Why mobility pilots succeed and scale-ups fail.

Every mobility program we have been called in to rescue had one thing in common: it ran the scale phase like a bigger version of the pilot.

The pilot works. Twelve-month rollout in one city, one vendor, one operational model. Metrics land, stakeholders are happy, the steering committee approves scale-up.

Then the scale fails. Eighteen months later, the program is 30 percent behind plan, the budget has doubled, the country leads are frustrated with HQ, and someone on the executive team is asking whether they should have just built it in-house.

This is not a rare pattern. It is the default path for mobility programs that have not thought about scale as a separate problem. The pilot proves the concept. The scale-up is where most programs actually die, and they die quietly, over 18 months, rather than in a visible single failure.

Why pilots and scales are different problems

A pilot optimises for one question: does this work in one market with full attention? A scale optimises for a completely different question: does this work in 15 or 30 markets with reduced attention per market?

The operational model that works in a pilot rarely works at scale without significant redesign. This is where most programs break, because the team that delivered the pilot assumes "more of the same" will work at scale, and HQ signs off on the assumption rather than on the actual math.

Three specific things break. We have seen the same pattern across global mapping programs, fleet operations rollouts, and MaaS platform expansions.

The three places mobility scale-ups actually break

Break 1: Vendor orchestration at scale

In the pilot, you probably had one vendor, a direct relationship, and the ability to escalate to a founder or partner quickly. At scale, you have 8 to 15 vendors per market or per function, each with their own contract structure, each with different SLAs, each hired at different moments in the program's evolution.

The vendor stack that looks manageable from HQ's slide deck is often unmanageable on the ground. In a multi-country rollout we diagnosed recently, the country leads were spending roughly 60 percent of their working time on vendor coordination alone. The program was not behind schedule because the work was slow. It was behind schedule because nobody had capacity to do the work.

Break 2: Country lead hiring and autonomy

In the pilot, you had one excellent person running one market. At scale, you need 15 to 30 excellent people, and you do not have them. Most scale-ups compromise: they hire country leads who are "good enough" and then centralise decisions back to HQ to compensate.

This backfires. The centralisation adds cycle time to every decision the country lead should be making in hours rather than weeks. Meanwhile, the good-enough country lead gets frustrated, disengages, and either leaves or underperforms. The fix is usually reopening the search for two to three critical countries and giving those leads real autonomy.

Break 3: The operational rhythm does not translate

In the pilot, you had weekly or daily standups with the whole team. At scale, you cannot have 15 country leads on one call. The cadence has to change.

Most scale-ups keep the pilot rhythm for three to six months, realise it is not working, and then swing to the other extreme: no cadence at all beyond monthly reporting. Both failure modes are common. The actual answer is a tiered cadence: country leads sync weekly with a regional ops lead, regional leads sync biweekly with HQ, HQ runs monthly program-level reviews. Setting this up takes deliberate design, not evolution.

What good scale execution actually looks like

The mobility programs we have seen succeed at scale share three traits.

They treat scale as a separate program with a separate plan

Not "phase 2 of the pilot." A real project with its own operational model, its own hiring plan, its own vendor strategy, and its own success metrics. The pilot team does not automatically become the scale team. The scale phase needs different skills entirely, heavier on operations, lighter on innovation.

They hire country leads before they need them

The default mistake is to hire the country lead two months before launching in that country. The actual need is four to six months before, because local hiring, entity registration, onboarding, and ramp-up all take longer than expected. Programs that pre-hire country leads three months before the hard date finish roughly on schedule. Programs that hire at the two-month mark finish three to six months late.

They centralise tooling, decentralise execution

Common platform for ticketing, vendor management, and reporting. Local ownership of all day-to-day decisions within defined guardrails. This is the structure that scales without grinding to a halt on HQ bottlenecks or fragmenting into 15 different operational models.

How to structure the pilot-to-scale transition

If you have a mobility pilot that is working and you are planning scale, three moves make the transition cleaner.

Separate the pilot retrospective from the scale design. Pilot review is about what worked and what did not in one market. Scale design is about what structure will work in 15 markets. These are different conversations with different people.

Write the country lead role before you recruit anyone. The job description for a country lead at scale is different from the pilot lead's role. More operational autonomy, more vendor management, less direct founder access, more HQ coordination. Clarify this before hiring anyone you will be stuck with for two years.

Budget for the operational overhead of scale. The HQ operations headcount is the single most underestimated line in most mobility program budgets. Get this one right and everything else becomes easier to manage.

The practical next step

If you have a mobility program moving from pilot into scale, the cheapest move you can make this month is to answer three questions:

• Have you designed the scale operating model as a separate project, or are you assuming pilot plus resources will work?
• Have you costed the HQ operations headcount needed to coordinate 15 or 30 country operations, and is it in the budget?
• Have you written the country lead role for the scale phase, distinct from the pilot lead role?

If the answer to any of these is no, you are setting up for the quiet 18-month failure pattern. It is cheaper to redesign now than to rescue later.