A telecom's customer-billing AI pilot succeeded in one regional call center after an envision workshop and a limited experiment phase. Leadership wants to move directly to a company-wide rollout across all call centers next quarter. What is the most important step before that?
Select an answer to reveal the explanation.
Short Explanation
Think of the pilot as proving the idea works in one city, not that it'll work everywhere — you still need to test it somewhere bigger and messier before you bet the whole company on it. A launch phase does exactly that: it checks the solution against more varied, representative conditions first. Skip it, and you find out about regional differences the hard way, all at once.
Full Explanation
The phased approach moves from envision to experiment to launch before scale, and a launch phase's job is to validate the solution across a broader, more representative set of conditions than a single pilot site — exactly the step missing here. Duplicating the pilot's exact configuration in every call center immediately skips that validation step, and a national rollout is likely to surface regional differences in call volume, workflow, or customer mix all at once instead of catching them in a controlled launch first. Pausing all AI work until a fully documented long-term roadmap gets board approval overcorrects: the actual gap is a missing launch phase, not a missing roadmap, and halting everything sacrifices the momentum the pilot already built for no matching benefit. Replacing the pilot team with a new team dedicated to enterprise rollout throws away the institutional knowledge that team built about how the tool actually behaves in practice, handing the harder rollout work to people with no direct experience with it. The operating principle: each phase exists to reduce a specific category of risk before the next commitment, and skipping the launch phase is one of the most common causes of scaling failures traced back to unrepresentative pilot data. Operational check: before a company-wide rollout, ask whether the solution has been tested anywhere beyond the original pilot's specific conditions.