The moment everything broke: leading a 1,000-question RFP

Several years ago, I was responsible for managing one of the largest RFPs of my career. The questionnaire had around 1,000 questions, spanning security, architecture, product capabilities, implementation, service delivery, commercials, CVs, and case studies. Our response team included over ten contributors across sales, presales, marketing, engineering, solution architecture, support, CSM, and professional services. To complicate matters, we were spread across different time zones.

Despite the scale, everything was managed through a single Excel spreadsheet.

There was:

• no centralised knowledge base
• no document governance
• no version control
• no workflow structure
• no consistent review process
• no standard tone or message framework
• and no reliable historical answers — the ones we had were usually outdated

Many product, platform, or integration answers depended on whoever happened to remember the details. Even when strong content existed somewhere, we lost time hunting it down, rewriting it, or trying to reconcile conflicting versions.

By the time we submitted, the quality was surprisingly solid — thanks to an enormous amount of human effort — but the process was chaotic, slow, and exhausting. It was also repeatably inefficient. The problems weren’t due to people or skills; they were due to low process maturity.

That experience forced me to start thinking differently about proposal operations.

Why traditional RFP workflows break down

After leading enough bids, a consistent pattern emerges. Organisations struggle not because the “writing” is difficult, but because the system around the writing isn’t designed for scale.

Symptoms of low-maturity workflows

• Knowledge is scattered across PDFs, SharePoint sites, and inboxes
• Past answers are unreliable or stale
• Work is tracked in spreadsheets
• Contributors overwrite one another’s updates
• SMEs become bottlenecks
• Messaging varies depending on who answered what
• Reviews happen too late to influence quality
• Rework consumes days before deadlines

Industry data backs this up. APMP-aligned research shows that low-maturity BD teams waste up to 30 percent of total proposal effort simply through rework, unclear roles, and disorganised knowledge. Teams operating at “ad hoc” or “informal” maturity levels consistently exhibit the same patterns I saw on that 1,000-question RFP.

The solution is not working harder. It’s understanding and improving process maturity.

Understanding process maturity

Process maturity is a simple concept: the more consistent and repeatable your processes are, the better your outcomes become.

High-maturity processes are:

• predictable
• governed
• documented
• repeatable across staff turnover
• measurable
• continuously improved

Low-maturity processes rely on:

• memory
• heroic effort
• spreadsheets
• individual personalities
• ad-hoc coordination

I'm sure you can easily spot where your organisation sits.

To formalise this, the Business Development Capability Maturity Model (BD-CMM) is often used in the world of proposal management. BD-CMM is a framework that describes five levels of maturity, from ad-hoc to continuously optimised. It is essentially a roadmap from reactive scrambling to operational excellence.

Here’s the simple version:

• Level 1 – Ad Hoc - Rework, chaos, inconsistent answers, heavy reliance on SMEs.
• Level 2 – Managed - Some repeatable processes; tools still inconsistent.
• Level 3 – Defined - Organisation-wide standards, knowledge governance, structured workflows.
• Level 4 – Quantitatively Managed - Metrics drive process control and predictability.
• Level 5 – Optimising - Continuous improvement; performance improves systematically every cycle.

These levels match almost exactly what I’ve seen in real teams.

And the data is striking. Shipley and APMP case studies have observed that teams moving from Level 1 to Level 3 see:

• win-rate increases of 15–30 percent
• significant reductions in cost of sales
• fewer proposal defects
• higher staff satisfaction due to clearer processes

Best-in-class proposal teams operating at the top of these maturity scales often report 70 percent win rates for new business and 98 percent win rates for recompetes.

Understanding this changed the way I approached every proposal after that 1,000-question bid.

Five lessons this experience taught me

1. Most delays come from knowledge chaos, not writing speed

Team members usually have an idea of how to answer questions — they just can’t find the right knowledge to confirm. This is a classic Level-1 pattern.

CMMI and BD-CMM studies show that teams with inconsistent repositories spend 25–40 percent more time searching for usable content. Standardising content and establishing governance cuts cycle times by up to 30 percent.

2. Collaboration collapses without a structured workflow

In my large RFP, we had intelligent, committed people — and yet contributors repeatedly overwrote each other’s work, duplicated answers, or missed changes. That’s not a people problem; it’s a workflow problem.

Moving from ad-hoc practices to Level-3 defined workflows reduces variability, improves accountability, and dramatically boosts predictability.

3. Inconsistent messaging is unavoidable without governance

Without controlled templates, tone rules, GTM-guidance and review stages, the final proposal always feels stitched together. Capability-model research shows that when teams introduce defined standards and structured reviews, proposal error rates drop 25–30 percent.

