A consulting agent reads an inbound RFP, pulls matching case studies, drafts a proposal, and routes it for partner and sales-lead approval. Notion holds the capability library. Google Workspace holds the draft. Pitch holds the deck. Dock holds the agent's interpretation and the two sign-offs required before sending. The proposal goes out only after both reviewers approve the row.
Notion, Google Workspace, and Pitch stay the system of record for the raw data. Dock is the system of record for what the agent interprets. Each Dock row carries a pointer back to the platform record, agent identity, decision, reviewer, and timestamp. The agent re-fetches platform data via fresh API reads when it needs current state.
The Proposals surface
| RFP | Client | Notion case studies cited | Pitch deck | GDoc draft | Agent | Proposed fee | Partner | Sales lead | Status |
|---|---|---|---|---|---|---|---|---|---|
| Acme retail GTM | Acme Co | 3 (CPG-rebrand, DTC-launch, channel-mix) | pitch.com/d/acme-gtm | gdoc/acme-prop-v3 | proposals-agent | $185k / 8wk | jordan@ approved 05-29 | priya@ approved 05-30 | sent |
| FinServ ops review | NorthBank | 2 (regional-bank-ops, core-migration) | pitch.com/d/nb-ops | gdoc/nb-prop-v2 | proposals-agent | $240k / 12wk | jordan@ approved 05-29 | priya@ rejected 05-30, fee low | revise |
| Healthcare digital | Mercy Health | 4 (provider-EHR, patient-portal, billing-ai, claims) | pitch.com/d/mercy-dig | gdoc/mercy-prop-v1 | proposals-agent | $320k / 16wk | pending | pending | partner-review |
Every row points to the source RFP email, the Notion case study IDs, the Pitch deck URL, and the Google Doc revision. The agent identity column is the agent's own credential, not a partner's borrowed login. Both reviewer columns must flip to approved before the row's status can move to sent.
A worked workflow
A new RFP lands Monday at 9am. The agent reads the requirements, queries Notion for case studies tagged with matching industry, and finds three. It drafts a 12-page proposal in Google Docs from the firm template, generates a Pitch deck from the slide library, and proposes a fee band from similar past engagements. It writes a Dock row with the RFP pointer, cited case study IDs, deck URL, doc URL, and fee.
The partner reviews positioning on Tuesday and approves. The sales lead reviews fee and timeline on Wednesday and rejects NorthBank as too low for scope. The agent reads the rejection, adjusts the fee, regenerates affected slides, and updates the row. The sales lead approves the revised row. The proposal goes out Thursday. The row is the durable record of why this proposal said what it said.
Why it matters
APMP research found 62 percent of proposal professionals work over 40 hours per week on proposals, and 77 percent describe their processes as suboptimal [1]. The bottleneck is not drafting. It is the loop between RFP, capability library, partner judgment, and sales-lead pricing. An agent compresses drafting from days to an hour. Partner and sales-lead reviews stay human and attributed. Dock makes both a row state, not a Slack thread.
Related rails: Dock for consulting and agencies, Dock for research for capability reads, Dock for design for decks, Dock for sales for fee logic, and agent audit and compliance for the trail procurement asks to see.
Try Dock for your next RFP cycle.
FAQ
Does the agent send the proposal? No. It drafts and stages. Sending requires partner-approved and sales-lead-approved on the row, each stamped with reviewer identity and timestamp.
What if the capability library is messy? The agent cites only tagged, dated Notion pages. Untagged pages do not enter the draft. Tag hygiene becomes a partner habit.
How does the fee get set? The agent proposes a band from comparable past engagements and writes the reasoning into the row. The sales lead approves, edits, or rejects. The agent does not negotiate.
Can a client see the audit trail? Yes. Each row exports a record of inputs cited, agent identity, reviewers, and timestamps. Procurement teams ask for this more each quarter.
[1] Responsive, "RFP Statistics," responsive.io/blog/rfp-statistics. Cites APMP membership data on proposal hours and process satisfaction. [2] Forrester, B2B Marketing research on go-to-market transformation, forrester.com/blogs/category/b2b-marketing.
