How to write cold emails that work in 2026

What changed in cold email between 2024 and 2026

The cold email playbook from 2022 does not work anymore. Three shifts changed the practice between 2024 and 2026, and any team writing cold email today has to account for all three.

Shift 1: AI saturation. The cost of writing a "personalized" cold email dropped from $5 per email (human research) to $0.05 per email (AI). Every B2B prospect now receives 40 to 80 cold emails a week, most of them AI-generated. Inboxes have adapted. Buyers triage faster, skim more, and pattern-match harder. A cold email that looked personalized in 2022 looks like a template in 2026.

The implication: the bar for "personalized" has moved. A token-replacement opener ("Hi Sara, I saw you work at Acme") no longer reads as personalization. It reads as automation. The bar is now research-grade openers: a sentence that proves someone (human or AI) read something specific and recent about the prospect.

Shift 2: Deliverability standards. In February 2024, Gmail and Outlook tightened authentication requirements. Any sender exceeding 5,000 messages per day to their networks now needs SPF, DKIM, and DMARC. Without them, the message is throttled or rejected. The change broke programs overnight, especially programs sending from a single primary domain.

The implication: deliverability is no longer a niche concern. It is the floor of the program. The best message in the world goes to spam if the authentication is broken. We will cover deliverability briefly here, but the deeper version is in our deliverability playbook.

Shift 3: The attention floor dropped. The average B2B knowledge worker spends 30 to 90 seconds on email triage in the morning, processes 100 to 300 messages, and replies to under 5 percent of them. Cold emails that take 25 seconds to read get deleted. The new bar is the 8-second skim: the opener, the ask, and one proof point have to register inside 8 seconds. Anything longer requires the prospect to slow down, and prospects do not slow down.

The implication: length, density, and visual structure matter more than they did. 200-word cold emails are dead. 75-to-125-word cold emails work. The structural compression is what makes the message fit inside the attention floor.

The 2026 anatomy of a high-reply cold email

A cold email that works in 2026 has five parts. Each part has a job, and each part is one sentence (the exception is the problem statement, which is one or two sentences).

1. Subject line. 3 to 7 words. Lowercase. Specific to the prospect. The job is to get opened.
2. Opener. One sentence. References something specific and recent about the prospect. Proves the sender did the work.
3. Value bridge. One to two sentences. The problem you solve, phrased in the buyer's language, with one specific proof point.
4. Ask. One sentence. A yes/no question with low friction.
5. Signature. One line. Name and role. Nothing else.

Total length: 75 to 125 words. Total reading time: 8 to 15 seconds. Total time to write (with research): 5 to 10 minutes per email if you do it by hand, 30 to 90 seconds per email if you use a competent AI for the opener.

That structure is the entire game. Everything that follows is depth on each component.

Subject lines: 3 to 7 words, lowercase, specific

Subject lines are misunderstood. Most cold email writers treat them as the first sentence of the pitch. They are not. The subject's only job is to get opened. The body's job is to get a reply.

The subject patterns that work in 2026:

• Topic + company: "outbound at acme", "deliverability at acme", "sdr math at acme". Reads like an internal email, not a vendor pitch.
• First name + topic: "sara, quick one", "marcus, your q1 hire", "priya, sdr question". The first-name prefix triggers attention because most automated email uses the formal "Mr." or full name.
• Specific reference: "your post on 4-step sequences", "your hire last month", "saw your stack on builtwith". The specificity is the open rate driver. Generic subject lines like "quick question" still work, but worse than they did three years ago.
• Direct question: "still doing outbound in-house?", "open to a 15-min call?". Direct questions get opened at higher rates than statements, but they have to feel earned, not pushy.

The subject patterns that get filtered or ignored:

• All-caps anywhere. Triggers spam filters. Reads as desperate. Drop.
• Emoji. Performance dropped sharply in 2024 as inbox providers began down-weighting subject-line emoji. The lift is gone; the penalty remains.
• Re: or Fwd: prefixes on cold emails. Buyers learned the trick. It reads as deceptive now and can be a CAN-SPAM violation in the United States.
• Personalization tokens with no context. "{first_name}, an idea" feels like an automation, not a person. Token-only subject lines under-perform any specific subject.
• Length above 50 characters. Mobile preview clips anything longer. The truncation usually clips the part that mattered.
• Clickbait. "You will not believe this" or "the secret to outbound" reads as marketing email and gets ignored.