I saw this repeatedly: introducing even lightweight governance immediately elevated clarity, consistency, and credibility.

4. SMEs become bottlenecks when content patterns aren’t reusable

On that 1,000-question RFP, the same SMEs were asked the same questions, phrased slightly differently, across dozens of sections. To make matters worse, SMEs often get multiple variants of similar questions across multiple RFPs - leading to further bottlenecks. Often SMEs also have a 'day job' that competes for their precious time too.

BD-CMM identifies SME over-dependency as a hallmark of low maturity. When teams introduce reusable knowledge components and governed content, SME load drops 20–40 percent — freeing them to focus on differentiating content and/or their day job.

5. Continuous improvement is the real differentiator

High-maturity organisations don’t wait until the next RFP to improve. They refine their content, workflows, and messaging continuously.

Signals like:

• recurring gaps
• incomplete answers
• repeated SME rewrites
• commonly flagged risks
• inconsistent messaging across bids

…all point to opportunities to move toward Level-5 optimising behaviours.

This was one of the biggest shifts for me personally: maturity is not a destination, but a habit.

A practical playbook you can apply this week

The steps below reflect both my experience and the common patterns maturity-model research highlights across industries.

1. Centralise and govern your knowledge

Create a single source of truth for:

• product documentation
• security policies
• architecture details
• service delivery commitments
• case studies
• CVs
• organisation profiles

CMMI case examples show that implementing content governance can deliver 35 percent improvement in on-time delivery and reduce rework by 30 percent.

2. Standardise workflows and assignments

Define:

• a multi-stage response flow for questions
• clear ownership
• progress dashboards
• status lifecycles
• review checkpoints

Teams moving to Level-3 standardisation report double-digit improvements in accuracy, compliance, and overall quality.

3. Use AI to generate grounded first drafts — not guesses

AI should retrieve and synthesise answers from tightly-governed content, not hallucinate new material.
Used this way, AI can one-shot answer many questions without requiring any human adjustment. And for other answers, it reduces the need for repetitive writing, gives SMEs stronger starting points, and keeps responses consistent and with better grammar and spelling.

4. Review strategically, not reactively

Introduce structured checks for:

• completeness
• tone
• clarity
• risk
• contradictions
• GTM messaging

APMP-aligned consultancies indicate that these types of checks can reduce late-stage rewrite effort by 25–50 percent.

5. Build continuous improvement into every cycle

After each bid, review:

• which questions caused delays
• which answers were repeatedly rewritten
• what knowledge was missing
• which risks surfaced
• which themes or messages were unclear

This is where maturity jumps happen. Small, regular improvements compound quickly.

How AI-native platforms accelerate maturity (and where Respond fits in)

Over the last three years, advanced in large-language models (LLMs) have made it possible for organisations to achieve maturity jumps quickly — not in multi-year transformation programs, but often in a single quarter.

The reason is simple: The best AI-native RFP platforms embed the behaviours of high-maturity teams into the tooling itself.

Using Cognaire Respond as an example:

Accelerating Level 1 → Level 2

• predictable multi-stage workflows
• structured assignments
• audit trails
• shared workspaces
• consistent templates

Supporting Level 3 (“defined processes”)

• governed corpus and content rules
• reusable knowledge components
• consistent language rules
• document versioning and approvals

Enabling Level 4 (“quantitatively managed”)

• completeness and risk scoring
• quality metrics
• strategic analysis
• progress dashboards
• corpus gap identification

Driving Level 5 (“optimising”)

• analysis of human vs AI enhancements
• repeated pattern detection across bids
• content improvement recommendations
• topic-level governance tuning
• continuous improvement loops baked into the workflow

Research-backed benefits such as shorter cycle times, higher-quality content, lower SME burden, and improved win rates align directly with these maturity jumps.

Cognaire Respond operationalises the same practices that maturity models such as BD-CMM recommend.

If I were starting again today

If I could redo that 1,000-question RFP knowing what I know now, I’d change the process immediately — not the team. I’d bring structure in earlier, manage knowledge more intentionally, standardise workflows, and consider the use of AI only when grounded in factual, governed content.

The difference in efficiency, quality, and team experience would have been enormous.

Next step

If you want to see what moving one maturity level looks like in practice, try applying the playbook above to your next questionnaire — or explore how platforms like Cognaire Respond embed these maturity principles directly into the proposal workflow.

References

• Shipley Associates – Business Development Capability Maturity Model (BD-CMM)
• APMP – Proposal Management Best Practices and Benchmark Studies
• CMMI Institute – Capability Maturity Model Integration
• PwC – Process Maturity and Operational Performance Research

‍