth explaining:

Goal

Simply put, what kind of results do you hope to achieve in the design after going online? For example, Youtube’s search function, the key goal in Task Success is: the user finds the most relevant video or channel faster when searching.

Signs

The goal is to determine, what signal that the design has reached or did not achieve the goal? For example, the sign of Youtube in Engagement is the number of users watching videos on Youtube and the time it takes. Another example of the failure of the search function in terms of Task Success is that the user did not click on any search results after searching.

There may be many signs that can be used to determine whether or not to achieve the goal. At this time, the trade-offs must be made in light of the actual situation. For example, is it easy to trace this sign? Can it observe obvious changes as your design changes?

Metric

The indicator is more grounded than the logo. It is very close to the raw data we have obtained. For example, the “Time when users watch videos on Youtube” indicator is represented by the indicator “Per-minute video minutes per person per day.”

Through the “Target → Signs → Indicators” process, combined with the HEART framework, you can clearly know what aspects of the design you want to validate, and what data you need to focus on to achieve your goals.

To sum up

The importance of data in design is self-evident. It allows designers to have a more general understanding of the product to the user, and allows us to come up with more objective reasons to prove why our design is good.

If you don’t know how to formulate data metrics, Google’s approach may be useful to you. According to the principle of “Target → Signs → Indicators”, combined with the HEART framework, it is possible to determine which aspects of the design should be validated and which specific data to focus on.

Beware of pitfalls in the data. Sometimes it will bias us. Care must be taken to distinguish between “observed phenomena” and “inferred conclusions.” From this perspective, design is also a subject that requires Critical Thinking. In addition, data should not be a fundamental purpose at all times. If you want to understand what your product’s core goals and values ​​are, it’s easy to take the wrong place.

Finally, as a means of aiding design, data also has its limitations. It can tell you what happened, but it does not explain to you why it happened. The designer needs to find the problem with other means.