The problem
A software engagement is shaped long before anyone writes a story. It is shaped by the RFP the client sent, the statement of work both sides signed, and the schedules attached to the contract. Those documents run to dozens of pages, they are written by several people over several weeks, and they rarely say the same thing twice in the same way.
Someone has to read them closely and pull out what matters for delivery: what exactly is in scope, what is explicitly out, which integrations and environments the client will provide, what service levels and acceptance criteria apply, and which clauses quietly move risk onto the delivery team. That reading usually happens under deadline, by one or two people, and the result is a list in a document that loses its link to the clauses. Three months later, when a scope disagreement comes up, the team is back in the PDF trying to find the sentence.
Good looks like this: every scope item, assumption, obligation and risk on one list, each pointing at the clause it came from; anything vague, contradictory or missing called out as a question; and the questions ordered so the most expensive ambiguity is raised first.
How it works
Forge Contracts is a Blueprint. We build it on demand in your engagement. This section describes the approach we would take.
It reads the document set: the RFP or SOW, any schedules, and clarification answers or addenda. It converts PDF and Word files to text while keeping section and clause numbers, so every item it writes can point back to a clause.
It works clause by clause. Each section is read on its own and its items extracted: deliverables, exclusions, client dependencies, environments, data and security obligations, service levels, acceptance terms, payment milestones tied to delivery. A second pass merges the items across the whole document, and that is where it finds the problems: a deliverable in one section that an exclusion in another contradicts, an acceptance criterion with no measure, a dependency on the client with no date.
It is steered by a checklist you agree with us: the clauses your organisation always wants to see, the terms it never accepts, and the risks it has been caught by before. That checklist is written down and versioned, so a reviewer can see why something was flagged.
Then it is checked. Every item has to cite a clause, and the citation is matched against the text; an item with no source is rejected. A checklist item that the document never addresses is reported as missing, not silently skipped.
People decide. Your delivery lead, and your commercial or legal reviewer where needed, go through the questions and risks first. Contracts prepares the review; it does not give legal advice, and it does not decide whether to bid or sign.
Where it gets it wrong
- Legal language is subtle. “Shall use reasonable efforts” and “shall” carry very different risk, and a summary can flatten the difference. Contracts quotes the exact wording next to every obligation, so the reviewer reads the clause, not a paraphrase.
- Meaning spreads across documents. A definition in the master agreement can change what a line in the SOW means. Contracts reads every document you give it together, but it cannot read the one you did not include, and it says which referenced documents were missing.
- Tables and scanned pages lose structure. Pricing tables, requirement matrices and scanned annexes are where text extraction is weakest. Contracts flags pages it could not read cleanly so a person checks them by hand.
- It does not know your appetite for risk. A clause that one organisation accepts routinely is a deal-breaker for another. The agreed checklist carries that judgement; without it, Contracts lists the clause and leaves the call to you.
We would rather Contracts list a clause as a question than tell you confidently that it is fine.
Why not just prompt a chatbot
You can upload an RFP to a chatbot and ask for the risks. You will get a reasonable-looking list with no clause references, it will miss what the document never mentions, and you cannot tell what it skipped. What makes Contracts useful is the clause-by-clause pass, the citation that is checked against the text, the agreed checklist that turns silence into a flagged gap, and an output your delivery lead can work through line by line. That is the part we build, and the part you keep.
What you get
- In your engagement: built on demand in your engagement, and run on the SOW and its schedules before kickoff, so scope questions are settled before the first sprint.
- Under your brand: services firms can run it inside their own presales or delivery-assurance process, with their own checklist.
- Built to own: we tune it to your templates, clause checklist and review workflow, and hand over the source.