Campfire Announces Forecast: An End to the Mid-Month Guessing Game
Campfire Announces Forecast: An End to the Mid-Month Guessing Game
Ask any CFO what happens between the 1st and the close, and you'll hear a version of the same story. Someone opens a spreadsheet. They pull an export from the ERP. They eyeball the trend line from last quarter, factor in a gut feeling about the pipeline, and give the board a number they're maybe 70% confident in. Then they do it again next month, from scratch, because the spreadsheet doesn't remember what it assumed last time.
This is the mid-month forecasting problem, and it's one of the most common gaps in modern finance operations. It's also one of the clearest examples of a broader pattern: the tools finance teams use to look backward (the ERP) and the tools they use to look forward (planning software) don't talk to each other. Campfire built Forecast to close that gap.
What is Campfire Forecast?
Campfire Forecast is a near-term income statement forecast built directly into the ERP. It shows where the current month will land, on any day of the month, starting from posted actuals and every schedule already committed in the general ledger's subledgers, with Ember, Campfire's AI agent, estimating only the portion of the month that hasn't happened yet.
It's not a replacement for annual planning. It's built for a narrower, more urgent question: on the 12th of the month, with the board meeting on the 20th, where are we actually going to land?
The problem with how this question usually gets answered
There are really only two ways finance teams answer this question today, and both have the same structural weakness.
The spreadsheet and gut check. Someone exports a trial balance or GL detail, rebuilds a forecast model by hand, and layers in judgment calls about what's likely to happen before month end. It works, but it's rebuilt from scratch every month, it's only as current as the last export, and it lives outside the system of record. Nobody can see what assumptions went into last month's number without asking the person who made it.
The planning tool. FP&A platforms are excellent at long-range planning, but most of them sit downstream of the ERP and only receive trial balance data, not the underlying subledger detail. That means a planning tool forecasting this month's revenue doesn't actually know what's already been contracted, invoiced, or scheduled to amortize. It has to model the whole month as an assumption, even the parts that are already, functionally, known.
Both approaches are guessing about things the ERP has already answered.
How Forecast actually works
Forecast is built on three layers of a single income statement, and the distinction between them is the whole point.
Actuals are the transactions that have already posted to the general ledger. This is Campfire's official record: customer invoices, vendor bills, payroll syncs, manual journal entries, and the month-end postings the accounting team runs for revenue recognition, depreciation, and prepaid amortization.
Committed is everything already scheduled in Campfire's four subledgers, revenue, prepaids, fixed assets, and leases, shown as if it had already posted. This is the layer that makes Forecast fundamentally different from a spreadsheet or an external planning tool. A signed contract has a revenue recognition schedule sitting in the system the moment it's signed. A prepaid bill has an amortization schedule the moment it's paid. A lease has a monthly expense schedule for its entire term. None of that requires a forecast, because it's already known. Committed is deterministic, ties out line by line to the underlying contract, bill, asset, or lease, and involves no AI at all.
Estimated is Ember's projection for the part of the month that genuinely hasn't happened yet, using the company's own historical trend and seasonality as a baseline, with Committed as the floor it estimates on top of. Click into any number and Ember shows what it assumed in a single sentence, and finance teams can shape that estimate directly: type a note about an upcoming contractor spend, attach a headcount plan, or connect a CRM or HRIS through Campfire's MCP integration. Ember re-estimates in response. Nothing in any of this posts to the general ledger automatically. Estimates are advisory, and the accounting team still reviews and posts at close exactly as they do today.
The result is a forecast where most of the month isn't actually a forecast. It's already known, it's just not usually visible until close.
Why this matters more than it sounds like it should
The reason this is worth a dedicated product, rather than a feature buried in a dashboard, comes down to where the data lives. A planning tool or a spreadsheet is always working from an export, a snapshot of the ERP from whenever someone last pulled it. Campfire's subledgers update continuously as transactions happen, which means Committed is only possible for a system that already owns the ledger. That's a structural advantage, not just a UX one: no tool sitting outside the ERP can produce the Committed layer, because it doesn't have access to the schedules that generate it.
This is also why Forecast and a continuous close solve adjacent but different problems. A continuous close, the model built around posting and reconciling transactions in real time throughout the period, is about when the books get updated. Campfire's Committed view already gives finance teams that same forward-looking picture, every schedule read forward, without changing when things actually post, so the accounting team still reviews and posts at close, which is what auditors expect. Forecast then goes a step further and estimates the part of the month nobody has posted yet, which isn't something a continuous close model addresses on its own.
What a finance team is saying
"Three views of Actuals, Committed & Estimates built from our GL, Subledgers and AI insights gives exactly the breakdowns we need. This is a forecast tailor made for accounting and finance teams to see how their numbers are shaping up mid month in a transparent, useful way," said Zach Doyle, CPA, Senior Manager, Accounting at Advisor360°, who tested Forecast ahead of launch.
| Spreadsheet | Planning tool | Campfire Forecast | |
|---|---|---|---|
| Data source | SpreadsheetManual export, point in time | Planning toolTrial balance export | Campfire ForecastLive subledgers and GL |
| Committed dollars (contracts, prepaids, leases, depreciation) | SpreadsheetManually rebuilt each month | Planning toolUsually not visible | Campfire ForecastNative, tied out automatically |
| Shows its assumptions | SpreadsheetDepends on the person who built it | Planning toolSometimes | Campfire ForecastYes, one sentence per number |
| Updates during the month | SpreadsheetOnly when rebuilt | Planning toolRarely in real time | Campfire ForecastAny time, on demand |
| Posts to the ledger | SpreadsheetNo | Planning toolNo | Campfire ForecastNo, advisory only |
| Built for | SpreadsheetAd hoc mid-month check-ins | Planning toolAnnual and long-range planning | Campfire ForecastThe mid-month "where do we land" question |
Where Forecast doesn't replace other tools
Worth being direct about this: Forecast isn't an annual planning tool, and it isn't trying to be. Long-range planning, scenario modeling, and headcount planning across multiple years are still better served by dedicated FP&A software, which is precisely why Forecast can pull in a headcount plan or a driver file as an input rather than trying to rebuild that function itself. Forecast is narrowly built for near-term accuracy on a question planning tools structurally can't answer as precisely, because they don't have subledger-level access to the ledger.
Getting started
Forecast is available now to existing Campfire customers, either directly in-app or through your CS contact, and is included as part of implementation for new customers.
