“But I still haven’t found what I’m looking for.”
This classic lyric from Bono and U2 hails from their 1987 song of the same name. Unfortunately, this introspection often typifies many users’ experience when using the search functionality on a site or platform. Ask yourself. How many times have I been on a website and was unable to find (discover) what I was looking for using search? How often at work was I unable to locate a specific document?
What did you do? How did you begin your search? What did you type in as search terms? How do we decode all of this?
First, let’s spend a minute defining “search” and what it encompasses. Search is the practice of identifying and indexing specific content that can then be discovered and displayed to authorized (i.e., entitled) users. Put another way, “[search) is the art and science of making content easy to find.”[1] The word content is key here, as it refers to all types of objects, from videos to people to products to workflows and actions, etc.
Now that we have a working definition, how can we measure the impact of search? The main consideration across the industry is speed/performance. For example, how long does it take for the results to appear? Amazon calculated that a page load slowdown of just one (1) second could potentially cost it $1.6B in annual sales. Google calculated that a search results delay of just four tenths (0.4) of a second could result in a loss of up to 8 million searches per day – along with the associated ad revenue.[2]
Beyond product and Google searches, let’s also consider the impact to employees using enterprise search to locate critical information for their role in an organization. One study from Adobe “…shows that [employees] waste a significant amount of time each week dealing with a variety of challenges related to finding and working with documents. This wasted time costs organizations ~$20K per worker per year and amounts to a loss of 21.3% in the organization’s total productivity. For an organization with 1,000 people, addressing these time wasters would be tantamount to hiring 213 new employees.”[3]
As you can see, search is kind of important.
The impact of the search experience touches not only monetary aspects for an organization, but also a user’s trust, loyalty, and advocacy for that particular product/platform (known as the Loyalty Ladder). Every interaction or transaction by a user with a product/platform moves that user up or down the Loyalty Ladder based on the overall experience. When users have a bad search experience, that transaction moves them down the Loyalty Ladder and further away from brand advocacy. In contrast, when users find exactly what they are looking for on the first try, they move up the ladder.
We have looked at search and its impacts across a variety of factors. Now let’s look at users’ search behaviors when they get to a new platform. Fifty percent of visitors to a new page on a website go straight to the internal search box in order to navigate. After viewing their search results, only 15% of people continue to use search following their initial query. So why is that number (15%) so low?
A few factors contribute to this, both positive and negative. First, the search may have been 100% successful and directed the user right to where he needed to go. Taking this one step further, based on how search results are formatted for a platform, the search results may turn out to be the user’s entry point to begin browsing to find exactly what the user wants.
For example, the user searches on “Sony 4K” on an e-commerce site and is brought to the 4K TVs category with Sony selected in “Brand” filters. From there the user can easily browse all of the Sony 4K TVs without the need to search again.
There are certainly contributors for unsuccessful searches, with the largest contributor being this: we (users) are terrible at searching. That revelation may sting a little bit, but let’s take a look at why this statement is true.
- Search requires domain knowledge – While most users consider themselves pretty savvy with search, many times the culprit behind a failed search is the user’s lack of domain knowledge, which forces the user to perform multiple searches or abandon the site altogether. For example, a user shopping for a new 4K television probably knows some basic features (attributes) that they want, but is (likely) not completely knowledgeable about all the different brands and options available.
- Higher (perceived) interaction cost than navigation/browse – Users have a perception that the interaction cost of performing a search, which may or may not be successful, is greater than browsing/navigation through a platform. In essence, users believe that a “failed” search wastes more of their time than clicking on categories and filters to discover their object. This stems from a “boolean” mindset that directs users to start all over if their search fails to hit the mark. For well-designed experiences, this becomes less of an issue when objects (e.g., categories) are included in the search results.
- Users don’t know how to form a search query – Put simply, users don’t speak in terms or style that are conducive to a search query. Instead, they speak in natural language so many times their query is exactly what they would say to another user. Most users don’t think or know how a SQL-like query is formed or understand the best way to convey search terms for a given platform. So they often just start typing terms that they believe relate to the object they are looking for, resulting in a nonsensical query. Inevitably, no matter what the search query/terms are, some form of transformation will be required to change that query into something the platform can act on.
Now, knowing some of the challenges and opportunities, how does an organization begin to design and evaluate their search experience? In Part 2 of this blog, we will direct our attention to three main areas of focus when considering the search for a platform: technology, experience, and data. Stay tuned!
[1] https://www.aiim.org/What-is-Enterprise-Search#
[2] http://www.fastcompany.com/1825005/how-one-second-could-cost-amazon-16-billion-sales.
