Past the Pilot: What It Takes to Scale Skill Building Across the Org
Edmundo Barrientos · 2026-07-22 · 3 min read
TL;DR: Most L&D pilots die in the sandbox. The ones that scale aim at real pain, not eager volunteers; live inside the processes and channels people already work in; and are judged by whether the skill showed up in real work, not by a satisfaction score.
The sandbox is where pilots go to die
In Chief Learning Officer, Artem Ivanenko tells a story L&D leaders know. A team ran an AI coaching pilot for managers practicing difficult performance-review conversations. Twenty motivated managers got on-demand access; over the whole pilot, they logged ten minutes combined, not each. The tool worked. Almost nobody used it.
The diagnosis is worth borrowing: pilots recruit the enthusiastic and grade themselves on satisfaction. A "4.5 out of 5" from volunteers means they enjoyed it, not that the skill reached those who struggle.
Put the practice where the job already is
Ivanenko's fixes are practical. Aim at the point of pain: managers with the worst review-completion record, not volunteers. And build it into the existing workflow, not a separate destination, because every extra click, login, or window switch can drop adoption by double digits.
Take that literally. "Inside the job" means the process step and the channel people already live in. A review happens on a cadence, inside a conversation; its practice should surface there: a prompt in Slack the week reviews open, a rehearsal in the same Google Meet used for the real one, not a separate app to open. Your docs, calls, and recordings already contain the training; turn them into practice where the work is. This holds beyond one team, in leadership, compliance, onboarding, and customer success, not just the sales team.
Adoption is necessary, but not the finish line
Ivanenko's third move goes past adoption: judge the pilot on operational viability, activation within a day, stickiness without nudges, not a satisfaction score. That is the right bar for whether a tool gets used, but using it is not the same as building the skill. Those managers could log in daily and still run the same evasive review.
So "in the flow of work" carries two jobs: where the skill is practiced, and where its transfer is measured. The channel that hosts the rehearsal is where you watch the behavior change, which is what separates the emerging category from a feature. This runs on top of the LMS and coaches, not instead of them.
The proof that earns a rollout
The number that earns the rollout is measured skill transfer, in the work, at scale: not that managers practiced, or even kept practicing, but that reviews got more direct, escalations fell, the ramp got shorter. That evidence already lives in the systems where the work happens. The same process step gave the skill its priority in the first place. A pilot deserves to scale when it shows the skill moved in real work, not when twenty volunteers liked it.
Related reading
- Autonomous Skill Building in the Flow of Work: Defining the New L&D Category
- Course Completion Is Not Skill: What L&D Should Measure Instead
- No Process, No ROI: Aligning Skill Development to the Work It Serves
References
- Chief Learning Officer. Beyond the Sandbox: A Playbook for Scaling L&D Innovation Enterprise-wide (Artem Ivanenko), 2026.
About the publisher: this article is published by Grasply (grasply.ai), a capability builder. Grasply turns what a company already knows, its docs, calls and recordings, into skills people use at work, for employees, partners and customers. Practice turns knowing into doing, and Grasply shows proof that the work changed.