What makes a contact refresh worthwhile?
A refresh is worthwhile when its expected avoided cost or recovered value exceeds the cost of the check. Use evidence from the relevant cohort, separate field types, and include the chance that a check resolves the issue. A record’s age alone does not determine whether the refresh will pay back.
Apollo’s CSV enrichment documentation notes that enrichment can consume additional credits. More generally, every refresh process has a cost, whether paid in credits, supplier fees or staff time. Do not describe a new download as a refresh unless new evidence was actually obtained.
Choose the field that matters to the campaign. Rechecking a role may be useful for a relevant account; repairing an email for a company outside the target is not the same investment. Save the original record and the new observation so recovered rows can be distinguished from merely copied rows.
Sources: Apollo CSV enrichment documentation
How do you calculate a refresh threshold?
Compare per-row check cost with the expected avoidable loss from an unresolved defect. Multiply the defect probability by the chance the refresh resolves it and the loss avoided. Use the same units throughout. The resulting threshold is a planning tool whose value depends on the assumptions and evidence behind it.
Hypothetical inputs: checking costs $0.60 per row; an unresolved relevant defect creates $12 of avoidable staff work; a check resolves 80% of such defects. The break-even defect probability is $0.60 / (0.8 × $12) = 6.25%. Above that assumed probability, expected avoided work exceeds checking cost in this simplified model.
For 300 rows with an assumed 10% defect rate, expected resolved defects are 300 × 0.10 × 0.8 = 24. Avoided work is 24 × $12 = $288; checks cost 300 × $0.60 = $180; expected net benefit is $108. These are hypothetical expectations, not guaranteed savings or a measured decay rate.
| Input or result | Value |
|---|---|
| Check cost | $180 |
| Expected resolved defects | 24 |
| Expected avoided work | $288 |
| Expected net benefit | $108 |
| Break-even defect probability | 6.25% |
How do you replace assumptions with evidence?
Pilot the refresh on a defined cohort and measure what changed, what was resolved and what work was actually avoided. Keep the before-and-after evidence and review ambiguous cases. Use those observations to update the decision, rather than turning an assumed defect rate into a permanent rule for every list.
Track each changed field and its consequence. A corrected title is not automatically a recovered opportunity; a new address may still belong to the wrong buyer. Include checks that produced no resolution in the cost denominator. Otherwise the model rewards only successful work while hiding failed attempts.
For new research, specify the target, evidence and checked-date requirements up front. HEED’s accepted-row definition and review process provide a basis for evaluating its deliverable. Compare a fresh scoped order with a refresh using the same campaign criteria, and account for the value of existing records you can legitimately retain.
- Use cohort-specific defect evidence rather than an industry-wide slogan.
- Count unresolved checks in total cost.
- Separate avoided staff work from speculative future revenue.
Sources: HEED service and record acceptance terms
Define a B2B research briefSpecify the audience, required fields and exclusions before choosing a batch.
Common questions
Is 6.25% a universal threshold for refreshing data?
No. It follows from the hypothetical $0.60 check cost, 80% resolution rate and $12 avoided-work assumption. Change any input and the threshold changes.
Should a newer file always replace the old one?
No. Compare field evidence and keep valid ownership, history and exclusions. A later import date alone does not establish better information.