Why Are My Marketing Emails Going to Spam? Who to Send To and How Often

Direct answer: Marketing emails land in spam when too many people who have stopped engaging receive too much mail too fast. Sticky Digital recommends sending default campaigns to people who clicked, bought, or visited your site in the last 30 days, adding the 60-day and 90-day groups at a lower frequency, and never sending a sale to the full list on day one. Read results by inbox provider, not in total, because Apple inflates opens and Gmail does not report spam complaints. For a Black Friday send on November 27, 2026, the rewarm starts about four weeks out, which is this week.

What we see across accounts

The deliverability problems we get called into rarely start with a bad email. They start with a calendar. A brand has a revenue gap, a sale, or a launch, and the list on the send gets bigger than the list on the last send.

We have run Klaviyo for DTC brands in beauty, wellness, and subscription consumables long enough to know the pattern, so we checked it. We pulled 90 days of Klaviyo deliverability data from Hiro, our analytics platform, for three accounts. The accounts are anonymized here. Figures cover July 30 to October 28, 2026 unless noted.

One account is a wellness brand with about 412,000 delivered emails in the window. One is a beauty brand with about 17.4 million. One is a subscription consumables brand with about 11.4 million. The sizes differ by a factor of forty, and the same three behaviors show up in all of them. Apple opens look great. Outlook behaves differently from Gmail. Bounces hide in flows.

Who should you send to? The 30, 60, and 90 day model

Start with how you define active. Do not define it by opens. Apple Mail Privacy Protection preloads messages, so an Apple open can happen with no human anywhere near the inbox. Define active by actions: a click, an order, a checkout started, or a visit to your site.

30 days: the default audience

Every campaign goes to this group first. These are people who did something in the last month, and they are the ones whose behavior tells inbox providers your mail is wanted. If you only have the capacity to get one segment right, build this one.

60 days: added on a schedule

People who engaged 31 to 60 days ago get your stronger sends: a launch, a restock, a real offer. They do not get every campaign. Think of this tier as a lower-frequency lane, not a bigger version of the first.

90 days: the last tier added

The 61 to 90 day group joins only for sends with a real reason to return, and only after the first two tiers have held steady. Past 90 days, move people into a winback or sunset flow and suppress them from campaigns if they do not respond.

The windows are a starting point. A brand with a long repurchase cycle, such as home goods, should stretch them. A replenishment brand with a 30-day cycle should tighten them. The principle holds either way: the further someone is from their last action, the less often they should hear from you.

How often should you email? Set it by segment, not by brand

Most frequency advice gives one number for the whole list. That number is wrong for most of the list, because a 30-day buyer and a 90-day lapsed reader are two different audiences.

Judge each segment on revenue per recipient, unsubscribes, and spam complaints together. Klaviyo’s deliverability score treats an unsubscribe rate of 0.1% or lower as excellent, and a spam complaint rate of 0.005% or lower as excellent. Our three accounts ranged from 0.13% to 0.36% on unsubscribes and 0.008% to 0.010% on spam complaints. All three sat well under the ceiling Google sets for bulk senders, which is a spam rate below 0.10% in Postmaster Tools and never 0.30% or higher.

Staying under Google’s line is the minimum. If you raise frequency and a segment’s unsubscribe rate roughly doubles within a few sends, step back for that segment. Do not apply the cut to the whole list.

Why are your emails hitting spam? How to read each inbox provider

Apple: opens are inflated, clicks are not

Apple open rates were 85.1% in the wellness account, 82.7% in the beauty account, and 78.0% in the subscription account. Gmail ran 47.2%, 55.8%, and 33.0% in the same accounts. That is 1.5 to 2.4 times higher on Apple. Click rates told the opposite story: Apple’s were at or below Gmail’s in all three. Treat Apple opens as a delivery confirmation, not an engagement signal.

Outlook and Hotmail: watch the gap to Gmail

Our working line is 34%. When Outlook and Hotmail opens fall under 34% on an account where they normally run higher, we start asking questions about placement. The more useful comparison is Outlook against Gmail on the same account. In the subscription account, Outlook opens were 34.2% and Gmail’s were 33.0%, so that account is running normally at a low baseline. In the wellness account, Outlook opens fell to 28.5% in May while Gmail held at 46.3%. That gap is the signal.

Outlook also reports spam complaints, and it bounces more. In the beauty and subscription accounts, Outlook bounce rates were 0.52% and 1.38%, against 0.11% and 0.33% at Gmail.

Gmail: the silent one

Gmail does not report spam complaints to Klaviyo, so a complaint problem there shows up as falling opens first. In the wellness account, Gmail opens dropped from 56.5% in February to 24.0% in March. Nobody sent a complaint notice. The opens just went away.

