'%20fill='%23fff'/%3e%3c/svg%3e)
How Blackstone leverages TextQL to enhance its data and analytics capabilities
Thousands of tables, hundreds of fields per table, billions of rows. Blackstone’s team brought TextQL in to make every one of them queryable — without depending on the small group of people who know how everything joins together.
1000s
tables made queryable in plain English
Billions
of rows reasoned across by Ana
1 prompt
to schedule a recurring analytics workflow that runs itself
"When I walked out of the first meeting with TextQL, I thought, wow, this is a product that many of our portfolio companies could benefit from."
John Stecher, Chief Technology Officer, Blackstone
About Blackstone
Blackstone is a leading global alternative asset manager with $1.3 trillion in AUM, spanning private equity, real estate, credit, and insurance.
Industry
Financial Services
Company Size
~5,000+ employees
Headquarters
New York, NY
Pain Point
Answering a business question meant waiting on the few people who knew where the data lived and how it fit together.
Products Used
Your data team, fully leveraged.
See what Ana does with the data you already have.
Request Demo →Request Demo →
Blackstone works against petabytes of data — spread across cloud warehouses and on-premises systems, with millions of dollars in market data layered on top. The raw material for almost any business question lives somewhere in there. The challenge isn’t getting the data. It’s getting to it.
Before TextQL, the path from a business question to a defensible answer ran through that short list. They knew which tables held which facts, which ones could be safely combined, and how to phrase a query so it didn’t quietly miscount along the way. Reconciling company names, removing duplicates, lining up the right records — every question routed back to the same handful of experts.
[ THE PROBLEM ]
When the data layer is a group of people
Thousands of tables, hundreds of fields per table, billions of rows. A short list of people knew how it all fit together.
When the data layer is a group of people
PETABYTES OF DATA · A SHORT LIST OF EXPERTS
PORTFOLIO_HOLDINGS_DAILY
12.4B rows
MARKET_DATA_NAV_HISTORY
8.7B rows
LP_COMMITMENT_LEDGER
410M rows
FUND_ENTITY_OWNERSHIP
92M rows
+ thousands more
?
3 people
know how this fits
Senior leader's question
What's our cross-fund exposure to financials, segmented by vintage?
Day 4 · still in the queue
every question routes back to the same handful of experts.
FIG. 01 — BEFORE TEXTQL
A question from a senior leader, pulling across several datasets, segmenting by time period, rendering into a dashboard, would land in an analyst’s queue and come back in days, sometimes a week. None of the institutional knowledge that made the answer possible was written down in a form a tool could use.
Generic AI query tools couldn’t fill the gap. They could produce a query, but they didn’t know which tables were trustworthy, how they connected, or where the cleanup logic lived. In a regulated, high-stakes business, a confident wrong answer is worse than no answer at all.
Scaling that model meant one thing: hiring more and more people.
"I’ve been surprised often by its capability. I asked it to make me a “playbook” — give me a runbook for solving a problem. It turns out TextQL has a feature called Playbooks. It knew about its own capabilities and configured a scheduled playbook itself. To this day, every Monday morning, I get the results of that question in my email."
Rob Wisniewski, CTO, Credit & Insurance Technology, Blackstone
[ THE SOLUTION ]
Solution
Pointing Ana at the warehouse
Blackstone validated TextQL’s agent, Ana, against analytics they already maintained — a benchmark with answers they could verify line by line. Ana matched them, faster, and the team could keep asking follow-up questions in the same thread.
The setup is the underappreciated part. You point Ana at your databases, and it explores the data, works out how everything connects, and writes down what it all means. What usually takes months of painful manual mapping happens in days — with the kind of precision a firm like Blackstone actually needs.
Validated line by line.
GROUND-TRUTH BENCHMARK · ANA MATCHED IT
47 joins
312 entities
12 semantic terms
FUND
- • fund_id
- name
- vintage_year
- strategy
LP
- • lp_id
- commitment_usd
- jurisdiction
NAV
- • as_of_date
- value_usd
- fund_id
HOLDING
- • holding_id
- fund_id
- portco_id
- cost_basis
PORTFOLIO_CO
- • portco_id
- sector
- geography
VINTAGE
- • vintage_year
- cohort_irr
- tvpi
FIG. 02 — what usually takes months of manual mapping happens in days.
[ THE RESULTS ]
The gap between question and answer, closed
The biggest impact wasn’t a faster query. It was the time before the query — finding the right tables, working out how they fit together, knowing what to trust. That’s where the days got spent.
QUARTERLY IMPACT
Key Results
1000s
tables made queryable in plain English
Billions
of rows reasoned across by Ana
1 prompt
to schedule a recurring analytics workflow that runs itself
From days to seconds
Locating the right data and figuring out how to combine it used to depend on the few people who held that knowledge. Ana does it directly against the warehouse, in seconds.
Correctness you can defend
Every answer ships with the work behind it: which tables it touched, how it combined them, what filters it applied. In a regulated business, that’s the difference between a useful answer and an unusable one.
Self-aware tooling
Ana knows its own capabilities. Asked for a “playbook,” it set one up on its own — and now it delivers the answer to email every Monday morning.
More people working with the data
With Ana handling the legwork, the bar to ask a useful question dropped from “expert” to “anyone who can describe what they want.”
The bottleneck has never been the analysis itself. It’s everything that has to happen before the analysis can start: finding the right data, remembering how it fits, phrasing the question in a way that won’t quietly produce the wrong number. Move that work from a small group of people into a tool the whole organization can use, and the firm can scale insight without scaling headcount.
"The more questions you can ask, and the quicker the turnaround time from asking the question to getting results back, the more you can actually start to ask more and more informed questions."
John Stecher, Chief Technology Officer, Blackstone
Read more from TextQL
Announcements
Dashboards 2.0
We doubled down on dashboard infrastructure. 2.0 ships parallel data loading, full Plotly Dash support, parameterized queries, dedicated sandcastles, incremental edits, draft/publish state, and built-in usage analytics.
Announcements
TextQL Raises $17M, Anchored by Blackstone Innovations Investments, to Help Bring Intelligence to Enterprise Data
TextQL closes $17M in strategic investment anchored by Blackstone Innovations Investments to scale AI-native analytics infrastructure for the enterprise.
Engineering
Introducing API Connectors
Your warehouse has your historical data. Your SaaS tools have everything else. Now Ana can reach both in the same conversation, securely, with no tokens exposed.
Go from question to conviction in minutes, not weeks.