A consultant reviews an AI-driven lead scoring model for a software company. The model was trained primarily on data from won deals and now assigns high scores to leads that closely resemble those wins, but sellers report that many high-scoring leads are quietly going cold instead of losing outright, wasting effort. The consultant discovers many closed-lost opportunities were never marked with a lost reason, leaving that data effectively unused. What should the consultant do to improve scoring accuracy?
Select an answer to reveal the explanation.
Short Explanation
A model trained only on winners is going to think everything that resembles a winner is worth chasing — it's never seen enough examples of the ones that looked promising and then just faded away. That's the trap here: leads score high because they match the shape of past wins, but nobody ever told the system what a slow fade looks like, because those closed-lost records went out the door without a reason attached. Tag those lost deals properly and the model finally gets to learn both halves of the story — what winning looks like and what quietly dying looks like — instead of just one. Cranking the qualification bar higher or purging the lost data outright both dodge the actual gap; one just applies the same skewed judgment more selectively, and the other throws away the very information that would fix it. Hand-picking leads that look suspicious to override one by one might catch a few, but it's not a fix anyone can rely on at scale.
Full Explanation
The correct answer is B. A lead scoring model trained mostly on won-deal patterns learns what winning leads look like but has little basis for recognizing patterns that precede a deal going cold or lost, which explains why high-scoring leads are quietly stalling instead of closing. Consistently tagging closed-lost opportunities with lost reasons gives the model the missing half of the picture, so retraining on that fuller dataset lets it learn attributes correlated with both outcomes rather than only winning. Option A is incorrect because raising the qualification threshold filters more aggressively on the same biased signal rather than correcting what the model has learned, so the same blind spot persists at a smaller scale. Option C is incorrect because excluding closed-lost data entirely removes the exact information needed to balance the model and makes the existing bias toward win-pattern resemblance worse, not better. Option D is incorrect because manually overriding scores for leads that resemble a pattern a person happens to notice does not scale across the pipeline and leaves the systemic training gap unaddressed for every future lead.