[ THE PROBLEM ]

No self-serve analytics. Every question took days.

PlayOn’s data environment spans multiple acquired companies, dozens of database systems, and wide tables exceeding 40 columns — a complexity that broke every AI agent they tried, including Snowflake Cortex and Secoda.

AI agent comparison

Two tools failed. One didn't.

Snowflake Cortex

Could not handle wide tables exceeding 40 columns or join across acquired company databases.

FAILED

S

Secoda

Broke on multi-source joins across the medallion architecture. Wrong answers killed trust.

FAILED

vs

TextQL Ana

• 7,000 datasets joined across bronze, silver, and gold tiers via ontology mapping.

• Board reporting automated with school-year corrections. Validated through playbooks before delivery.

VALIDATED

FIG. 01 — PlayOn evaluated Snowflake Cortex and Secoda before deploying TextQL Ana against their Snowflake medallion architecture.

Most tools require a little bit of complex SQL if you’re going to take multiple different data streams and predict against them. It requires a skill set that most business stakeholders don’t keep up to date. And realistically, you’re thinking of maybe a 2- or 3-day turnaround time for any analytics question.

One of the biggest challenges with rolling out AI is that you’re worried they’re not going to give you the right answer the first time. That can kill trust in a data organization and an initiative very quickly.

[ THE SOLUTION ]

Ana joins 7,000 datasets non-technical users never could

Ana succeeded where Snowflake’s native Cortex and Secoda both failed: handling complex, multi-source data accurately under real-time pressure.

PlayOn deployed Ana against their Snowflake data warehouse to automate weekly Friday board updates — tracking streaming bookings, active subscribers, ticket sales, and year-over-year growth with school-year corrections built in. Executives now get intraday snapshots via Slack without waiting for a human to pull the numbers.

PlayOn’s Snowflake warehouse follows a medallion architecture — bronze, silver, and gold layers — spanning data from multiple acquired companies. Ana’s ontology maps the relationships across all of them, letting business stakeholders ask questions in plain English that previously required a data engineer and days of turnaround.

• User Engagement Analytics

  User engagement patterns, MAU/DAU tracking, feature adoption, click analysis, user population segmentation.

• Business Metrics

  Subscription analysis, ticket sales, payment methods, donation tracking, pricing optimization, churn investigation.

• Data Validation

  Venue/team data mapping, ID reconciliation across platforms (VNN, MaxPreps, GoFan), data integrity checks.

• Board Reporting

  Automated weekly Friday updates on streaming bookings, active subscribers, ticket sales, and YoY growth with school-year corrections.

When a PlayOn executive asks “how are streaming bookings trending vs. last year with school-year corrections?” Ana plans the query across ticketing, subscription, and engagement data simultaneously. It handles the seasonal adjustments, joins the right tables from the medallion architecture, and returns the answer in Slack — no analyst needed.

The key unlock was validation. Ana can, at the end of her analysis, validate what she’s going to say — as long as you tell her how to do so through ontology and playbooks. That was what built trust across PlayOn’s organization.

[ THE RESULTS ]

What changes when every stakeholder can ask their own questions

Ana became PlayOn’s default analytics layer. Board reporting is automated for weekly Friday updates. New executives spend their first two weeks talking to Ana in Slack. The analyst on-call burden during peak football season is gone.

Analytics turnaround went from 2–3 days to 4–5 hours — that saves an order of magnitude over the course of a year across hundreds of analytics questions per month.