Why sending to the full list fails

The most common deliverability failure we see is a volume jump. It is not a bad subject line, a broken link, or a missing unsubscribe. It is a send that reaches several times more people than the one before it.

The wellness account is the clearest case in our data. Delivered volume on a trailing 30-day basis went from about 74,000 in February to about 334,000 in March, more than four times higher. In March, Gmail opens were 24.0%, Outlook opens were 14.9%, Outlook bounces were 2.0%, and Klaviyo’s deliverability score fell to 46, which is the poor band. In April, volume fell to about 27,000 delivered and the score recovered to 74. By the end of May, with volume back near 154,000, it sat at 76.

The recovery came from sending to fewer people, not from rewriting the emails. A sender who can get back to a healthy score by shrinking the audience was probably never short on creative.

Why a bounce-removal webhook belongs in your setup

Bounces concentrate in flows. In the beauty account, the flow bounce rate was 5.28% while the campaign bounce rate was 0.15%, about 35 times lower. Flows are where new signups land, and new signups include typos, throwaway addresses, and bots.

Klaviyo handles part of this on its own. The gap a webhook fills is everywhere else. A webhook that listens for bounce events and removes or suppresses that address across your other systems keeps your SMS platform, subscription tool, and store records from mailing or counting an address that cannot receive mail. It also gives someone a log to review, so a spike in flow bounces gets noticed as a pattern instead of a rounding error.

How to rewarm before a sale

If your biggest sale is Black Friday on November 27, 2026, you have about four weeks. Here is the sequence we use as a default.

• Week 1: Send only to the 30-day engaged segment at your normal cadence. Check that authentication is clean, review Postmaster Tools, and clear known bounces.
• Week 2: Add the 60-day segment to your two strongest sends.
• Week 3: Add the 90-day segment to one send. Compare Outlook opens to Gmail opens on the same send. If Outlook falls well below Gmail, or the Postmaster spam rate moves toward 0.10%, hold the tier where it is.
• Week 4: Run early access to the engaged tiers. The 90-day group joins on the sale itself only if the previous three weeks held.

Rewarming is not a cure for a list that has already tripped a threshold. In the wellness account, the score recovered after volume came down, not after volume went back up. If your Outlook and Gmail opens are already apart, shrink the audience first and rewarm after.

How Sticky Digital approaches who, how often, and why spam

We run this as a system, not a one-time fix. It is part of the email and SMS programs we manage for DTC brands.

• An engaged 30-day default segment built on clicks, orders, and site activity rather than opens.
• Scheduled 60-day and 90-day adds with a lower frequency and a stop rule if Outlook diverges from Gmail.
• A provider-level read of every account in Hiro, so Apple, Outlook, and Gmail are never blended into one open rate.
• A bounce audit of flows before campaigns, with a webhook that removes bounced addresses from connected systems.
• A rewarm calendar set four weeks ahead of the biggest send of the year, with a hold rule at each step.

See more on Klaviyo deliverability and retention in our blog, or meet the team behind the accounts.

FAQ

Why are my marketing emails going to spam all of a sudden?

A sudden drop is usually a volume or audience change. Check whether a recent send reached many more people than the one before it, and whether Outlook and Gmail opens diverged. In one wellness account we manage, delivered volume rose more than fourfold in a month and the Klaviyo deliverability score fell to 46. It recovered to 74 the next month after volume dropped to about 27,000.

How often should I send marketing emails?

Set frequency by segment. People who engaged in the last 30 days can take your normal cadence. People who engaged 31 to 90 days ago should hear from you less often and only for stronger sends. Track revenue per recipient, unsubscribes, and spam complaints together, and keep your spam rate below Google’s 0.10% target.

Are Apple Mail open rates accurate?

No. Apple Mail Privacy Protection preloads emails, so opens register without a person reading. In three Klaviyo accounts we checked, Apple open rates were 1.5 to 2.4 times Gmail’s while Apple click rates were at or below Gmail’s. Use clicks, orders, and site activity to judge engagement.

How long does it take to rewarm a list before Black Friday?

Plan on about four weeks. Start with the 30-day engaged segment, add the 60-day group in week two, test the 90-day group in week three, and run early access to the engaged tiers in week four. For a November 27, 2026 sale, that means starting now.

Should I send more emails if my email revenue is flat?

Not before you check who you are sending to. More volume to an audience that has stopped engaging tends to lower opens at Gmail and Outlook before it raises revenue. Fix the segments, then test frequency on the engaged tiers.

Brands that want this audit run on their own Klaviyo account can start here.

Article By: Mariel Kilroy, Co-Founder, Sticky Digital

Mariel Kilroy is the Co-Founder of Sticky Digital, a retention marketing agency specializing in email, SMS, loyalty, and subscription growth for DTC brands.