Kuiz berkaitan panduan · Sederhana Tahap

AI untuk Pembaharuan Insurans, Pengekalan dan Kuiz Jualan Silang

Menguji ciri model churn, sebab pemodelan peningkatan mengalahkan menyasarkan skor luput tertinggi dan masa untuk memasarkan semula pembaharuan sebelum pelanggan membeli-belah.

soalan 1 daripada 8

Which client profile would a churn model most likely treat as a high lapse risk?

Pratonton setiap soalan8 soalan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

  1. Which client profile would a churn model most likely treat as a high lapse risk?
    • A client with bundled home and auto policies, no rate change and ten years with the agency
    • A single-policy client with a large rate increase at renewal and recent billing complaints
    • A client who just renewed early and paid in full
    • A household that recently added an umbrella policy
  2. Why do many agencies review renewals 60 to 90 days before the renewal date?
    • Carriers ban contact closer to renewal
    • Premiums are lowest exactly 90 days out
    • It is required to order a CLUE report
    • It leaves time to remarket before the client gets the renewal notice and starts shopping
  3. Why isn't calling the clients with the highest churn scores first always the best strategy?
    • Some will leave no matter what; outreach helps most with clients in between
    • High scores are always wrong
    • Regulators forbid contacting high-risk clients
    • High-score clients never answer the phone
  4. What does an uplift model need that a basic propensity model doesn't?
    • Only the client's age
    • A list of competitor prices
    • Outcomes from both contacted and uncontacted groups, ideally a randomized holdout
    • A larger font on renewal letters
  5. A model uses the cancellation reason recorded after a policy lapsed as an input. What problem is this?
    • Calibration drift
    • Target leakage that inflates accuracy in testing
    • Entity resolution failure
    • A circular reference
  6. Why should lapse scores be calibrated before ranking accounts by expected premium at risk?
    • Calibration removes the need for any human review
    • It makes every score equal
    • It turns the scores into premium amounts
    • So predicted probabilities match actual lapse rates, which makes probability times premium meaningful
  7. A household adds a 16-year-old driver and has no umbrella policy. What cross-sell approach does the guide favor?
    • A simple rule tied to the documented exposure, prompting the agent to discuss higher limits and an umbrella
    • An opaque model that recommends the highest-commission product
    • Emailing every client about every product
    • Adding an umbrella automatically without asking
  8. Why does the guide link cross-selling to retention?
    • Cross-selling is required by law at every renewal
    • Single-policy clients never leave
    • Single-policy clients are widely observed to leave more easily than households with bundled policies
    • Bundled policies always cost more