An administrator enables predictive lead scoring for a Dynamics 365 Sales Premium environment that went live two weeks ago. The scoring column shows 'Not enough data to generate a score' for every open lead. The environment already has several hundred open leads and a functioning lead form. What is the most likely reason scoring has not activated?
Select an answer to reveal the explanation.
Short Explanation
A predictive scoring model can't guess at qualification patterns out of thin air; it has to study past examples of leads that actually turned into wins or losses before it can label anything else with confidence. Two weeks after go-live, there just hasn't been enough time for a meaningful number of leads to reach a closed state, so the model is still waiting on the raw material it needs, regardless of how many open leads are sitting in the pipeline. The distractors here invent things that don't exist: a form-publishing step that would show a missing field rather than this message, a manual training button that Sales Insights doesn't have, and a special privilege for administrators that isn't part of the security model. Nothing here is broken; it just needs time and a growing set of closed outcomes before scores start appearing.
Full Explanation
The correct answer is D. Predictive lead scoring is a machine-learning model that must be trained on the organization's own historical outcomes, so it needs a meaningful number of leads that have already been marked won or lost before it can learn which attributes correlate with qualification. An environment that has been live for only two weeks typically has not accumulated enough closed leads yet, even if the total volume of open leads is large, which is why every open lead still shows the insufficient-data message. Option A is incorrect because publishing the form affects whether the score field is visible on the form, not whether the underlying model has enough data to compute a score; an unpublished form would show a missing field, not a data-insufficiency message. Option B is incorrect because there is no manual 'model training' button in Sales Insights settings; training happens automatically on a schedule as historical data accumulates. Option C is incorrect because predictive scoring is governed by feature enablement and licensing, not by a 'Predictive Model Consumer' privilege attached to the System Administrator role, which does not exist. The team should wait for enough leads to close and allow the model to retrain.