Huber+Suhner Reached 99% AI Pilot Adoption — The Playbook

The single biggest mistake I see SMB owners make in AI rollouts is treating "pilot adoption" as the goal. Huber+Suhner's reported 99% pilot adoption is striking — but the playbook behind it is what makes it transferable, not the headline.

What Huber+Suhner actually did

The reported 99% figure isn't from a 5,000-person blanket deployment. It's from a deliberately scoped pilot — a controlled population, hand-picked roles, structured weekly cadence. That distinction matters.

Three program-design moves drove the number:

1. Population scoping. They didn't randomly assign licenses. They picked roles where AI fit was likely high — knowledge workers with repetitive text-based workflows, supported by managers who agreed to model usage.
2. Manager-led participation. Every pilot participant had a manager who was also a participant. This single move kills the "my boss doesn't get it" failure mode.
3. First-hour use case per participant. Before the pilot started, every participant had a concrete first task they would attempt with AI in their actual workflow. Not "explore the tool" — a specific deliverable.

Definition: First-hour use case — for AI rollouts, the specific work task each participant will attempt with the tool in their first session. Not a tutorial; a real piece of work that produces a real output.

Why pilots can hit 99% and rollouts rarely do

Pilots and full rollouts are different animals. A pilot is volunteer, scoped, supported. A full rollout is mandatory, broad, often under-supported. The question to ask is: what part of pilot dynamics survives the transition?

The answer, drawn from the Huber+Suhner pattern and many others: the program design survives. The volunteer-self-selection bias does not.

The four components of pilot success that DO transfer:

• Role-fit profiling before licensing
• Manager-included cohorts (not just IC's trained while managers stay outside)
• First-hour use case for every participant
• Per-role measurement from day one

The components that DON'T transfer:

• Volunteer self-selection (everyone in pilot wanted to be there)
• Heightened executive attention (you can't replicate "the CEO is watching" at scale)
• Forgiveness of mistakes (pilots are forgiven; rollouts are scrutinized)

What this means for an SMB

You shouldn't try to hit 99%. Aim for 70-85% MAU at full rollout — that's the durable, achievable tier (where JCB sits). What you SHOULD copy from Huber+Suhner is the pilot rigor, applied to your full rollout.

Apply pilot-grade scoping to a full rollout. Don't blanket-license. Profile roles by AI-fit, prioritize the top 60-70%, defer the rest until week 3.

Always include managers. Every cohort should have managers in it. The "train ICs, hope managers absorb it later" pattern produces drift.

Pre-write the first-hour use case for every participant. Before training day one, every employee has a concrete first task — written down. Champions help with this in the week before.

Measure from day one. MAU by role, time-saved by role, use-case library entries per role. No general averages — the average lies, as the UK pilot data showed.

Tool tip (Course for Business): Our 6-week program borrows the rigor pieces from pilots like Huber+Suhner — role-fit profiling, manager-included cohorts, pre-written first-hour use cases, per-role measurement — and applies them to full company rollouts at 30-500 person scale. AI Champions (1:15-20) carry the program through weeks 2-6. Augment, don't replace is the framing every cohort opens with — every employee ships their first AI automation in week one. https://course.aiadvisoryboard.me/business

What the playbook looks like at SMB scale

Strip Huber+Suhner's pilot down to its components and you get a 6-week program for any 30-500 person company:

Pre-week: Champions selected (1:15-20 ratio). Each champion writes a first-hour use case for each colleague in their cohort.

Week 1: Cohort labs (15-25 people each, including managers). Every participant ships their first automation using the pre-written use case. Use-case library starts.

Week 2: Champions run clinics. Per-role measurement starts. Shadow-AI hygiene addressed.

Week 3-4: Wave 2 and 3 cohorts (deferred roles). More sophisticated use cases. Manager modeling check.

Week 5-6: Integration into team SOPs. MAU review by role. Hand-off to internal champion structure.

This produces durable MAU in the 70-85% tier. Don't aim for 99% — aim for the durable tier and you'll outperform most pilot-then-rollout patterns.

