Product · Investment banking

Comps and pitchbook engine: ticker in, branded deck out

Enter a ticker and get a peer set, trading comps, a valuation football field, a formulas-only Excel model, and a pitch deck rendered into the firm's own PowerPoint template, with every number traceable to the filing it came from.

Obsidian Group product, built for investment bankers
Every cell
traced to its source filing
Your template
decks render into the firm's own .pptx
Formulas only
Excel output, no pasted values
The build
Stack
PythonFastAPINext.jsSEC EDGAR XBRLCloud RunCloud SQL

Why we built it

The first week of any banking pitch is the same week every time: pull filings, build the comps, reconcile the numbers, format the deck. Associates do it by hand, and the output is only as trustworthy as the last person who touched the spreadsheet. We built the tool we wished we had.

What it does

  • Provenance first. Financials are pulled from SEC filings and stored with their source. Every figure in the summary clicks through to the filing and line it came from. A number without a source does not exist in the system.
  • Comps that behave like a banker’s. Peer sets with medians that exclude the target, flags for fiscal-year mismatches and captive finance arms, and last-twelve-months math done the way a reviewer expects.
  • An Excel model you can defend. The comps sheet is formulas only, over a data sheet of hardcodes, with provenance in cell notes. Nothing is pasted as a value.
  • A deck in your template. Upload the firm’s own PowerPoint template once. The engine reads its fonts, accent colors, and slide size and renders the pitch pages into it, so the output looks like the firm made it, because it did.
  • Valuation and private companies. A football field across methods, a DCF, and support for private comparables entered with their own provenance.

Where it is

Live and gated, in use for our own work and available for firms that want the engine running on their templates and their peer sets. It is also the foundation we build from when a bank asks for a sprint: the data layer, the provenance model, and the rendering pipeline are already there.

Want something like this built for your firm?

Two weeks and one working tool is the usual starting point. Tell us the workflow and we will scope it on the first call.