[EmpathyX] [UX] Next Queries

Given an initial query, the Next Queries service provides a new set of queries that are related to the initial one. These queries are inferred from patterns in other user sessions (a.k.a. collaborative filtering) and provide a mechanism to anticipate what users might look for in their next query.

The API is very easy to integrate as Next Queries are global and only depend on the query and language. For instance, if we wanted to get the Next Queries for "eggs" in our UK store, we could do the following request:


https://api.empathybroker.com/search/v1/query/{instance}/nextqueries?q=eggs&lang=en_UK

And voilà, the endpoint would return something like this:

[
  { "query": "bacon" },
  { "query": "bread" },
  { "query": "ham" },
  { "query": "milk" },
  { "query": "butter" }
]

But... how can we use that information?

Next Queries can be implemented in multiple ways, but the general guideline is to use them only in spaces where the user is likely to perform a new query. Let's take a look into three of those situations:

1. The user goes back to the search box

This is the most straight-forward situation, since the user is actively trying to perform a new query.

In this case, one typical option is using the predictive layer to display Next Queries as suggestions:

Going back to the search box in Massimo Dutti

2. The user adds a product to the cart

When this event happens, we can safely assume the user is going to look for something else. Even if that is not the case and the user is planning to proceed to checkout (or leave), Next Queries are a nice feature for cross-selling and inspiring the user with ideas that were out of their initial intention.

Some options for this scenario are presenting Next Queries next to the product, or even on top of the results grid, as we can see in Ocado after adding bread to the cart:

Adding bread to the cart in Ocado

3. The user scrolls down a number of rows (or paginates) without clicking on any product

That succession of events is an indicator that the user is struggling to find a result that satisfies their need. In that case, it is generally a good idea to display query refinement options (such as Related Tags) or alternatives to guide the user into new, related sections of the catalog. Next Queries can help to tackle the second scenario in different ways.

One possible implementation is embedding a block of results for one or more Next Queries in the results grid. I personally like to display each query as a carousel, especially on mobile devices, to avoid consuming much vertical space. Pull&Bear is one of the stores that implements this option, after scrolling down 12 rows.

Scrolling down 12 rows in Pull&Bear

Another alternative is displaying Next Queries as labels without products, but still within the results grid, as we can see in Google Images' Related Searches section:

Scrolling down a few results in Google Images


Conclusion

Next Queries are a great tool to help your users in situations where they are likely to perform a new query. Besides acting as a shopping assistant, they can help users to discover new products and generate cross-selling opportunities.

If you are considering to use the Next Queries service, drop us a line and we will be happy to enable the endpoint for your instance.