Team scan (what AI champions report after week 1)

• Cohort completion: 95%+ when managers are included; falls to 60-70% when only ICs attend
• Adoption: 75-90% trained staff using AI for real work ≥3x/week
• First-hour use case completed: 90%+ of participants when pre-written; <50% when ad-hoc
• Saved time per person: 30-55 min/day in week one (high end of typical)
• Manager-led demos surfaced: champions report 3-5 manager-modeled wins per week
• Use-case library: 25-40 entries by end of week one
• Shadow AI flags: typically 1-2 incidents — addressed in week 2
• Resistance pockets: <10% when role-fit profiling is rigorous (vs 15-20% with blanket licensing)
• Drop-off candidates: roles flagged in pre-week profiling — deferred, not abandoned
• MAU trend: rising into week 3, steady-state by week 5

What NOT to copy from Huber+Suhner

Two specific traps:

• Don't think pilot adoption == rollout adoption. A 99% pilot can be followed by a 30% rollout if program design weakens at scale. The 70-85% durable tier is more honest.
• Don't try to engineer "99%" through forced participation. That produces compliance, not adoption — which is the same as the Microsoft 300K rollout pattern (mandatory licenses, no usage).

Tool tip (Course for Business): The Huber+Suhner pilot rigor — role-fit profiling, manager-included cohorts, first-hour use cases, per-role measurement — is exactly what the 6-week program is built on. We don't promise 99% (that's a pilot artifact). We aim for the durable 70-85% MAU tier where programs sustain past month one. AI Champions (1:15-20), Shoulder-to-Shoulder hot seats, every employee ships in five days. https://course.aiadvisoryboard.me/business

Micro-case (what changes after 7-14 days)

A 110-person industrial services firm runs a Huber+Suhner-style scoped wave: 60 priority-role employees plus their managers in week 1, deferred 50 in week 3. Pre-week, champions write a first-hour use case for each of the 60 priority participants. By day 7, 90%+ have completed their first use case successfully and have shipped a working automation. By day 14, MAU is sitting at 82% in the priority wave, the use-case library has 32 entries, and the deferred wave is being onboarded with stronger profiling. The CEO, who participated in the manager cohort, sees adoption pull through — the modeling effect is visible.

Note on this case: This example is illustrative — based on typical patterns we observe with companies of 30-500 employees, not a single named client. Specific numbers are rounded approximations of common ranges, not guarantees. Huber+Suhner's 99% pilot adoption is the publicly reported figure from Huber+Suhner.

FAQ

Was the 99% real, or marketing? Both can be true. The figure is real for the scoped pilot — and the scoping is what makes it possible. Apply that scoping rigor to your full rollout and you'll hit 70-85% MAU durably, which is more useful than a peak number that fades.

Should I run a pilot before a full rollout? For SMBs, usually no — your population is small enough that the "pilot" and "rollout" are the same thing. Skip the pilot phase, apply pilot-grade rigor to week one of full rollout.

What's the right ratio of managers to ICs in cohorts? Mixed — ideally every cohort has 2-4 managers among 15-25 ICs. Don't run "manager-only" cohorts; the modeling needs to happen alongside team members.

How do I write a "first-hour use case" for someone in a role I don't fully understand? That's what champions are for. The champion sits with the person 15 minutes in pre-week, asks "what's the most repetitive text-based task you do every week," and turns the answer into a first-hour use case.

What if some roles genuinely don't fit AI? Flag them explicitly in pre-week profiling. Don't force the license. Revisit in 90 days when the tooling and your role's workflow may have evolved.

Conclusion

Huber+Suhner's 99% pilot adoption is striking but not directly copyable — pilot dynamics differ from rollout dynamics. What IS copyable is the rigor: role-fit profiling, manager-included cohorts, first-hour use cases, per-role measurement.

Apply pilot rigor to your full rollout. Aim for the durable 70-85% MAU tier. That's better than a 99% peak that decays.

If you want every employee to ship their first AI automation in five days — book a 30-min call and we'll map your team's first week: https://course.aiadvisoryboard.me/business