The subject line writing test: would a colleague write this to you about your work? If yes, ship it. If it sounds like a vendor pitching, rewrite.

The opener: three patterns that beat "I see you're at {Company}"

The opener is where most cold emails lose the prospect. A generic opener signals automation, and the prospect stops reading before the ask. The opener exists to prove research, not to sell.

Three opener patterns that work in 2026:

Pattern 1: The recent signal

Reference something the prospect's company did in the last 30 days. A hire, a launch, a funding round, a public initiative, a quote in the press.

Working examples:

• "Saw your Series B announcement last week, and the team hire pattern suggests you are rebuilding outbound in the next quarter."
• "Noticed your new VP RevOps started 6 weeks ago. Most teams in that transition rebuild the SDR motion in their first 90 days."
• "Your product launch on Tuesday led with the partner-integration story. Same pattern we are seeing across the category."

What makes these work: they cite something the prospect can verify in their own head. They suggest a thesis (the implication of the signal) rather than just acknowledging it. They do not flatter; they observe.

Pattern 2: The peer comparison

Reference what comparable companies are doing or have done. The prospect is in the peer set, and the opener positions the email as a conversation among peers, not a vendor pitch.

Working examples:

• "Three of the last five fintech CFOs I talked to all had the same problem with revenue forecasting after their Series B. Worth seeing if you do too."
• "Saw [Comparable Company] just shifted from 5-step to 4-step sequences last month. The data on email 5 was their main reason. Curious where you land on it."
• "Most 50-to-200-person SaaS teams hit a wall at 8 meetings per week per SDR. Wondering if you are running into the same math."

What makes these work: they position the prospect inside a peer set rather than treating them as a generic target. The peer comparison invites the prospect to either confirm or correct, both of which are replies.

Pattern 3: The content callback

Reference something the prospect themselves published: a LinkedIn post, a podcast appearance, a conference talk, a tweet, a public Substack.

Working examples:

• "Your post on shifting from 5-step to 4-step sequences. The data on email 5 surprised me too. Have a hypothesis on why email 3 is still the high-yield slot."
• "Heard you on the SaaStr podcast last week, the segment on inbound-to-outbound conversion specifically. The 90-day math you cited matches what we are seeing."
• "Your tweet thread on AI SDR adoption from Q3. The part about reply triage being the bottleneck is the right diagnosis."

What makes these work: they prove the sender actually consumed the prospect's content. The verification is unmistakable, and the prospect cannot dismiss the email as automation.

What does not work

Openers that signal automation rather than research:

• "I see you work at {Company}." Tells the prospect nothing they did not already know.
• "Hope this email finds you well." Wastes a sentence and signals template.
• "I came across your profile on LinkedIn." Says you used a tool, not that you did research.
• "Congrats on the recent [funding round / hire / promotion]." Generic, gets used by every other vendor, reads as flattery rather than observation.
• "My name is Nitish and I am the founder of ReachIQ." The prospect can read your signature. Burning the opener on self-introduction is a wasted sentence.

The test: read your opener out loud. If a colleague writing to you about your work would write it, ship it. If it sounds like a vendor template, rewrite.

The 75-to-125-word constraint and why it works

Cold emails between 75 and 125 words consistently outperform shorter and longer ones in the data we see. The boundaries are not arbitrary. Below 75 words, the email usually skips one of the four required elements (opener, problem, proof, ask). Above 125 words, the email exceeds the prospect's attention floor and gets glazed.

The numerical breakdown of the 75-to-125-word range:

• Opener: 15 to 25 words.
• Problem statement: 20 to 40 words.
• Proof point: 15 to 30 words.
• Ask: 8 to 15 words.
• Signature: 4 to 8 words.
• Greeting and sign-off: 4 to 10 words.

The 75-to-125-word constraint forces every word to do work. If a sentence is not serving one of the four elements, it gets cut. Most cold emails get longer because the writer hedges, adds a feature mention, or stacks two asks. Each of those instincts produces a worse email.

One specific failure mode: stacking proof points. Writers who have multiple wins want to mention all of them. "We worked with X and Y and Z and got results A and B and C." This reads as bragging and burns the proof slot. Pick one proof point. The most specific one, ideally with a named comparable.

How to use research to write a 1-sentence opener

The opener is where research lives. The right research process takes 3 to 5 minutes per prospect by hand, 30 to 90 seconds with AI. Either way, the inputs are the same.

