Skip to article
ecomservices.io
Chargebacks · How to run this
MatchingNot on this page yetPricingCalculatorIntakeCalculatorReason codesCalculatorRepresentmentCalculatorThresholdsCalculatorAllocationCalculatorOptimisationCalculatorPrevention stackCalculatorProviders

Library · Chargebacks · How to run this

Overturn is not review lag

How to run thisChargebacks3 steps
CalculatorManual review overturn rate · Calculator lives in the articleOpen

Overturns ÷ reviews make overturn %. Type both.

Pair: Manual review overturn rate on Chargebacks. Lag $: manual review lag. Depth: review queue depth.

What you type

Overturns — manual decisions that reverse the automated allow/block call.

Reviews — total manual reviews in that same window.

Overturn % = overturns ÷ reviews × 100. Retune rules when overturn % is fat — do not invent an overturn chip.

What not to do

Do not invent overturn %. Do not mix auto-clears into reviews. Do not treat step-up abandons as overturns. Do not average every brand when one owns the queue.

Related: Chargebacks, manual review lag, review queue depth, fraud review queue cost.

Procedure

  1. Type overturns and reviews.
  2. Read overturn %.
  3. Retune rules when overturn % is material.

Failure modes

  • Inventing overturn %.
  • Mixing auto-clears into reviews.
  • Treating step-up abandons as overturns.

Worked example

None on this page yet. When one lands it will be a real case: what broke, the number, and what changed. Not a composite.

Glossary

Review overturn %
Manual overturns ÷ manual reviews × 100.

Work with us — We run this job on your accounts. Services

FAQ

Is this manual review lag?
Sibling queue clock. [Manual review lag](/manuals/sift-manual-review-lag) prices lag hours × AOV. This page is overturns ÷ reviews.
Is this review queue depth?
Related staffing. [Review queue depth](/manuals/sift-review-queue-depth) is open ÷ capacity. This page is how often humans reverse the model.
What overturn % should I invent?
None. Type overturns and manual reviews from the same Kount / fraud queue window.