Productizing content into a demand engine

The situation

Greenplaces is a sustainability and carbon accounting platform that helps mid-market companies measure, reduce, and report emissions under the same disclosure requirements the Fortune 500 faces, usually without a dedicated sustainability team. As the company matured past founder-led marketing, it needed a repeatable demand engine and content that could translate a complex regulatory landscape into a buying journey. I owned content, campaigns, webinars, lifecycle and the product-led acquisition work, alongside an in-house marketing and sustainability team.

Regulation Radar: value before the ask

The default move in this category is to write another explainer eBook. I turned the education into a tool instead. Climate reporting rules were changing fast, SB 253, SB 261, SEC disclosures, CSRD, and buyers were confused about which applied to them. I took Regulation Radar from concept through copy and build direction: an interactive assessment that gave each visitor an immediate, personalized answer to “which regulations apply to me,” while capturing qualification data useful to sales. The library behind it covered 19 regulations across four regions. The value arrived before the ask, which flips the usual gated-PDF pattern, and that’s why it converted.

Then I pointed the same tool at two more audiences. A sales version let a rep generate a personalized regulatory report for a target account, and a CS version did the same for existing customers. The decision that mattered there was a small one: internally generated reports deliberately don’t enroll anyone in HubSpot, because a rep pulling a report on a prospect shouldn’t drop that person into a marketing nurture sequence. Each version shipped with the enablement around it, an announcement email, a four-step sales sequence and a CS template, so it actually got used.

The judgment call I’m proudest of on the content side was one I stopped. A draft eBook for staffing firms claimed those firms risked penalties for non-compliance. I pushed back. Most of the frameworks in question are voluntary, and any penalty depends on jurisdiction and public status, so the claim wouldn’t survive a knowledgeable reader. In a category where the buyer’s entire job is compliance, one shaky regulatory claim is a credibility event that costs the deal. I caught it before it shipped, and did the same on a conference deck where the regulatory framing was wrong.

“

The best lead magnet does something for you before it asks for anything.

The Supplier Database: an AI pipeline with a human checkpoint

Regulation Radar answered “what must I comply with.” The next question the same buyer asks is “what does my biggest customer require of me,” and that one had no answer anywhere. Sustainability teams kept hitting the same wall, with the information buried across supplier codes, sustainability policies and long PDFs. So I built the other half of the map: a searchable database covering 80-plus companies.

The interesting part is that nobody researched and wrote 80 profiles by hand. I designed a pipeline. Clay ran the research and generated a structured first draft for each company — its requirements, source documents and industry — which staged into Google Sheets, and then hit the step that mattered most: human review. Long supplier-code PDFs are where the workflow was least reliable, since a requirement can be missed, misread, or attributed to the wrong source, so a reviewer validated every entry against its underlying documents before it moved to the production sheet. SpreadSimple published the approved dataset as a scalable web experience, and I later began migrating it into Greenplaces’ WordPress stack.

The decision worth naming isn’t “I used AI to build a database.” It’s that I built an AI-assisted content operation and put the human checkpoint exactly where the workflow was weakest. That’s what let a small team produce a proprietary asset at a scale manual work couldn’t reach, without shipping errors into a compliance-sensitive audience.

“

I put the human checkpoint exactly where the workflow was weakest.

02 · Supplier database · production pipeline

Diagram · not a screen capture

Where the human goes

1 · Input

Target companies

Named accounts to research.

2 · AI research

Clay

Enrichment plus first-draft generation per company.

3 · Staging

Google Sheets

Generated drafts held for review.

4 · The intervention

Human QA

Every generated answer validated against the source PDFs. Nothing publishes unread.

5 · Production

Google Sheets

“in prod”

6 · Published

SpreadSimple

80+

global company pages

Stages 1–3: Content production engine

Stages 5–6: Publishing engine

A person, where the AI was weakest

AI drafted it. A person cleared it.

Pipeline designed & operated by Comarketer.ai

One engagement

Full scope

The stories above carry the argument. Everything below ran inside the same engagement — one demand engine, catalogued.

12

workstreams

1

public, linked

All 12Public only · 1Expand allCollapse all

01Content & campaigns

4 workstreams · 1 live

Vertical guide programGuides + nurture

A guide per vertical, each with its own nurture track and landing page.

Webinar engineCampaign system

Full campaigns: landing page, promotion, paid and organic push, follow-up.

Tiny Climate ActsParticipation campaignLive

A participation campaign: 650 people across 52 teams, 16,935 actions.

View it live →

Competitive teardownsResearch

Competitor guides pulled before each build, so the library stayed differentiated.

02Demand & field

4 workstreams · none public

Request a Demo rebuildConversion page

Rebuilt around executive research, reporting deadlines and customer proof.

Field marketing & eventsEvents

A conference booth, Climate Week, Legal ESG London, an IR-lead roundtable.

Executive speakingDeck + talk track

Deck and talk track written for the VP of Sustainability’s London session.

Newsletter programEmail

Migrated into HubSpot and run on a fixed cadence.

03Lifecycle & standards

4 workstreams · none public

Segmentation & automationsLifecycle

The lifecycle layer running underneath every campaign.

Deliverability engineeringEmail ops

HubSpot throttling and suppression, so the cadence didn’t burn the list.

Attribution plumbingCRM

Registration source carried through into the CRM.

Brand voice & style guideStandards

Authored, and it governed content briefs across the program.

The bottom line

The voice guide became the shared standard, and nearly every initiative became a system rather than a one-off: the assessment, the database, the webinar engine, the guide program, the newsletter. That is the method, and it is the reason a small team kept shipping. Tiny Climate Acts ran the same way — 650 participants across 52 teams logged 16,935 actions over three weeks in April 2025. Pipeline and revenue attribution live in the client’s systems.