Five research inputs, in priority order:

1. LinkedIn activity: The prospect's last 30 days of posts, comments, and likes. Their public LinkedIn shows what they are thinking about, hiring for, or building.
2. Company news: A Google search for "[Company name] news" filtered to the last 30 days. Funding rounds, hires, product launches, customer wins, press mentions.
3. 10-K or annual report: If the prospect's company is public, the most recent 10-K names the strategic priorities. The "Risk Factors" section names the problems they actually worry about. The "MD&A" section names the metrics they actually track.
4. Podcast or conference appearances: Search the prospect's name plus "podcast" or "conference" or "panel." A 30-minute appearance gives you 30 specific things to reference.
5. Tech stack: BuiltWith, Wappalyzer, or Apollo's tech-graph data. The stack tells you what categories they have already invested in, which tells you what gap your product fills.

Across the five inputs, you only need one usable signal. The opener writes itself once you have that signal.

The workflow that works at scale (with or without AI):

1. Pull the prospect's LinkedIn URL and company URL into the research input.
2. Pull the last 30 days of company news and the last 30 days of LinkedIn activity.
3. Identify one usable signal: a specific thing the prospect or their company did in the last 30 days.
4. Write the opener: one sentence that names the signal and implies a thesis.
5. Check the opener against the test: does it prove research, or does it read as automation?

AI personalization done well vs done badly

AI changed the economics of cold email personalization. The bad version of AI personalization has flooded the inbox and trained buyers to filter it out. The good version of AI personalization is functionally indistinguishable from human research and lifts reply rates to 5 to 12 percent. The difference is in the prompt design and the input quality, not in the model.

Bad AI personalization

What it looks like in production:

Hi Sara,

I noticed you work at Acme Corp and saw your impressive background on LinkedIn. As a fellow professional in the B2B SaaS space, I wanted to reach out about something I think could really help Acme grow.

At ReachIQ, we help companies like Acme accelerate their pipeline through AI-powered personalization and intelligent sequencing. We have worked with hundreds of B2B SaaS companies and have helped them achieve incredible results.

Would you be open to a quick demo this week to see how we can help Acme reach its sales goals?

Best regards,
Nitish

Why this fails:

• The opener says nothing specific. "I noticed you work at Acme" is true of every prospect on the list.
• The "impressive background" is filler. Buyers learned to filter on it.
• The product description is a feature list, not a problem statement.
• "We have worked with hundreds" is volume without context.
• The ask is "a quick demo," which is a high-commitment ask phrased as low-commitment.
• The whole email reads as a template. It would not change if the prospect were named Marcus at Bravo Corp.

The reply rate on this email pattern is 0.5 to 1.5 percent. The cost to send is low, but the cost to the program's deliverability over time is high because the spam complaint rate is elevated.

Good AI personalization

Same prospect, same product, different prompt design:

Hi Sara,

Saw Acme just brought on Marcus as VP Sales six weeks ago and started hiring 4 SDRs in parallel. Most 50-to-200-person B2B SaaS teams that do that hit a wall around 8 meetings per week per SDR in the second quarter, and it is usually research time, not selling time, that breaks the math.

We worked with a Series B fintech in your category who went from 14 to 38 meetings per week in 90 days by moving the research layer to AI and keeping the SDRs on conversations.

Worth 15 minutes next Tuesday to walk through what we did?

Nitish

Why this works:

• The opener references two specific signals (the VP Sales hire, the SDR hiring) and implies a thesis (the math breaks at this scale).
• The problem statement is in the buyer's language: meetings per week, research time, selling time. Not "pipeline acceleration."
• The proof point is specific: a named comparable, a number, a timeframe.
• The ask is low-friction: 15 minutes, a specific day, yes/no answerable.
• The email would change substantially for a different prospect because the signals are specific.

Length: 95 words. Reading time: under 15 seconds. Reply rate target on this pattern: 5 to 12 percent. The prompt difference between the two emails is not the model. It is the input: feeding the AI the LinkedIn activity, the company news, and the hiring data, then asking for a single-sentence opener that names the signal and implies a thesis. The bad email feeds the AI a token; the good email feeds the AI research.

