Almost everything we do in our lives requires making a decision. What should I wear today? What should I eat for dinner? Which set of headphones should I buy? Which account tier do I really need for this new service? In today’s world, there is often an abundance of choice. When faced with many alternatives, how do people avoid choice overload and decide which option to pick?
For example, there are hundreds of bathroom vanities available at Lowes — far too many options to consider individually! To quickly narrow the selection, most online shoppers first use the filter options to select criteria that are nonnegotiable: for instance, the dimensions of the space available and the number of sinks. Depending on the number of results remaining, they may continue adding additional filters for preferences such as color finishes. Then, once the result list is reduced to a feasible size, they browse through it and select a few promising alternatives to compare.
This process illustrates the two decision-making strategies that people use:
- A noncompensatory decision-making strategy eliminates alternatives that do not meet a particular criterion.
- A compensatory decision-making strategy weighs the positive and negative attributes of the considered alternatives and allows for positive attributes to compensate for the negative ones.
Noncompensatory decision making is used when there are many alternatives to choose from; it allows people to quickly narrow down the number of options to one or a few. Compensatory strategies put a load on the working memory and, because of that, can be used only when there are few alternatives.
Understanding why and how people use these two decision-making strategies allows us to design interfaces that best support users.
When People Use Each Method
The number of options is the primary factor that determines what strategy people will use. That’s because this number determines the amount of effort necessary to sift through them.
When we evaluate just a handful of alternatives (around 5–7), comparing the attributes of each is a feasible task. By thoroughly reviewing these attributes, we can determine whether certain aspects are more important to us than others, and then allow the positive values to outweigh negative ones when assessing each alternative — this is the compensatory strategy.
However, when there are many options, it would be daunting — if not completely unreasonable — to comprehensively compare the pros and cons of each alternative. In these situations, we turn to the noncompensatory strategy of eliminating any alternative that does not meet some key criterion. This approach allows us to narrow down the set of options quickly and easily, at the expense of not fully considering each.
Filters Support Noncompensatory Strategies
When people use a noncompensatory strategy, they select one or more nonnegotiable key characteristics and eliminate all the alternatives that don’t have them. This procedure is repeated until the set of options is narrowed down to a manageable number.
Filters and faceted navigation on product-listing pages are the most common tools that support this noncompensatory criteria selection, allowing users to quickly eliminate options that don’t possess those chosen traits. For example, Wayfair.com provided many filter options for various sofa characteristics, allowing users to narrow down a list of 20,903 sofas to view only those that meet certain qualifications, such as size, seating capacity, and price.
For sites with large inventories, filters are critical. Without them, users can easily feel overwhelmed by all the choices and will often abandon the site. In addition, filters are so commonplace that users expect them on every site.
For instance, a user shopping for a mirror on the Crate & Barrel mobile app was disappointed that she could not filter out mirrors that did not have a gold finish: “I was looking specifically for a gold, which I did see on top (of the listing page), but I was just checking to see if there were other cool ones. So, I’m seeing there’s 41 of them, but there’s no way to filter it. … There’s got to be a way to filter them. But there isn’t, which is odd. Usually, you know, you can, so I’m a bit bothered.”
Of course, simply having any filter isn’t enough — the filter attributes must be relevant to users and match the characteristics that they care the most about. Presenting a filter that only includes one or a few irrelevant attributes is almost more maddening than not presenting a filter at all. For example, one user shopping for a rug on the Interior Define mobile website became frustrated that the only available filter attribute was Collections. She stated, “I wonder if you can narrow down what kinds of rugs I’m interested in. So, I see the filters up here […] I’m not familiar with the names of collections, so that doesn’t help me with that. […] Where would you be able to do that? Because all I see are collections when I do filter. Usually you’d be able to filter by size, by color, by feature.”
To determine what filters are likely relevant to your users, check your site’s search logs to see what criteria users are searching for, talk to brick-and-mortar salespeople (if you have them) and customer support representatives to hear what users are most concerned about, or listen for what criteria users bring up during usability testing.
In addition to using filters when browsing alternatives, another noncompensatory tactic users may employ is searching using a multiword query. In our past ecommerce research, in-site searches contained an average of 2.3 words per query, which included characteristics such as color, size, and brand. Users expected these characteristics to act like filters and were confused and disappointed when sites didn’t prioritize items matching every word in the result set. For instance, a user searching a site for “wooden box” would expect to receive a list including only boxes that were made of wood — not all items on the site that are wooden plus all items that are boxes.
Compensatory Decisions Need Comparison Tools
When evaluating a small number of alternatives, thoroughly considering each option and its various pros and cons is a manageable task. UI tools that allow users to see and compare multiple items and their individual attributes on the same page support compensatory decision making.
Well-designed comparison tables break down the characteristics of each alternative, allowing users to compare the merits of each option. Because comparing multiple items is a cognitively demanding process, these tables must be designed to support easy scanning: align each item and its attributes into consistent columns and rows, avoid lengthy text within table cells, and ensure the included attributes are meaningful (and available for each item in the table). Ecommerce sites with just a handful of items per category or services with varying pricing tiers for multiple account levels benefit from displaying easily findable comparison charts to help users choose between the few options.
Compare tools that allow users to select a few products and directly compare them are also helpful. Remember that there is no need to support more than about 5 products in each of these tools — since people will not use noncompensatory strategies with a higher number of alternatives.
During a recent usability study, for example, one user was shopping for a refrigerator on Home Depot’s mobile app and wanted to determine why similar-looking fridges varied in price. If one option was more expensive, was it worth that higher price? To uncover these feature differences, she added 2 models from the product-listing page to the app’s comparison tool. While looking at the resulting table, she stated, “I like the side-by-side comparison because […], if something costs a little more money, I can see what the tradeoff is and if that is something that is of value to me to pay the $200 or $130 more.”
Surfacing key details about each alternative is critical to supporting compensatory decisions, so people can consider each individual attribute. Even when evaluating a single item, people must see both the pros and cons to determine whether there are enough benefits to outweigh any negative aspects.
Many Decisions Combine Both Strategies
These opposing decision-making strategies are not conflicting — rather, people often employ each of the strategies when moving through the stages of whittling down the potential choices and evaluating options. As illustrated in the earlier Lowes example, commonly people first narrow down a large set of alternatives using a noncompensatory strategy such as filtering, then, once they are left with a smaller number of results, they compare the negative and positive attributes of individual items.
Using each decision-making strategy, in turn, allows people to efficiently reduce a large selection of alternatives to just a handful, and then thoroughly compare the positives and negatives of these few choices to make the optimal final decision.
Conclusion
Understanding how people make choices allows us to design tools that support users’ decision-making strategies and increase usability. Providing some way to filter out options that don’t possess selected characteristics is critical for noncompensatory decision making, while surfacing key traits and offering comparison tools supports compensatory strategies.
Learn more about how people think and problem-solve in our training course on The Human Mind and Usability.
Share this article: