Planning and FP&A
Governed forecast versions, content hashing, parent diffs, and decision-report adoption.
Solutions · Head of FP&A
Forecast from today's actuals, not from last month.
Every decision you make is only as good as the data behind it, and that data has always had to be pulled together by hand before the analysis could even start. Sebastian keeps planning on the same graph the close runs on, so each quarter builds on the last instead of starting from a blank sheet.
The planning function
Five FP&A workflows, all on the same actuals as the close.
Planning and forecasting
Driver-based models built from actuals rather than a blank spreadsheet, with line-item support behind every assumption you have to defend.
driver-basedDepartment-level planning
Plan at the granularity the business actually runs at: what marketing spent on events, what engineering consumed to ship. Not a top-line roll-up.
granularBudget versus actual
Variance against plan with drill-down to the transaction, so a conversation about a miss starts from evidence rather than a guess.
plan to actualFlux analysis
Period-over-period movement with the driver already identified, prepared for review instead of reconstructed the night before.
preparedBoard and management reporting
Board decks and business reviews assembled from the same numbers the ledger holds, with every figure traceable to source.
board-readyPlanning and forecasting
The forecast stops waiting on the close.
A forecast normally waits for accounting to finish, then waits again while someone reconciles those actuals against the plan in a separate system, which adds days on top of the close. When planning reads the same graph the close runs on, the forecast updates as accounting lands.
sebastian · fp&a / forecast versions
Q3 forecast · drivers
re-projected on July actuals- Headcount34 hires land Aug 12,312 → 2,346
- Support softwarescales with headcount$9.10 / seat / mo
- Partner blended raterenegotiated May$164 / hr
Suggested revision
Raise Q3 opex +$1,847K: thirty-four hires land Aug 1 across three offices, and support software follows the seats.
Accept into forecastAdjust drivers
Append-only versions with authorship and parent diffs
Conflict protection when two people work the same version
The plan of record changes only when an eligible owner adopts it
Department-level planning
Plan at the level the business actually operates.
Most FP&A tools sit too high to answer the questions department leaders actually care about. Sebastian joins department systems to accounting history on one graph, so you can build and defend a plan at the level of the spend that drives it.
- Department drivers joined to transaction-level history
- Attribution for the investment behind a line, not just its total
- Zoom from a department question out to the consolidated P&L
sebastian · fp&a / department drivers
FY26 to FY28 plan
BaseUpsideDownside
LineFY26FY27FY28
Revenue$858M$1,029M$1,214M
Gross margin60.0%62.8%65.1%
Opex$423M$542M$628M
EBITDA$91.8M$104.2M$162.3M
FY26 opens from closed actuals,not a re-keyed export
Sebastian
Unify accounting data with FP&A to drive your business.
Budget versus actual
Variance you can drill, not just display.
Actuals and plan live on the same record, so budget-versus-actual is a drill-down rather than a data request. The line-item support sits under every number, which means the conversation with a department lead starts from the transactions instead of from an argument about whose figures are right.
- One version of actuals shared with accounting
- Line-item support behind every planned figure
- No reconciliation between two systems’ views of the same metric
sebastian · fp&a / budget versus actual
Review the parent diff
Version comparisonCurrent versionVersion 12
Parent versionVersion 11
Changed inputsDiff retained
Stale writeRejected on conflict
Conflict checks prevent an older browser state from silently replacing a newer save.
Flux analysis
The explanation is ready before the meeting.
When actuals miss, the investigation usually happens under time pressure the night before a board meeting, the worst possible moment to be reconstructing a story. Flux runs off the same graph, so the driver behind a movement is already identified and waiting for you to review and sign.
- Period-over-period movement with its driver attached
- Commentary proposed for review, never published unreviewed
- Timing, one-offs, and genuine run-rate change told apart
sebastian · fp&a / flux
July performance story
vs FY26 budget- Implementation labor−$2,418K
- Cloud hosting credits+$847K
- Granite Peak renewal+$783K
Drafted by Sebastian
Board and management reporting
Build the board deck from actuals from the ledger.
Board decks and business reviews are built from the same record the close produced, so a figure in the deck and a figure in the ledger cannot quietly disagree. Every number carries its lineage, which is what lets you defend an assumption in the room.
- Reporting sourced from the same graph as the close
- Every figure traceable to the transaction behind it
- Narrative drafted for review, with the reviewer named
sebastian · fp&a / board reporting
Board memo · July 2026
drafting…Executive summary
Every figure drills to the ledger. No number in the memo is free-typed.
What changes
Planning that keeps up with the business.
Forecast the day close lands
Planning reads the same graph the close runs on, so there is no separate reconciliation step adding days on top of the close.
Department-level detail
Answer what marketing spent on events or what engineering burned to ship, instead of planning against a top-line roll-up.
Nothing to reconcile
Accounting and FP&A read one set of transactions, so the deck and the ledger cannot quietly disagree.
Keep your ERP. Close and plan in one pane.
AI for accounting and FP&A on the ledger you already run. No migration. No rip and replace.
Agents at work · Matching an invoice to its PO line…