Three AI personalization rules

1. Feed the AI research, not tokens. Pull the prospect's last 30 days of LinkedIn activity, the company's last 30 days of news, and any public signals. The AI cannot personalize on data it does not have.
2. Constrain the output to one sentence. The opener is one sentence. Anything longer dilutes the personalization signal and starts to look like padding.
3. Audit a sample weekly. Spot-check 20 emails per week. If the openers are reading as template, the input pipeline is broken or the prompt is too loose. Fix it at the input layer.

The platform side of this is covered by hyper-personalization: research inputs, prompt design, and output constraints handled at the platform level so the per-email work is research-grade.

Follow-up email patterns: touches 2, 3, 4, and 5

80 percent of meetings come from email 2 through 5 in a sequence. A program that sends one email and moves on is leaving 4x its pipeline uncollected. Each follow-up email has its own job, and writing them as "bumping this up" or "in case you missed it" is the most common failure mode.

Touch 2: The reframe

Job: approach the same problem from a different angle than email 1. If email 1 was about pipeline volume, email 2 is about deliverability. If email 1 was about cost, email 2 is about quality. If email 1 was a peer comparison, email 2 is a recent-signal opener.

Working template:

Hi Sara,

One more thread on this. The reason most teams in your category hit the SDR wall I mentioned is not the SDRs themselves; it is the research-to-send ratio. 60 percent of an SDR's day in 2024 was researching prospects. In 2026, that should be 10 percent.

Worth 15 minutes Tuesday to walk through what shifted?

Nitish

Length target: 60 to 90 words. Tone: less introductory than email 1, more conversational. Reference email 1 implicitly without re-summarizing it.

Touch 3: The proof point

Job: deliver the strongest proof point in the sequence. Email 3 is consistently the highest-reply email in our data because prospects who were warming convert here. Lead with a named comparable, a specific number, and a timeframe.

Working template:

Hi Sara,

One concrete example. Karlo runs RevOps at a Series B fintech (50 employees, growing 8 percent month-over-month). They had the same SDR math problem you might be running into. Last quarter they went from 14 meetings per week to 38 by shifting the research layer to AI and keeping the 4 SDRs on conversations and reply triage.

Karlo wrote up the whole thing on LinkedIn last month if useful. Otherwise, worth 20 minutes to walk through the exact setup?

Nitish

Length target: 80 to 110 words. This is the only email in the sequence allowed to run slightly longer because the proof point is doing the work.

Touch 4: The pattern break

Job: do something different from the previous three touches. Most sequences read as four versions of the same email by touch 4. The pattern break re-engages the prospect by changing the format.

Working pattern breaks:

• The short one: "Sara, two questions. Are you in market for outbound tooling in Q1, and if so are you looking at the platform layer or the managed-service layer? 90 seconds is enough."
• The honest one: "Sara, going to be straight with you: I have sent three emails and not heard back. Two options. Either this is not the right thread, in which case let me know and I will close it out. Or you have been busy. If it is the latter, want me to send a 90-second Loom showing the approach so you can watch it whenever?"
• The thesis one: "Sara, building a thesis on why outbound at the 50-to-200-employee SaaS tier is breaking specifically in Q1. Three patterns I am seeing: the SDR math, the deliverability collapse, the reply backlog. Curious which of those (if any) maps to your situation. Worth 15?"

Length target: 50 to 80 words. The pattern break is shorter than the proof email because the format itself is doing the lifting.

Touch 5: The breakup

Job: give the prospect a low-friction yes/no decision. The breakup email is consistently the second-highest reply driver in the sequence. The "close the loop" framing implies finality without finality and triggers more "actually, let me see what you have" replies than any other format.

Working breakup templates:

• "Sara, should I close the loop on this one? Happy to either drop it or send a 2-paragraph summary if there is a real chance it is useful."
• "Closing out unless you tell me otherwise. If now is not the right week, no worries; can check back at the end of the quarter."
• "Mind if I close the loop here? If outbound is on the roadmap for next quarter rather than this one, just let me know and I will circle back."

Length target: 40 to 70 words. The breakup email is the shortest email in the sequence. It is also the most effective on a per-email basis.

What does not work in the breakup slot:

• "Bumping this up" or "in case you missed it." Reads as automation. Reduce reply rate.
• "Did not hear back, let me know." Passive. Does not invite a decision.
• "Last note from me on this." Works, but less well than "close the loop."
• "Final follow-up." Sounds threatening. Avoid.

