Email list cleaning is removing the addresses that hurt deliverability. Verifiers such as ZeroBounce, NeverBounce and Kickbox check whether each address exists and accepts mail. Gmail and Yahoo set their limits on a different number, the spam complaint rate on delivered mail, which a list of entirely valid addresses can still breach.
That gap is why a freshly verified list can still land in spam. This guide quotes the mailbox providers, the verifiers and two email platforms’ help centers as served on October 8, 2026, then gives a step-by-step cleaning method and a guide to choosing a verifier. It is published by Relvino, which decides per shopper whether to send at all across email, SMS and pop-ups, so read it as informed but interested.
Page one for “email list cleaning” splits between two kinds of company. The verifiers describe a check on the address. NeverBounce’s home page promises to “protect your sender reputation by removing invalid emails, hard bounces, and spam traps before you hit send.” Kickbox sorts every result into three buckets, “Deliverable”, “Risky” and “Undeliverable”, where risky covers disposable addresses, role addresses such as info@ and accept-all domains, and undeliverable covers “Invalid or fake email addresses” and “Typos & syntax errors”. ZeroBounce lists the same family: “Detect invalid, disposable, and inactive addresses” and “Identify spam traps, abuse emails, and catch-alls”.
The email platforms describe a check on the reader. Klaviyo’s help article titled “How to clean your email list to maintain good deliverability” says: “The process of list cleaning for deliverability involves 3 steps:” and the three are “Create a segment of unengaged contacts.”, “Exclude that segment from sends.” and “Suppress the Never engaged segment.” None of the three asks whether an address exists. Every one asks whether a person who receives the mail opens it, clicks it or buys.
Both are real list cleaning, but only one is measured by the number Gmail and Yahoo publish.
Google’s email sender guidelines set the limit in its own tool: “Keep spam rates reported in Postmaster Tools below 0.10% and avoid ever reaching a spam rate of 0.30% or higher.” Yahoo’s sender requirements set the same ceiling, “Keep your spam rate below 0.3%”, and add how it is counted: “Spam rate is calculated in our system based on mail delivered to the inbox”.
Read those two sentences together. Yahoo says its rate is counted on delivered mail, and a spam report in either inbox needs a message to report, so the rate both providers grade is made of complaints from people who received the message. An invalid address cannot complain, because nothing was delivered to it. Relvino’s reading: every complaint that counts comes from a mailbox that accepted the message, which is what an address check is built to confirm. Verifiers do flag some known complainers (ZeroBounce lists “abuse emails”, and Kickbox’s Threat Score marks addresses tied to complaints), but none can tell how one reader will respond to one brand. Verification lowers bounces; the complaint rate is built from addresses a validity check has no reason to reject.
The providers say what moves that rate. Yahoo, under a heading about removing invalid recipients: “Sending email to users who are not reading them, or who report them as spam, will harm your delivery metrics and reputation.” Google’s subscription guidance is shorter: “Consider unsubscribing recipients who don’t open or read your messages.” No verifier can act on that instruction, because the record of who opens a brand’s own mail sits with the platform that sent it.
Klaviyo’s sunset flow article says mailing inactive profiles “will lower your open rate, which is the most important signal used by mailbox providers to determine where future emails go (the inbox or the spam folder)”. Google’s page says “Google doesn't track open rates.” The practical reading: an open is the sender’s best sign that a person is still there, worth pairing with clicks, site visits and orders, while the number Gmail publishes is the complaint rate. Clean against the proxy; check the result against the grade.
The order protects reputation first and handles cheap fixes before expensive ones.
One caution on timing. Klaviyo’s cleaning article says brands “may choose to include unengaged contacts in certain high-priority campaigns, like new product announcements or Black Friday/Cyber Monday.” Bringing a dormant segment back is a reactivation, which Spamhaus warns provokes high complaint rates, so if it goes back in, it should go in small and early, as the Black Friday email marketing guide sets out.
All three describe the same core check, whether an address exists and accepts mail; the differences are in what each adds and how it charges. The table quotes each vendor about itself only, with prices as published on its pricing page on October 8, 2026.
Kickbox’s own home page concedes the limit of the category: “A valid address isn't always a safe one.” Its Threat Score and ZeroBounce’s Activity Data are signals about an address in general, not about how that person responds to one brand. For a store, real-time checking at the signup form usually matters more than bulk cleaning, because it keeps typos out instead of removing them later; all three offer an API for it.
Engagement-based cleaning runs on thresholds: in Klaviyo’s example, a profile counts as unengaged after five emails received and 90 days without an open or a click. Relvino moves that judgment to every message. For each shopper and each moment it decides whether to send at all, and on which channel, across email, SMS and pop-ups, so a person who has stopped responding is not mailed while waiting to cross a threshold. Relvino’s own claim is 80% less spam, lower send costs. The full comparison with a flow-based setup is in the Klaviyo vs Relvino guide.
A cleaning project is done when two numbers look right, each owned by a different tool.
A list can pass the first and fail the second. When it does, another verification run will not move it; the people to stop mailing are already on the list, valid and silent.
Start with the removals that are obligations, not choices: unsubscribes (Yahoo asks bulk senders to honor them within 2 days) and anyone who reported a message as spam through a provider feedback loop. Then let the platform drop hard bounces, verify any addresses it has never mailed, and only after that build the unengaged segment and run a short sunset. Doing it in that order means the riskiest addresses leave first.
It depends on where the bad addresses come from. If they arrive through signup forms, the deciding feature is a real-time API that rejects typos and disposable addresses at the form, before they reach the list. If the problem is an old or imported file, bulk cleaning with free handling of unknown results matters more. If the problem is complaints, no verifier is the answer; the sending platform’s engagement data is.
Free verification tiers are small. Kickbox says the first 100 verifications are free, ZeroBounce offers 100 free monthly verifications when signing up with a business or premium domain, and NeverBounce gives free credits at signup. The part that is free at any size is engagement-based cleaning inside the email platform: Klaviyo can create a Never Engaged segment from its deliverability dashboard, and Mailchimp can archive inactive contacts from a segment.
Find out which list and which address is listed, then fix the cause before asking for removal. Spamhaus says no one but Spamhaus decides what it lists or removes, and that the only way to be removed is to fix the problem that caused the listing. On a shared sending IP, the email platform usually handles it; Google’s sender guidelines tell shared-IP senders to make sure that IP is not on any blocklist.
Engagement cleaning is continuous rather than periodic: Klaviyo’s guidance is to keep excluding the unengaged segment from campaigns on an ongoing basis, with a shorter window for brands that mail daily. Verification is event-driven: run it before importing a list, before a migration, and before mailing any segment that has not been mailed for months. Google asks senders to monitor their spam rate in Postmaster Tools regularly, and that reading tells whether the routine is working.
About 30 minutes for the technical cutover: connect the Shopify store, point the sending domain, connect SMS, and shopper data ingests automatically. Bring the full suppression list across with it, so unsubscribed, bounced and complained addresses stay off. Revenue is then measured in a 14-day pilot, and from the first send Relvino decides per shopper whether a message should go at all.