The Scenario
You are an e-commerce product manager. Your site's search results page has twelve filter attributes in the facet rail. They have not changed since the site launched three years ago.
Your hypothesis: customers use three of them constantly and ignore the rest. If you can prove it with data, you can promote the top filters to the primary rail and move the rest to an "Advanced Filters" expandable, simplifying the UI without removing functionality.
The data is in Algolia analytics. You need it in an Excel workbook for the product review meeting on Wednesday.
The bad version of Monday:
- You look for filter analytics in the Algolia dashboard and cannot find them in the main navigation
- You find a reference in the documentation pointing to the analytics API
- You ask the data team for a pull
- The data team has a three-day queue
- Wednesday arrives and you present the hypothesis without supporting data
The fast version is one prompt, Monday afternoon.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your Excel workbook that calls the Algolia filter analytics API and writes the results into the workbook, no data team ticket required.
Open the SheetXAI sidebar and type:
Fetch the top filter attributes used in searches on my Algolia index 'products' in the last 30 days. Write attribute name to column A and usage count to column B of the Filter Analytics tab. Sort by usage count descending.
SheetXAI calls the Algolia filter analytics API and writes the attribute usage ranking into the Filter Analytics tab.
What You Get
A filter usage ranking with:
- Column A — filter attribute name as configured in Algolia
- Column B — how many times that filter was applied in the last 30 days
- Sort order — highest usage first
The usage count is the argument. If "brand" has 180,000 filter applications and "material" has 4,000, the case for promoting brand and demoting material is already made before the meeting starts.
What If the Data Is Not Quite Ready
Filter analytics are clean structurally, but the product decision usually requires more slicing.
When you want value-level usage for a specific attribute
Knowing "brand" is the top filter is useful. Knowing which brands customers filter by most is more useful.
For the 'brand' attribute in my Algolia index 'products', pull the top 20 filter values customers selected in the last 30 days. Write value name to column A and selection count to column B of the Brand Filter tab. Sort by count descending.
When you want to compare mobile versus desktop filter usage
Your site has different filter UIs on mobile and desktop. You want to know if usage patterns differ by device.
Fetch filter attribute usage from my Algolia index 'products' for the last 30 days, broken down by device type if available. Write attribute name, desktop usage count, and mobile usage count to columns A, B, and C of the Filter Analytics tab. Highlight rows where mobile and desktop rank the same attribute differently in the top three.
When you want the share of searches that used each filter
Percentages are easier to present than raw counts.
Fetch filter attribute usage from my Algolia index 'products' for the last 30 days. Write attribute name and usage count to columns A and B. Then fetch total search count for the same period and write the percentage of searches that used each filter to column C.
When you want to pair filter usage with conversion rates to make the business case
Some frequently used filters drive purchases. Some do not. You want the data to show which filters are worth promoting.
Fetch filter attribute usage and associated conversion rate from my Algolia index 'products' for the last 30 days. Write attribute name, usage count, and conversion rate for searches that used that filter to columns A, B, and C of the Filter Analytics tab. Sort by conversion rate descending. Add a column D that ranks each filter by the product of usage count and conversion rate as a proxy for total business impact.
The pattern: start with the ranking, drill into the attributes that matter, and pair usage with conversion to make a data-backed case for the UI change.
Try It
Get the 7-day free trial of SheetXAI and open a blank workbook, then ask it to pull your Algolia filter analytics. The Algolia integration is included in every SheetXAI plan. For related workflows, see how to pull zero-results query data from Algolia to Excel or the Algolia in Excel overview.