Reply handling templates

Half the program is writing the emails. The other half is handling what comes back. Four reply types account for 95 percent of the volume: positive, soft positive, objection, ghost. Each one has a template that works.

Positive: "yes, when works?"

Action: book the meeting, send confirmation, get out of the way. The prospect already said yes. Do not re-pitch.

Hi Sara,

Tuesday at 10am Pacific or Wednesday at 2pm Pacific work? Calendar link if easier: [link]. Will send a 1-line agenda once you confirm.

Nitish

Length: under 30 words. Time to send: under 1 hour. Do not include a calendar invite that auto-blocks the time before they confirm; that reads as presumptuous.

Soft positive: "tell me more"

Action: give them the next layer of detail and 2 specific time options. Do not send a giant info pack; that buries the meeting ask.

Hi Sara,

Quick version: ReachIQ is the platform, Done-for-You SDR is the managed service. Most teams in your spot start with the platform and add the managed-service layer when they want to move faster on reply handling. The Karlo case study I mentioned was a platform-only setup.

Want to grab 20 minutes Tuesday at 10am Pacific or Wednesday at 2pm Pacific to walk through which one fits? Either way, will send a 90-second Loom of the platform tonight so you have a visual.

Nitish

Length: 80 to 120 words. Time to send: under 4 hours. Always include two specific time options; open-ended scheduling reduces reply rate.

Objection: "we already use X"

Action: acknowledge the objection, name the dimension on which you differ, do not knock the incumbent. Most objections are not real "no" replies; they are tests.

Hi Sara,

Makes sense, [Competitor] is the right call for a lot of teams in your tier. The dimension where teams usually move to ReachIQ is the reply triage layer: most platforms send well but leave reply handling to the human SDR, which is the operational bottleneck at your scale.

If you are looking at that specifically, worth 15 minutes to compare notes. If not, totally fair; will close the thread.

Nitish

Length: 70 to 100 words. Time to send: under 24 hours. Never bash the incumbent; the prospect chose them, and bashing the choice means bashing the prospect.

Ghost: no reply across the full sequence

Action: the breakup email is the ghost handler. Past the breakup, move the prospect to a 90-to-120-day nurture cycle rather than re-sequencing them immediately.

Why not re-sequence: a second sequence to prospects who completed the first sequence without replying produces 1.0 to 1.5 percent reply rate, versus 5+ percent on fresh prospects. Re-sequencing the same list is operationally expensive and pipeline-weak. The 90-to-120-day nurture lets the prospect's situation change (new role, new fiscal year, new initiative) before you try again.

The nurture touch at day 90 to 120:

Hi Sara,

Circling back. A few months ago I reached out about [topic]. Two things have changed since then: [specific change 1, specific change 2]. Wanted to see if either of those moves it back onto the list, or if Q3 is still not the right quarter.

If still not the right thread, just let me know and I will leave it.

Nitish

Length: 60 to 90 words. Reply rate target: 3 to 6 percent. Lower than fresh-list, but materially higher than re-sequencing.

A/B testing methodology

Most A/B testing in cold email is noise. The tests that produce signal are the tests run at the sequence level, on samples of 200+ prospects per variant, over a 14-day window. The tests that produce noise are the tests run on subject lines or sign-offs, on samples of 50 prospects, over 3 days.

What to test:

• Sequence length. 4 emails vs 6 emails. Often the 4-email variant wins on reply quality even if 6-email wins on reply volume.
• Channel mix. Email-only vs email plus LinkedIn vs email plus LinkedIn plus phone. The lift from adding channels is usually significant enough to detect at modest sample sizes.
• Opener pattern. Recent signal vs peer comparison vs content callback. Some categories favor one pattern over another.
• Proof point type. Numbers vs named comparables vs case-study-style anecdotes. The right one varies by audience.
• Breakup framing. "Should I close the loop?" vs "Closing out unless you tell me otherwise." Both work; which one wins varies by audience.

What not to test (settled questions or noise):

• Subject-line capitalization. Lowercase wins. Move on.
• Emoji. Settled. Do not use.
• Sign-offs (Best, Thanks, Cheers). Noise.
• Send time within working hours. Effect is real but small. Test once, then settle.
• Single-word changes inside the email body. Below the noise floor at sample sizes you can run.

The right cadence for A/B tests: run two tests per quarter, each at the sequence level, each with a 200+ prospect sample per variant. Read the results at 14 days. Hold a winner for one quarter before testing the next variable.

