--- title: "Refresh List Lab | Email List Quality Research & Methodology" description: "Explore the Refresh List research format for email list quality: sampling rules, denominators, verification outcomes, privacy boundaries, and limitations." url: https://refreshlist.net/lab last_updated: 2026-09-25 --- [Refresh List](/) › The Lab RESEARCH DESK / REFRESHLIST LAB # Measure the list. Show the method. Refresh List Lab is the home for first-party research into email list quality. We publish a finding only when the sample, denominator, verification context, and privacy boundaries can be explained beside it. Evidence-led research notes · Definitions and limits shown · Last reviewed 18 September 2026 THE SHORT ANSWER ## What is Refresh List Lab? It is an evidence-first guide to understanding list quality without turning a demonstration into a statistic. The Lab separates observed outcomes from assumptions, keeps unknowns visible, and puts limitations next to every conclusion. STATUS FORMAT DEFINED The reporting framework defines what a responsible summary should disclose. EVIDENCE RULE NO NUMBER WITHOUT A DENOMINATOR. Every rate must say what was counted, what was excluded, and when it was measured. 01 / THE RESEARCH CONTRACT ## Before a chart, there is a contract. A reader should be able to reproduce the question a study answered—and identify the question it did not answer. ### Define the population State where records came from, which period they represent, and why the sample belongs to the question. ### Declare the denominator Separate source rows, unique addresses, eligible checks, completed outcomes, and unresolved results. ### Protect the data Minimize identifying content, document authorization, and never treat a customer upload as public research by default. ### Show the limitation Coverage gaps, provider behavior, timing, and missing signals stay beside the finding—not buried in fine print. 02 / HOW A STUDY WOULD RUN ## From source rows to a defensible finding. - Frame Write the question, population, inclusion rules, and collection window before looking at rates. - Inspect Count source rows, malformed values, duplicates, and eligible unique addresses without charging or inferring outcomes. - Verify Record provider context, completed results, and unknowns. A provider outcome is evidence at a point in time, not a delivery promise. - Report Publish the numerator, denominator, date, coverage, exclusions, and limitations together. EXAMPLE REPORTING FIELD Unique checked addresses “Unique checked addresses” is not interchangeable with submitted rows. A future report would show both, explain the difference, and state which one supports each rate. Submitted rows not collected Unique checked not collected Unknown outcomes kept separate 03 / EMAIL HEALTH INDEX ## A score needs a visible denominator. An Email Health Index should organize defined measures within a stated assessment window. It should not compress incomplete evidence into a universal quality percentage. ### Potential measures - Duplicate rate, using a stated source-row denominator - Malformed or skipped rate, separate from mailbox results - Valid, invalid, and unknown outcomes within the checked population - Supported role, catch-all, or disposable signals with coverage stated REPORTING STANDARD Show the evidence Every score should show its denominator, date, verification context, and coverage so readers can judge what it does—and does not—represent. [Read the index format](/email-health-index) 04 / RESEARCH NOTES ## Questions worth investigating. [NOTE 01 What does list hygiene actually measure? Separate acquisition context, suppression rules, duplicates, verification, and later delivery feedback.](/email-list-hygiene) [NOTE 02 When does a domain answer stop being a mailbox answer? Catch-all behavior can preserve uncertainty even when a receiving server accepts many names.](/catch-all-email) [NOTE 03 How should an unknown result be handled? Keep the provider reason, checking context, and uncertainty visible instead of inventing confidence.](/verification-methodology) COMMON QUESTIONS ## Research questions, answered. Does Refresh List publish an email quality benchmark today?+ Refresh List only presents score-like summaries when the sample, denominator, date, verification context, and privacy boundaries are visible. What would a real Lab study disclose?+ A real study would disclose its population, sampling and inclusion rules, collection period, duplicate treatment, denominator for every rate, provider coverage, privacy method, and known limitations alongside the findings. Would customer lists become public research data?+ No. Customer lists do not become public research data by default. Any future research dataset would require an explicit, documented privacy and authorization process, with identifying content minimized or removed. How does the Lab treat unknown results?+ Unknown remains separate from valid and invalid. A result that cannot be established is not silently converted into a failure, a pass, or a confidence score. What is the Email Health Index?+ It organizes defined measures such as duplicate rate, malformed rate, verification outcomes, and supported signals when genuine observations and denominators are available. THE STANDARD ## NO INVENTED INSIGHTS. When a study is ready, the method and the finding will arrive together. [READ THE METHODOLOGY](/verification-methodology) [Review the trust center](/trust) ## Sitemap - [All Public Pages](https://refreshlist.net/sitemap.md) - [Email Glossary](https://refreshlist.net/glossary) - [Read This Page And Use Interactive Features](https://refreshlist.net/lab)