Customer experience is often presented as an aggregate number across the entire nation. But customer experience is a local phenomenon and shoring up issues in specific areas and for specific towers can increase satisfaction, reduce churn, and improve each customer's experience.

Improving customer experience, tower by tower

While customer experience is measured on a nationwide level, customers actually experience your network on a tower by tower basis. CARTO and Google Cloud allow you to see this data in detail.

This demo will show you how to identify areas that need improving, understand performance data using Generative AI, and see issues at a tower level across all core performance metrics.

MAP 1 OF 3

Understanding coverage nationwide

This map allows us to see our coverage across the country in different categories, here styled by download speed performance.

Hover over a hexagon (which aggregates all the towers beneath it) to see the performance cateogories and competitors for that specific cell.

Move the map in to see a more granular view in different areas. Interact with the map using the widgets at the left.

MAP 2 OF 3

Unusual use patterns

Finding and detecting issues at a tower level is a top priority among many telco providers. However, identifying towers that have performance far outside the norm can be difficult.

However for providers that have invested in collecting usage data at a tower level, they can now pinpoint towers with performance issues at various periods of time and find users who are misuing the network.

Toggle the "Warning Category" widget to find towers that have service irregularities and click on them for more details.

MAP 3 OF 3

Usable results with generative AI

Of course, making these results readable can be a challenge as well. Ensuring that users internally can decipher the in depth data stored in each locaiton is key.

Using Google's Gemini AI we can label the results without leaving BigQuery and update them as data is made available.

Click on a cell to see the AI summaries of the data in that cell along with potential recommendations in that area.

Tools used: CARTO Builder

Orchestration at scale

Everything you have seen today is orchestrated with CARTO Workflows, a low code interface to Google BigQuery.

This allows you to construct complex spatial and non-spatial data pipelines to analyze your data at scale and keep analytics up to date, even with no background in data engineering.

Thank you!

A story map by