character·quilt
One engine.
700 bookstores.
How Follett's B2C and B2B teams put a single creative engine behind 700+ campus bookstores and ~120 RFPs a year — on-brand, deployed in HubSpot, hands-off.
character·quilt
700+ bookstores.
~120 RFPs a year. One creative bottleneck.
Follett runs marketing across 700+ campus bookstores — each one a different university, a different palette, a different mascot, a different calendar. Move-in day at Emerson lands a different week than rivalry day at Michigan.
The B2B RFP team — three people, ~120 RFPs a year, Nov–Mar crunch — sits on the same engine. CharacterQuilt is the layer that finally lets the variation scale.
- AI customization existed — deployment never closed
- Marketers were hand-duplicating templates 700 times
- RFPs consumed the team's bandwidth all winter
character·quilt
One source asset. Re-brand it for 700.
The "university-branded NYU shirt" campaign auto-regenerated as the on-brand version for Emerson, BYU and Michigan — different logos, different palettes, different campus context.
character·quilt
700 bookstores.
One workflow.
The same workflow runs across every store — no designer touches each one, no marketer duplicates the email 700 times.
Bookstore-level brand customization
One source asset re-generated for every campus — logos, palettes, mascots, store context. NYU → 700 in one pass.
On-brand creative for every campus moment
Move-in at one school. Rivalry weekend at another. Finals-week coffee. Pulled from each store's actual calendar — reads as locally made.
DAM-first, generate when missing
Has the asset? Pull from the DAM. Doesn't exist (Emerson's "Thinking Cup" café)? Generate it on-brand so creative never bottlenecks.
Closes the deployment loop
700 personalized variants assembled into HubSpot email templates, scheduled, sequences live — no per-store hand-offs. The gap nobody else closed.
character·quilt
Pitches that look like
they took a week — built in an afternoon.
RFP response generation
Knowledge base ingests ~15 past wins to learn voice, structure, and proof points. New RFP in → Content Agent drafts first pass → team reviews instead of writes.
1–2 weeks → no longer a fire drill.
Proactive ABM landing pages
Identify high-probability RFPs before they hit. Deck and landing page already built when the request lands. Reactive becomes proactive.
POC: University of Oregon ABM page shipped.
Sales leave-behinds at pitch speed
One-pagers, HTML follow-up emails, supporting collateral — generated alongside the deck. AEs leave the room with a full kit, not a promise to send something next week.
character·quilt
Days of hand-offs Minutes from portal to live.
Follett already used AI for some customization. The gap was always deployment — taking 700 variants and actually getting them live. The Deployment Agent closes the loop.
character·quilt
A library of agents.
Not a tool to learn.
Four specialists do the work. The Orchestrator conducts. No new UI for Follett's team to master.
Creative Agent Visual
Bookstore campaigns, co-branded decks, ABM pages, store renderings, social graphics — across 700+ universities.
Content Agent Copy
Bookstore campaign copy, RFP responses, GVP outbound, follow-up emails — trained on Follett's actual past wins and brand voice.
Deployment Agent Bridge
The Salesforce ↔ HubSpot bridge. Asset built in portal → live in HubSpot → ABM or bookstore sequence firing, hands-off.
Internet Researcher Context
Gathers the external context that makes each campaign feel local — campus events, programs, recent news, competitive landscape per account.
Orchestrator Conducts
Sequences multi-asset projects, manages the campaign-planning document for complex builds, routes feedback back through the system. The other four don't run alone — the Orchestrator is what makes them feel like one engine.
character·quilt
The stack Follett already runs on.
No migration. The agents operate Follett's existing tools — read access for analytics, deploy access for publishing.
Email templates, ABM sequences, bookstore campaign sends. Built in portal → live in HubSpot in minutes, not days.
Past meetings, opportunity stage, contact-level signals — and the trigger layer for sequenced follow-up.
Pulls reusable response components, ingests new wins back into the library — so the next RFP starts from a stronger baseline.
Store photos, lifestyle shots, brand guidelines, university partnership marks — wired in on Day 1.