Why I am writing about email (and not personal finance) today
- In mid 2026, email deliverability is the small-business problem that keeps repeating – not because email is dead, but because inbox providers are stricter and more automated than ever.
- I keep seeing the same pattern: good businesses send decent content, but a quietly messy list drags everything down.
- This post is the routine we settled on after a couple of painful months: fewer bounces, fewer spam-folder surprises, and more stable results.
The pain point: nothing is “wrong” – yet results keep sliding
- Open rates drift down, but content quality did not change.
- Replies slow down, even from customers who used to respond.
- You run a promo and it performs fine one month, then falls flat the next.
- You look for a fancy fix: new subject lines, more personalization, different sending days.
- The unglamorous truth: inbox providers treat list quality like a signal of whether you are a responsible sender.
- If you keep mailing dead addresses, role accounts, and typo domains, you train the system to expect bounces and complaints.
- And once that reputation slips, every future campaign starts from a worse position.
The lesson we learned the hard way
- We assumed “small list” meant “safe.” We were wrong.
- A list of 3,000 can be riskier than 30,000 if it is old, imported from multiple places, and never cleaned.
- We also assumed unsubscribes were the main metric. They matter, but bounces and silent non-engagement matter just as much – sometimes more.
- If you want predictable email performance, treat list hygiene like bookkeeping: regular, boring, and non-negotiable.
The workflow we use: simple, repeatable, and measurable
- This is not a one-time cleanup. It is a loop.
- We run it monthly, plus a mini version after every large import (trade show list, partner referral list, old CRM export).
- The goal is not perfection. The goal is to keep “bad signals” from piling up.
Step 1 – Control how addresses enter your world
- We stopped accepting “email lists” as a single spreadsheet that floats around. Every source gets tagged.
- We keep a simple intake note: where it came from, when, and what consent means in practice.
- Why it works: when a campaign goes sideways, you can trace the damage to a source instead of blaming your whole program.
- Concrete example: we had a “customer list” that included receipts forwarded by an office manager. Many were temporary or mistyped addresses. That one segment caused most of our bounces.
Step 2 – Normalize and deduplicate before you do anything else
- Before validation, we normalize:
- Lowercase everything.
- Trim spaces.
- Split combined fields like “Name <[email protected]>” into separate columns.
- Deduplicate on email address, not on name.
- Why it works: verification and bounce handling are less useful if you keep multiple versions of the same contact. Also, duplicates are an easy way to annoy people and trigger complaints.
Step 3 – Verify addresses before sending (especially after imports)
- We use eMail Verifier for list checks before a new segment gets mailed.
- We do not treat verification as “permission.” It is only a technical check: does the mailbox look deliverable?
- Why it works: it prevents avoidable hard bounces, which are among the fastest ways to harm sender reputation.
- What we do with results:
- Invalid – remove immediately.
- Unknown – keep out of the main list and test cautiously, or request a fresh address from the customer.
- Role accounts (info@, sales@, support@) – we usually exclude unless the business relationship clearly expects it.
Example decision rule we use
(if you need something concrete):
If status = Invalid -> Suppress
If status = Unknown -> Hold for manual review
If role account -> Suppress unless explicitly opted in
If status = OK -> Eligible to mail
Step 4 – Send in a way that protects reputation
- This is where people jump straight to templates and subject lines. We focused on risk management first.
- We send to engaged contacts first, then expand.
- Why it works: engagement acts like a positive signal and can buffer you when you mail less-engaged segments.
- We use MaxBulk Mailer for campaigns where we want tight control over sending, segmentation, and content versions on a desktop workflow.
- We keep segments simple:
- Recent customers (last 12 months).
- Older customers (12-36 months).
- Leads who asked for quotes.
- Newsletter signups who never bought.
- Why segmentation works (in plain terms): you stop forcing one message to do all jobs. A quieter message to older customers can outperform a louder promo that feels irrelevant.
Step 5 – Process bounces every time, not “when we get to it”
- This was the step we used to skip. It was the step that mattered most.
- We now process bounces after every campaign, usually the next morning.
- We use eMail Bounce Handler to parse bounce messages and classify them.
- Why it works: bounces are feedback. If you keep mailing hard bounces, you are telling mailbox providers you do not maintain your list.
- Our handling rules:
- Hard bounce (user does not exist, domain does not exist) – suppress immediately.
- Soft bounce (mailbox full, temporary failure) – allow 2-3 attempts over time, then suppress if it repeats.
- Spam complaint – suppress immediately and review what was sent and to whom.
A small-business-friendly cadence
Day 0: Send campaign
Day 1: Process bounces, suppress hard bounces
Day 7: Recheck repeat soft bounces
Day 30: Monthly hygiene run (verify + dedupe + segment review)
Step 6 – Use a suppression list like a seatbelt
- We maintain a suppression list that includes:
- Unsubscribes.
- Hard bounces.
- Repeat soft bounces.
- Addresses that asked to be removed (even if they never used the unsubscribe link).
- Why it works: people accidentally re-import old contacts all the time. A suppression list prevents that one mistake from becoming a deliverability incident.
Step 7 – Keep acquisition honest (and boring)
- We stopped using “add everyone you met” behavior.
- Instead, we send a one-time follow-up to new contacts and ask them to confirm what they want.
- Why it works: the first email sets expectations. Clear expectations reduce complaints and improve engagement.
- If someone does not engage after that first follow-up, we do not keep poking them forever. We would rather have a smaller list that performs than a bigger list that harms us.
What changed for us (and what did not)
- What improved:
- Fewer hard bounces within two sends.
- More stable inbox placement – fewer “I did not get it” messages from regular customers.
- Less time spent guessing whether content or deliverability was the problem.
- What did not magically improve:
- If the offer is weak, hygiene will not save it.
- If you email people who did not ask for it, verification will not protect you from complaints.
- If you change your sending behavior wildly (huge spikes, long gaps), you still create risk.
A quick note on list building tools and scraping
- You can extract addresses from sources in a lot of ways. If you use eMail Extractor, treat it as a research and data-entry helper, not a license to spam.
- Why I am cautious here: unsolicited mail is where small businesses get into trouble fastest – legally, reputationally, and operationally.
- The practical middle ground we use: only mail people when we can explain (to them) why they are receiving the email and how to stop it.
The metrics we actually watch
- Hard bounce rate – should be very low. If it spikes, stop and investigate the source segment.
- Spam complaints – even a small number is serious for a small list.
- Reply rate (for relationship emails) – tells you if you are sending something humans want to respond to.
- Click-to-open rate (when links exist) – more useful than raw opens in 2026.
- Why these work: they connect to sender reputation and actual business outcomes, not vanity stats.
Common mistakes I would avoid if I were starting again
- Mailing an old list “just once” – that is often all it takes to cause a reputation dip.
- Ignoring soft bounces forever – repeated temporary failures are still a bad signal.
- Treating verification as consent – it is not.
- Over-segmenting too early – 4-6 segments you can maintain beats 25 segments you cannot.
Where to learn more (one link)
Checklist
- Tag every list source with date and consent context
- Normalize and deduplicate emails before importing anywhere
- Verify new imports and quarantine “unknown” results
- Send to engaged contacts first, then expand carefully
- Process bounces after every campaign and suppress hard bounces
- Maintain a suppression list and always apply it on imports
- Review metrics monthly, not only when something breaks
3 Actionable Takeaways
- Run a bounce-processing step the morning after every send – do not wait for a quarterly cleanup.
- Quarantine any list you did not personally watch being collected, and verify it before it touches your main audience.
- Keep your segmentation simple enough that you will still be doing it six months from now.