Founder voice templates: no salesy language

Founder-led outbound (the founder themselves sending the email, not an SDR) outperforms SDR-led outbound in early-stage B2B by 1.5 to 2.5x reply rate. The reason is partly title (a founder reaching out gets attention) and partly voice (founders write differently than SDRs).

The founder voice is plainer. It does not pitch. It observes, names a thesis, and asks a small question. The patterns that distinguish founder voice from SDR voice:

• "I" instead of "we." "I have a thesis about why this is breaking in your category" beats "we have helped many companies solve this." The first sounds like a person; the second sounds like marketing.
• Specific over universal. "The Karlo case I mentioned" beats "our many satisfied customers." Specifics are founder-grade; universals are vendor-grade.
• Honesty about the cold-outreach origin. "This is a cold email; you can ignore it if it is not the right week" outperforms pretending the relationship is warm. Buyers reward honesty.
• Short asks. "Worth 15?" beats "would you be open to scheduling a 30-minute discovery call to see how we can help your team achieve their sales goals?"
• One concrete thing per email. One signal, one problem, one proof, one ask. No stacking.

A founder-voice cold email template:

Hi Sara,

Saw your Series B announcement last week and the 4-SDR hire in parallel. Building a thesis on why teams in your tier hit the meeting wall in their second quarter; the data is pointing at research time, not selling time, as the bottleneck.

Got 15 minutes next Tuesday or Wednesday to compare notes? I run ReachIQ, so I am partial, but I will be more useful than promotional.

Nitish

Length: 80 words. The "I am partial, but I will be more useful than promotional" line is the founder-voice signature: honest about the bias, confident about the value.

Mistakes that kill cold email programs in 2026

Across the programs we have reviewed, the recurring mistakes are operational, not stylistic.

Sending from the primary domain. One bad day and the CEO's email is in spam. Always use a secondary domain. The setup guide is in our deliverability playbook.

Skipping warmup. A new domain sending 100 emails on day one gets flagged within a week. Warm for 14 to 21 days, then ramp.

Generic opener. "Hope you are doing well" or "I see you work at [Company]" signals automation. Replace with one specific reference.

Multiple asks per email. Asking for a meeting and a webinar signup and a content download. The prospect does nothing.

Feature lists in the body. Features are for the demo. The cold email is about whether the prospect has the problem.

Pushy ask. "I would love to schedule 30 minutes to walk you through our offering" reads as desperate. Lower the threshold.

Heavy signature. Logos, banners, three phone numbers, embedded videos. Strip everything but name and role.

Sending the same email to everyone. Even templates need ICP-segment variants. Finance buyers need different language than sales buyers.

No follow-up. 80 percent of meetings come from email 2 through 5. Single-email programs leave 4x their pipeline uncollected.

Treating "I'll think about it" as positive. It is not. Tag it as a nurture and move on.

Re-sequencing silent prospects after 30 days. Produces 1.0 to 1.5 percent reply rate. Move silent prospects to a 90-to-120-day nurture instead.

Skipping the breakup email. Second-highest reply driver in the sequence. Sending 4 emails and going silent leaves 10 to 15 percent of pipeline uncollected.

How ReachIQ thinks about cold email

The platform side of ReachIQ runs the volume work: list discovery, enrichment, per-prospect personalization, sending, and reply triage. The Done-for-You SDR layer handles the conversations that need a human: the objection replies, the soft positives that need a real follow-up, the meetings that need a human voice on the line.

The configuration that works for most B2B teams in 2026: AI personalization at the platform layer (research-grade openers, not token replacement), human SDRs on the harder replies, and a tight reply triage SLA that closes the gap between "yes, when works" and a booked meeting. The combination is what makes the unit economics work.

The features that matter most for cold email specifically: hyper-personalization for the opener layer, automated sequences for the follow-up cadence, email health for the deliverability floor, lead discovery for the upstream list, and meetings for the calendar booking on positive replies.

For teams that want the platform without operating it, the Done-for-You SDR service runs the entire cold email program on the client's behalf. For teams that have an SDR but want them out of the volume work, the platform alone covers the orchestration. Pricing depends on configuration; the pricing page has the breakdown.

Glossary cross-references

Term definitions live in the glossary: cold email, personalization, sequence, cadence, deliverability, ICP, SDR, AI SDR, sender domain, domain warmup, intent data.

FAQ