Governed metrics, no warehouse

One set of numbers your whole company can trust.

Strickle connects to your PostgreSQL database, lets you define each metric once — owned, with lineage — and shows the exact query behind every answer. No warehouse to build, no data hire.

Early access. No spam, no sales blast.

MRR · last month

Example

€48,200

Verified

The query it ran

select sum(amount_cents) / 100.0
from subscriptions
where status = 'active'
  and month = date_trunc('month', now() - interval '1 month')

source: subscriptions · owner: finance

Every answer comes with the query it ran and who owns the definition — so you can check it, not just trust it.

Read-only, always

Shows every query

Definitions with named owners

No warehouse to build

The numbers shouldn't be a fight.

Two people open two spreadsheets, and the leadership meeting turns into a tug-of-war over whose number is right.

Board week means days of exporting reports, re-checking formulas, and fixing stale dates by hand.

Someone needs a figure, so they ping you — and then they wait, because you're the only one who can pull it.

Half the dashboards are stale, and nobody fully trusts the other half.

You've quietly become the single point of failure for every number the company reports.

It works until it doesn't — and it never scales past you.

How it works

Connect, define, ask — and see the proof.

Connect once, define once, and every answer is checkable.

1

Connect your PostgreSQL database

Point Strickle at your database with a read-only connection — a replica and a least-privilege role if you have one. No warehouse, no pipeline to build. (More sources coming.)

2

Define each metric once

Write a metric like MRR or active customers a single time, give it an owner and lineage, and that one definition is what everyone gets — no drift, no second version.

3

Ask in plain English — and see the query

Ask from your own AI assistant, connected to Strickle. Strickle turns the question into SQL against your governed definitions, runs it read-only, and shows you the exact query.

Why you can trust it

Built so a skeptic can check the work.

Trust isn't a tagline here — it's how the product is built.

Every answer shows its work

You see the exact SQL that ran and the source it touched. Nothing is a black box you have to take on faith.

One definition, one number, everywhere

Governed metrics have a named owner and visible lineage, so the figure in your board deck is the same one in Slack and on the dashboard.

Holds up when it matters

Numbers built from owned definitions are ones you can defend — in a board meeting, in front of a customer, and in diligence.

Tells you when it isn't sure

When a question is ambiguous or the data is stale, Strickle flags it instead of guessing. It won't show a number it can't stand behind.

Who it's for

Ask the questions you actually report on

Real questions about your business — answered from your own data, with the query shown.

  • What was MRR last month?
  • Which accounts churned last month?
  • How many active customers do we have by plan?
  • How many new customers did we add this quarter?
  • What's our net revenue retention?

If you own the numbers

Define each metric once and your whole team can self-serve — governed, so they get the right number instead of inventing their own. You stop being the human spreadsheet.

Founders · RevOps · Finance · BizOps

If you just need an answer

Ask in plain English and get a trustworthy, defensible answer — with the query right there. No SQL, no database access, no waiting on someone to pull it for you.

Sales · CS · Marketing · Ops · Execs

How it's different

Governed, not a black box.

vs spreadsheets

No more monthly hand-assembly or whose-number-is-right debates — one current set of numbers, the same everywhere.

vs BI tools (Looker, Power BI, Tableau)

No dashboards to build by hand and no per-seat license for someone who needs a figure once a month — governed answers you can ask for in plain English, not just charts to wire up.

vs ungoverned 'ask your database' AI

Point-AI-at-your-database tools guess and hand you a black box. Strickle compiles answers from your governed definitions and shows the exact query — so the number is one you can defend.

Pricing

Priced for everyone who needs a number

Your whole company — not a handful of paid seats.

The person who needs a figure once a month is included, without buying them a seat. Value-anchored against a fraction of the cost of a data hire — not per head.

Pricing is non-seat. Billing isn't live yet — early-access companies help us set it.

Get early access

Questions

Questions a skeptic asks first

Will it be correct on my messy data?
Strickle reads your real schema and runs against your actual database — no sample data, ever. Because every answer shows the exact SQL and the source it touched, you can verify the number rather than take it on faith. Governed metrics compile deterministically from a definition you control, so the result doesn't drift.
Who maintains the definitions — and what happens when that person leaves?
Each governed metric is owned by a team, not a single person, and carries visible lineage. The definition lives in Strickle with its history, so it survives a handoff instead of walking out the door with whoever wrote it.
Can I trust an automated number?
Only as far as you can check it — which is the point. Every answer shows its work, and Strickle flags ambiguity or stale data instead of guessing. It won't display a number it can't stand behind.
Is this another tool that becomes shelfware?
There's no warehouse to build and no modeling project to schedule. You connect a read-only PostgreSQL database, define the metrics you already report on, and start asking. The work is defining your numbers once — not standing up infrastructure.
What will it cost?
Pricing is non-seat — your whole company is included, not a fixed set of paid seats, so occasional users don't need one. We anchor it against a fraction of the cost of a data hire, not per head. Billing isn't live during early access; early companies help us shape it.
We already have Looker, a spreadsheet, or could just use ChatGPT — why this?
Spreadsheets drift and don't scale past you; BI tools mean dashboards to build and seats to buy; pointing a general AI at your database gives you a confident guess with no way to check it. Strickle sits in the middle — governed definitions you own, run against your own data, with the query shown every time.

Get one set of numbers your whole company can trust.

We're onboarding early-access companies now. Connect your PostgreSQL database, define your metrics once, and stop being the only person who can pull the numbers.

Early access. No spam, no sales blast.