Where this sits relative to BD-CMM
The established model in this field is BD-CMM, the Business Development Capability Maturity Model, developed by Shipley Associates with APMP and patterned after the Carnegie Mellon SEI capability maturity models. It has five levels and four key process categories, it is actively maintained — version 2.1 was published in February 2021 — and it is the right instrument for a formal capability assessment across an entire business-development function. What follows does a different job. It isolates the single axis that decides what proposal work actually costs an organization, and lets a proposal lead place their firm on it in about five minutes.
The question is not what tools a proposal team has. It is what the organization still has the morning after the submission goes out.
Why a tool inventory is a poor maturity score
Two firms can run the same software and sit three levels apart. The usual maturity questionnaires ask what a team has — a repository, a template library, a review workflow — and a team can answer yes to all three while still rebuilding every proposal from memory. A more useful test asks what is left once the deadline passes. Capability that disappears with the submission was never organizational; it was one person's week.
The five levels
Each level is defined by what persists after submission, not by what the team owns.
- Personal — capability belongs to individuals, and the answer to "have we done this before?" is a colleague's name. What survives: files.
- Collected — documents are gathered in one place, but without owners or approval states, so the drive holds four versions of the company description and no indication of which is current. What survives: files, findable by name.
- Governed — reusable content carries an owner, a version and an approval state, so a writer can tell approved material from a draft without asking anyone. What survives: content that can be trusted on sight.
- Connected — claims are linked to the evidence supporting them, and requirements to the clause they came from. What survives: the join between the two, which is the expensive part to rebuild.
- Compounding — outcomes, review decisions and clarification questions are recorded against the work they concern. What survives: the reasoning, which is why effort per proposal starts falling as the record grows.
One question that places you at each boundary
Ask them in order. The first question answered no marks the boundary you have not crossed, and that is the level you are at.
- Can a colleague who did not work on the last submission find what it drew on, without asking anyone?
- Can they tell which version of the company description is the approved one, without asking anyone?
- Can they see which project record a statement about past experience actually rests on?
- Can they see why the last bid was scoped the way it was, and what the review changed?
Why most organizations stop at level two
Because storage alone takes you there and no further. A shared drive or a document library moves a team from one to two in the week it is installed, and then stops: nothing in it records who owns a piece of content, which version is approved, or what a claim rests on. Levels three, four and five need governed content, evidence linked to the claims it supports, and outcomes recorded against the work they concern — which is precisely what a proposal intelligence platform adds on top of storage. The firms that climb fastest are the ones that make the recording part of a step they already perform, rather than a new step nobody has time for.
Primary sources
Quotations are reproduced from the published document at the date shown above. Procurement documents are revised, and the request for proposals you receive governs your bid — check the clause in your own tender before relying on it.