The situation
Uplimit was an AI learning platform co-founded by ex-Coursera leadership, with strong product and marquee logos but no settled story. Their framework led with efficiency in running cohorts, which describes a feature rather than a category, and it left the company sounding like a better Coursera. That is the one thing a founding team from Coursera cannot afford to sound like. What I saw was a lean marketing function with a clear-eyed executive team, and a brief to build the story and the machine that distributes it.
The decision
The framework I was handed was competent and inert. Rather than write a better one at my desk, I designed the positioning as an exercise: a structured set of questions put to the CEO, CRO and marketer, built to surface the story from customer proof rather than settle it by internal consensus. What came out of it I anchored in the two-sigma problem — Bloom’s finding that one-to-one tutoring outperforms classroom instruction by two standard deviations, against the long-standing assumption that you can’t scale it. That reframed Uplimit as the company solving a known, named problem in learning science, and handed the L&D audience something they already believed in to hang the product on.
02 · The anchor · Bloom’s two-sigma problem
Illustrative · after Bloom (1984)
A gap the audience already believed in
The framework I was handed
Efficiency in running cohorts
The reframe
Scaling what tutoring proves
The reframe wasn’t a slide that sat in a deck. It was written to open the company’s guide and it carried into the webinar program, where it had to hold up in front of a roster of former chief learning officers.
Then I built the pages it ran on. Three industry pages shipped on one template, and the front-end work is the part I’d point to: schema.org markup, complete Open Graph and Twitter cards, preconnect and DNS-prefetch hints, and CSS scoped to the theme’s universal class names so a single block held across all three templates without a fix having to be made three times. Uplimit didn’t need a second vendor to get from narrative to published page.
MOOCs made learning affordable, accessible, and flexible. AI is making it personal.
03 · Webinar program · public speakers
Speakers and event titles are public
Who the narrative had to hold up in front of
Live webinar · Uplimit
Beyond limits: Upskilling in the age of AI disruption
- Josh BersinIndustry analyst, HR and learning
- Jeanne MeisterAuthor and executive, future of work
- Karie WillyerdFormer CLO, Visa · SAP · SunByline ghostwritten
- Eric DinglerFormer CLO, Deloitte
- Adrianna AltermanSalesforce Ventures
Inside the same engagement
Full scope
The decision above carries the argument. Everything below ran inside the same engagement, catalogued.
- 7
- workstreams
01Narrative & pages
3 workstreams · none public
Messaging framework
An updated framework and guiding narrative that opened the company guide.
Persona landing pages
Three pages built and shipped on one template, front end included.
Ghostwriting
Executive blogs, and a byline for an external former CLO.
02Programs & plumbing
4 workstreams · none public
Webinar content engine
Landing pages, segmented sequences, speaker and brand social, recap and follow-up.
“Level Up” newsletter
Taken from an existing concept through to subject-line development.
Series A launch
Announcement collateral, launch-event content, and a toolkit built for investors.
Sales enablement
Attendee and absentee follow-up, plus outbound enterprise sequences.
The bottom line
The reframe carried the company through its Series A announcement and the webinar cycle that followed, which is the honest span of it — the story kept evolving after I left, as it should. The “Beyond limits” event title came out of a direct workshop loop with the CRO, his stem and my second half. This engagement is thin on confirmed downstream numbers, since registration and conversion data sit with the client, so I’ve kept it to what’s substantiated.