The Myth of Consistency

Consistency in UX is never questioned. It is seen as a cornerstone of UX design. It is sacrosanct. Incontrovertible. A first principle, never to be questioned—never to be compromised. However, I have long held that consistency is not human-centric; it is business-centric.

You have likely all heard the story of the Five Monkeys and the Ladder:

The story goes, that five monkeys were placed in a room with a bunch of bananas on top of a ladder. Spotting the bananas, one monkey jumped on the ladder and immediately was sprayed with cold water along with all the other monkeys. The first monkey jumped off the ladder and joined the other cold, wet, and now very confused monkeys.

But the bananas were too tempting so a second monkey started to climb the ladder—with the same results. When a third monkey tried for the bananas, the other monkeys blocked the ladder before any of them got sprayed.

Then one at a time, each of the five original monkeys were removed and a new monkey was brought as a replacement. As each new monkey spotted the bananas and naively started for the ladder, all the other monkeys promptly jumped into place and blocked all access to the ladder.

By the end of the experiment, none of the original monkeys were left, and even though none of the new monkeys had ever been sprayed with the cold water, they all shared the same belief; never go for the bananas.

This is not presented as any kind of actual scientific experiment but as an allegory about people following rules they don’t understand. There are many examples in the UX community where there are strongly held beliefs which are not understood but are still followed without question. For me the unwavering belief in consistency is one such example.

The original goal behind consistency was to ensure the person’s interaction with a product aligned with their expectations. Or to be more specific, the plan the person mentally formulated to accomplish their objective with the product/tool—based on their interpretation of what was being presented to them, when put into action, did in fact, achieve their desired outcome. That is, there was consistency between people’s mental models and the actions required to successfully achieve their desired outcome. (It’s unfortunate that mental models have fallen out of fashion, indeed the science of human perception and cognition is barely touched upon in design education.) Regardless, a better terms for describing this alignment would have been predictability.

At some point the underlying science was replaced with generic guidelines. As a consequence what was originally meant by consistency has been replaced by a common dictionary definition: “the quality or fact of staying the same at different times” or “conformity in the application of something”. To be clear, that type of consistency was never the goal of human-centered design but it did quickly gain adoption by the business.

In order to reduce costs and time to market, business’ recognized the value of conformity, whether it was UI, content, or production, doing the same thing over and over and getting the same result means mass production, and that means profits. Indeed, more often than not, consistency is applied like Lego™️ bricks, ensuring everything fits neatly into place even if it has sharp edges. But it is important to recognize this type of consistency is purely from the creators’ point of view.

Predictability by contrast comes from a human-centric point of view. While it provides greater freedom to innovate, to create new models of interaction, it also allows you to tailor your UI, content and experiences to the specific groups and individuals.

It is true that in order make a products/services predictable you need to do more design research and ideally even co-creation with various constituents, not just your customers, stakeholders, and advocates. Its requires more active engagement with groups who are likely to be under represented in your segmentation analysis, people who are lumped together based on impersonal demographic or psychographic profiles/personas.

Predictability can increase inclusivity and equity allowing your products to meet your customers where they are, to present functionality and services on their terms. It can help everyone feel respected, supported and valued—both in terms of the product’s interaction and its content by removing barriers and providing inclusivity in ways unachievable with traditional consistency.

Also consistency doesn’t create trust, which, with the rise of generative AI, is quickly becoming the primary concern of use experience designers. Ensuring people know the consequences of their actions—sharing personal data, changing a setting, liking a post, is what builds, or degrades, trust. In short making the system predictable goes toward making user confident in the system, adding transparency, clarity, and a concrete policy are more important to trust than consistency.

Consistency is business-centric.
Predictability is human-centric.

As always, please feel free to share your thoughts.

(please note, all posts are moderated for appropriate content)