Historical data analysis with AI-flagged budget risks before they become overruns.
Cost Estimation puts a working financial layer over every project: budgets, actuals, change orders, projected final cost. The AI watches the trajectory — pay items running ahead of budget, change order velocity, contractor invoice patterns — and surfaces risks while there's still time to act on them.
When you're bidding, the AI compares the SOQ against historical agency unit prices and flags lines where the bid is unusually high or low. When you're building, it watches actual placed costs against the estimated and predicts the final number months before the project closes.
Earned value, schedule performance, and cost performance indices update in real time as field data flows in.
Built for construction professionals — not generic SaaS.
Designed to compare bid unit prices against rolling averages from agency historical bid data and flag lines that are statistical outliers. Roadmap — agency data feed integration is in development.
PV, EV, AC, CPI, SPI updated continuously as daily reports accept. Project-level + WBS-level + activity-level drilldown.
Three EAC methods (CPI-weighted, SPI-weighted, blended) shown side-by-side. Variance from BAC explained with the top 5 contributors.
The AI tracks how fast change orders are accumulating compared to similar past projects. Spikes get flagged with the underlying pattern (specific pay item, specific contractor, specific scope).
Pattern detection on contractor billing — front-loading on early items, sudden quantity escalation, items invoiced ahead of measurable progress. Surface in the Financial dashboard for RE review.
S-curve projections for monthly cash needs based on schedule float, expected progress, and historical contractor billing cadence.
The cost layer sits over the same data that drives the Quantity Book — same line items, same source-doc references — so estimated vs. actual is always reconciled to the same source of truth. No double-entry, no spreadsheet drift.
Historical bid analysis is designed to leverage agency open-data feeds — for example, IDOT publishes letting results — ingested on a schedule and indexed by item code. The agency-data ingestion is on the roadmap; the analytical layer is built and ready to consume the data once the feed is live.
EAC forecasting uses the standard PMI formulas (CPI-only, schedule-weighted, blended) and reports all three. The AI doesn't pick one — it shows the spread and explains what assumption each method makes, so the PM stays in charge of the forecast they brief.