← All articles

Cold Email Follow-Up Cadence: How Many Touches, How Far Apart

October 5, 2026 · 10 min read · by the 0effort team

TL;DR: For cold email follow-up, a sound default is three follow-ups after the first email, with gaps of about 3, 5 and 7 business days, then stop. Treat it as a budget: any opt-out or clear no ends the sequence, out-of-office pauses it, and a touch that earns fewer replies than its complaint risk gets cut.

Most cadence advice is a template: seven emails, every two days, copy it. We think that's the mistake. A cadence is a budget. Every touch spends goodwill and domain reputation and buys fewer replies than the touch before it. The useful question isn't "what's the magic number" but "when does the next email stop paying for itself, for this person?"

Why follow up on a cold email at all?

Because silence is ambiguous. An unanswered first email covers a prospect who never saw it, one who saw it on a bad day, one who meant to reply, and one who isn't interested. Only the last is a no. A follow-up is a second chance at the other three.

You'll see statistics online about how many sales need five follow-ups, or what share of reps quit after one. We don't use them: we couldn't trace them to a primary source. If a vendor publishes reply rates by touch, read it as that vendor's dataset (their year, their sample), and not a general truth.

So we'll print assumptions instead. Suppose 1,000 prospects and positive replies per touch of 20, 10, 6, 3 and 1. Those figures are invented for illustration; swap in your own. The shape is the point: each touch goes to roughly the same list but earns a fraction of the one before, so the fifth earns one reply per thousand sends, a twentieth of the first.

How many follow-ups should you send before you stop?

Three, as a starting default, for typical B2B outreach. That makes four emails in total, the last one a closing note.

Why not more? Take that fifth touch: one positive reply per thousand sends. Complaints are counted per message, and Google's bulk-sender guidance is to keep your spam complaint rate below 0.1% and never reach 0.3% in Postmaster Tools (Google email sender guidelines, accessed 2026-10-05; they apply to mail sent to personal Gmail accounts). 0.1% is one report per thousand messages. One irritated recipient cancels the touch's whole yield on paper (an illustration only: Postmaster's spam rate is a daily, domain-wide figure, not a per-touch one), and the damage lands on the domain, including your first emails.

Go past three follow-ups only when deal size justifies research per touch and you have something fresh to say, like a funding round or a relevant hire. That makes it a different email instead of one more bump. And no cadence guarantees a reply rate or inbox placement; neither does any tool, ours included.

How many days should you wait between touches, and why should the gaps widen?

Start with this and adjust. Every number below is a business day, not a calendar day:

• Business day 1: first email.
• Business day 4: follow-up one, a gap of 3 business days.
• Business day 9: follow-up two, a gap of 5 business days.
• Business day 16: follow-up three (the closing note), a gap of 7 business days.

Counting in business days means no touch lands on a weekend, and it stretches the calendar: start on a Monday and the closing note goes out on calendar day 22, three weeks later.

Why widen? The short early gap catches the "meant to reply" crowd while your email is still near the top of the inbox. The longer later gaps stop you reading as pressure and leave room for a real event, like a colleague returning, to happen in between.

We'd avoid daily emails in week one. We know of no public data showing they earn more, and a daily nudge starts to look like the pressure people report as spam.

What should each follow-up say so it adds something instead of "just bumping"?

"Just bumping this to the top of your inbox" gives the prospect two emails and still no reason to answer either. Every follow-up should carry one new thing.

The first can add proof: a relevant result, a named company like theirs. The second can shrink the ask, from "30 minutes?" to "worth a two-line reply?". The closing note should offer an easy exit and ask for a referral: "if this isn't yours, who handles it?" A wrong-person reply is one of the better outcomes a cold sequence produces.

Should follow-ups stay in the same thread or start a new one?

Keep follow-ups one and two in the original thread. The prospect gets the context one click away, and you aren't forcing them to reconstruct what you said. Mail clients thread on message headers, not just the subject line, so a genuine reply to your own message normally stays attached to the original.

Start a fresh thread with a new subject when you change the angle or re-contact someone after a long gap. Never fake a thread, though: "Re:" or "Fwd:" on a first email implies a conversation that didn't happen, and CAN-SPAM prohibits deceptive subject lines (FTC compliance guide, accessed 2026-10-05).

How should cadence change by deal size and market?

Cadence should follow the economics. A low-ticket SMB product can't afford much research per touch, so run the default above, maybe a day tighter. A high-ticket enterprise deal flips it: fewer emails per person, more research per email, and the extra effort goes to other people on the buying committee and other channels (LinkedIn, phone) rather than a fifth message to one inbox.

Market matters more than deal size. In the US, CAN-SPAM works by opt-out rather than prior consent: among other requirements, you must not use misleading headers or subject lines, must include a valid physical postal address, and must honor an opt-out within 10 business days (FTC compliance guide, accessed 2026-10-05).

