Eliminative Research Should Be More Common and More Appreciated
It's time to step off the feature-multiplying treadmill and spend more time on streamlining
In healthy economic environments, companies like to hire more, launch more, and expand more. This pressure can create a treadmill effect, pushing a growing number of researchers to look for a growing number of opportunities to build new products. Success is often seen through the lens of feature launches rather than a more sustainable indicator of company health or performance.
This can all be pretty exciting but tragedy hits when that treadmill suddenly freezes. Employees who were running at full speed fly head first over the front of the machine in mass layoffs, and leaders realize a sudden need for greater streamlining and efficiency. Across tech today, this new pressure for cost-cutting and downsizing is more prevalent than ever.
Researchers are a big part of the solution to this problem, but the focus can no longer be dominated by a drive for more product. Instead, there needs to be equal or greater focus on identifying opportunities to find and eliminate product waste. The truth is there should always be a healthy balance of product growth and product pruning, but since it’s hyper-relevant now, here’s my deep dive into the psychological bias against subtraction and the value of product efficiency research.
The bias against subtraction
A study published in Nature back in 2021 showed that people have a peculiarly unbalanced way of solving problems. When people have a task to complete or challenge to work through, their minds naturally drift toward looking for something to add to the situation to get them closer to a solution. This remains true even when the ideal solution to that problem is a glaringly obvious subtraction from the situation.
In one example from the study, a university president asked stakeholders for formal suggestions of how to improve their institution. Out of 1000+ submitted recommendations, just 11% were subtractive strategies (e.g. eliminating legacy admissions) and 89% were additive strategies (e.g. more study grants). A mix of other experimental and observational studies found a very consistent effect: people generally gravitate toward fixing a problem by adding rather than removing.
This bias is alive and well in tech too. When brainstorming around a new problem, most ideas tend to focus on “building X” or “expanding Y”. It’s relatively rare that people suggest “streamlining X” or “eliminating Y”. Whenever those eliminations do happen, they often happen too late and are frequently unpopular decisions.
It’s perhaps not surprising that elimination decisions are less popular than launch decisions. Elimination makes people feel as though their efforts were wasted or potentially even not good enough, while launch decisions are a celebratory endpoint to hard work that people look forward to. Regardless of how good they each feel though, there’s a time and a place for both, and opportunities for elimination are relatively neglected.
Dedicating more time to eliminative research
Researchers are well-placed to identify opportunities for product elimination. Since they spend significant time on understanding user perception and evaluating product performance, they can identify product experiences that fall short of expectations after launch.
In some of my past work, this kind of research has helped to improve product efficiency in three key ways:
Identifying product risks and negative user perceptions before formal launch to avoid expanding a product without clear value.
Finding product features or interactions that are redundant or conflicting next to other existing features.
Identifying insufficient ROI on a product that is costly to maintain.
In all three cases, elimination is the outcome and it’s a profitable one. You end up with a simpler product experience with greater “joy density”. Instead of having a mix of many features—some good and some not so good—you end up with fewer features, all of which are meeting a high bar for excellent user experiences. Those kinds of products are always deeply pleasant to use because they have little friction and minimal opportunity for confusion within them. The joyful experience you’re trying to create isn’t diluted by a crowd of less cool features.
High joy density is what you can achieve with eliminative research. To be clear, “quality” isn’t judged along a single dimension as my simplified drawings above would suggest, and there may be many important features that are relatively joyless but important for the integrity or profitability or basic functionality of a product. I’m not recommending that you eliminate those. The point is a general one around bloat, redundancy, and excess in product development, which is much too common. As researchers, I think we have an opportunity to be stricter and more principled about what we recommend to be net positive experiences for users, and what we can afford to set aside for the benefit of creating a more streamlined product experience.
Top takeaways
Try to be conscious about giving equal weight to subtractive vs additive solutions in product development, research, and general problem-solving in life!
Don’t be afraid to recommend killing a concept or feature when it feels like the right thing to do. Reactions of disappointment in the short-term regularly evolve into feelings of positivity with time.
“Our wasted oil unprofitably burns,
Like hidden lamps in old sepulchral urns.”
~ William Cowper
If you enjoyed this newsletter, please spread the word by sharing it with others who might find value in it. If you’re new here, you can subscribe below.