FOR FINANCE
Variance explained before the meeting.
Ask about month-end variance, board numbers or plan versus actual. Check the SQL and rows before a number reaches the deck.
Read-only access · every number traced

Why is MRR below the deck?
$398.2k
$14.4k below the deck
37 past_due accounts were counted as paying in v7. Excluding them brings MRR from $412.6k to $398.2k.
Where to start
The questions already
on your agenda.
- 01
Why did gross margin drop 2 points?
The move by product line and cost centre.
- 02
Where are we behind plan on new ARR?
Actuals against the connected plan, by segment and month.
- 03
Which invoices over $10k are 30+ days late?
A live list of owners, ages and amounts.
One worked example
A Monday in Finance.
The brief arrives. Finance opens the trace, corrects the rule and sends it through review.
The variance brief lands in #finance.
The trace reveals 37 past_due accounts.
An approved MRR definition is released to the team.
A change becomes shared knowledge only after review and approval. Earlier definitions remain available.
$412.6k
Trace shown before correction
SELECT SUM(s.mrr_cents) / 100.0 AS mrr
WHERE a.status IN ('active', 'past_due')37 past_due accounts account for $14.4k.
| Account or campaign | Status | Value |
|---|---|---|
| Halden Robotics | active | $4,200 |
| Ostra Health | past_due | $1,850 |
| Pine & Fold | active | $960 |
Review → approval → team knowledge
Count active accounts only. Review before release.From a question to a routine
Close the week with the brief already written.
Schedule your variance review for Slack or email. Every line opens to its trace, and follow-ups stay with the discussion.
Explore scheduled routinesWeekly variance vs plan · week 43
- New ARR · $182.4k · −7.8%
- Gross margin · 71.2% · −2.0 pt
- Collections · $604.0k · +6.1%
Sources attachedIllustrative example
The working stays attached.
Read-only access
Your database stays yours. Connect with a read-only role.
One definition
Use the same reviewed metrics across answers and routines.
A trace you can inspect
Open the SQL and rows behind a number.
Your data. Your model. Your team.
Read-only connections. Use your own model keys.
PostgreSQL
MySQL
ClickHouse
Snowflake
Databricks
BigQueryRedshift
Hosted in the EU · made in Germany
One analyst. Different questions.
Bring your next question.
See OpenDash work with the questions your team already asks.