The EU and UK differ from the US and from each other. In the EU, GDPR requires a lawful basis for processing a prospect's data (Regulation (EU) 2016/679, Art. 6, accessed 2026-10-05), and whether you may send the unsolicited email at all falls under the ePrivacy Directive (Directive 2002/58/EC, Art. 13, accessed 2026-10-05), which each member state implements in its own national law. Germany's implementation, for example, treats email advertising without the recipient's prior express consent as an unacceptable nuisance, outside a narrow existing-customer exception (UWG § 7, accessed 2026-10-05), so the right number of touches there can be zero. The UK applies PECR instead, whose rule on unsolicited marketing email is written for individual subscribers (PECR reg. 22, accessed 2026-10-05); UK GDPR still governs how you process a named person's data, corporate address or not (UK GDPR, accessed 2026-10-05). Our GDPR guide covers the per-country detail, and none of this is legal advice.

One rule we apply in every market: an opt-out or a clear "no" ends the sequence. We'd honor it the same day, not on day ten, even where the law allows longer.

Which prospect signals should pause, slow down, or end a sequence?

This is where per-prospect cadence beats per-campaign cadence. Here are the signals we'd act on.

That last row needs a caveat. Apple's Mail Privacy Protection stops senders from collecting information with invisible pixels (Apple, accessed 2026-10-05), and corporate security scanners often pre-fetch links and images, so "opened four times" can mean a robot rather than interest. Optimize for replies and positive-reply rate instead.

Out-of-office is where manual sequencing breaks first: someone has to read the return date, hold the sequence and restart it. 0effort, an autonomous AI SDR that sources buyers, runs email and LinkedIn touches and answers replies over email and phone, reads every reply itself. See how AI SDRs handle replies for the sorting logic.

At what point do extra follow-ups cost more in complaints than they earn in replies?

We can't give you a per-touch complaint number, and nobody honestly can without your own data. Reply yield falls with every touch. Complaint risk plausibly doesn't fall as fast, since the people still on touch four are the ones who ignored touches one to three. Where the lines cross depends on your list, your offer and your domain.

To find yours, watch your domain's spam rate in Google Postmaster Tools against the 0.1% mark, and break out unsubscribes and negative replies by touch number. Postmaster only reflects mail to personal Gmail accounts, and most B2B lists sit on Google Workspace or Microsoft 365 mailboxes, so for a corporate list the per-touch unsubscribes and negative replies are your main signal. If touch three produces more unsubscribes than positive replies, you have your answer for that segment. If the Postmaster number drifts toward 0.1% after you add a touch, cut the touch before you rewrite the copy. Our deliverability checklist covers the rest of the diagnosis.

How do you test and tune your own cadence instead of copying someone else's?

Change one variable at a time, and measure replies, not opens. Sample size is the catch. Suppose a 2% baseline positive-reply rate and you want to detect a lift to 3%. A standard two-sided test at 5% significance and 80% power needs roughly 3,800 prospects per arm. Those are textbook assumptions rather than your data, but they show why a 300-contact test says almost nothing about a one-point difference.

Three tests are worth running, roughly in this order:

1. Hold out the last touch. Send the closing note to half the list, then compare positive replies and unsubscribes after the same number of days.
2. Compare gap lengths. Run 3-5-7 against 4-7-10 on randomly split lists.
3. Test a new-angle fifth email against stopping at four, only where deal size justifies it.

Complaints are too rare to measure per arm, so watch them at domain level and use unsubscribes and negative replies per test.

FAQ

Is it rude or spammy to follow up on a cold email more than once?

Not inherently. What gets mail reported is following up with nothing new to say, ignoring a no, or a pace that reads as pressure. A few spaced, relevant follow-ups with an easy opt-out is ordinary sales practice. Compliance rules still apply and vary by country.

What is the best day and time to send a cold email follow-up?

There isn't a universal answer, and we'd distrust anyone selling one. Send during your prospect's working hours in their time zone and skip weekends. Beyond that, the claims about the perfect weekday and hour aren't backed by data we can trace, so test it on your own list.

Should I send a "breakup" email as the last touch, and does it actually work?

We'd send a closing note, but an honest one: say it's your last email, ask whether someone else owns the topic, and make "no" easy to say. Skip the guilt trip ("should I assume you don't care?"). We know of no reliable public data showing breakups beat other final touches, so test yours with a hold-out.

How long should I wait before re-contacting a prospect who never replied to the whole sequence?

We'd wait until something has changed, or roughly a quarter if nothing has, then write a new email with a new angle rather than restart the old sequence. Silence isn't an opt-out, but it isn't an invitation either. Anyone who said no or unsubscribed stays out for good. In the EU, GDPR Art. 5(1)(e) also says personal data may be kept no longer than necessary for its purpose (Regulation (EU) 2016/679, accessed 2026-10-05), so set a retention limit for non-responders.

Should a follow-up be shorter than the original cold email?

Usually, yes: two to four sentences, one new fact, one small ask, since the first email did the explaining. But length isn't the goal. A follow-up that adds a new fact can run longer.

Can I keep following up with someone who opened my email several times but never replied?

Not on that basis. Repeated opens can come from Apple Mail Privacy Protection or a security scanner, and even a real open isn't a request for more email. Stay inside your normal cadence, stop at the end of it, and treat a reply as the only signal worth acting on.