Skip to content

ACCA Strategic Professional · Advanced Performance Management · Technology and information systems

Brindle Telecom uses a machine-learning model on customer usage and billing data to estimate each customer's probability of cancelling within three months. The retention team then targets the highest-risk customers with offers. What is the main performance-management risk of relying solely on the model's output when judging the retention team's success?

The key risk is that the model chooses which customers the team treats, so measured retention results may reflect the model's accuracy or selection bias rather than the team's effort. Judging the team solely on those outputs may therefore misattribute performance and give a distorted picture of controllable contribution.

  1. AThe model will automatically raise the budget for retention activity
  2. BCustomers flagged as high risk may be selected by the model itself, so team results may reflect model quality or selection bias rather than team effortCorrect
  3. CPredictive models cannot use billing data under any reporting standard
  4. DUsing customer data removes the need for any non-financial measures

Explanation

If the model decides who is targeted, outcomes such as churn reduction depend on model accuracy and how customers were selected, which can distort the assessment of the team's own contribution. The other options are unsupported: models do not set budgets, no standard bans billing data, and non-financial measures remain relevant.

Did you get it right without looking?

One question tells you little. A timed set on Technology and information systems shows your real accuracy, how long you take and where you lose marks.

Practise Technology and information systemsFree diagnostic mock

More Technology and information systems questions