Cold Outreach Copywriting Frameworks: The Complete Guide (2026)

Most cold messages drown in a sea of templates that all sound the same — same opener, same problem statement, same Calendly link. The reps who consistently book meetings have not memorized a magic line. They have internalized a handful of cold email frameworks that give their thinking a shape, then they fill that shape with specifics about the prospect in front of them. Cold email frameworks are scaffolding, not crutches. They tell you what order to think in, what to leave out, and where the message has to earn its next sentence. Once you stop staring at a blank page and start writing inside a structure that has been pressure-tested for decades, your reply rates start to behave. This guide walks through every framework that still earns its keep in 2026 — AIDA, PAS, BAB, PASTOR, STAR — plus value proposition, social proof, subject lines, CTAs, and how to adapt them to LinkedIn, voicemail, and video. Real examples throughout, no filler.

Why frameworks beat blank pages

A blank page asks two questions at the same time: what should I say, and how should I say it. Trying to answer both at once is why a draft takes you forty minutes and still sounds wooden. Frameworks separate the two. They lock the structure so your brain can spend its energy on the content. Pick PAS and you already know the message will be three beats long: a pain you have observed, why that pain compounds, what you propose to do about it. All you need to write is the specifics.

The second reason frameworks work is that they encode reader psychology. AIDA was lifted from print advertising in the late 1800s because that sequence — grab attention, build interest, create desire, ask for action — mirrors how people actually decide. PAS does the same thing for problem-aware buyers. BAB is built for vision-led pitches. PASTOR layered on storytelling because dry logic does not move people on its own. None of these are arbitrary. They are residue from millions of dollars of split tests across a century of direct response.

Third, frameworks make iteration possible. If you write every email from scratch, you cannot tell what is working. If every email follows AIDA, and your reply rate dips, you can isolate the variable: was it the attention line, the desire paragraph, the CTA? Frameworks turn copy into something you can debug.

The risk, of course, is that frameworks become formulas. PAS without specifics reads as accusation. AIDA without restraint reads as a billboard. The fix is to treat the framework as the skeleton and the prospect's reality as the muscle. The rest of this guide is about how to put real meat on these bones.

AIDA — Attention, Interest, Desire, Action

AIDA is the oldest framework on this list and still the most common in cold outbound. It works best when the prospect is not yet thinking about the problem you solve, because it is built to wake them up first and sell them second.

Attention is one line that earns the next one. Not a clever subject line — that is a separate problem we cover later. The attention line is the first sentence of the body. It must be specific, short, and about them. "Saw your team just opened a second location in Lyon" beats "Hope you are doing well."

Interest turns the attention into a relevant insight. You are not pitching yet. You are saying: I noticed something, and here is why it matters in your world. This is the part most reps skip, which is why their AIDA emails feel like a jump cut from greeting to pitch.

Desire is where the offer enters, framed around a future state the prospect actually wants. You are not describing your product. You are describing the result of using your product, in their language.

Action is one ask. One. Not "happy to send a deck, hop on a call, share a case study." A single, concrete next step.

A short example for an SDR selling a review-management tool to a multi-location dental group:

Subject: New Lyon clinic — quick question

Hi Marc — saw the team just opened the Part-Dieu clinic. Most multi-site dental groups I work with see review volume on a new location lag the older sites by 6 to 9 months, which slows local search ranking right when ad spend is highest. We help groups like Dental One and Sourire Plus close that gap inside the first 90 days by automating the post-visit ask without adding work for the front desk. Worth a 15-minute look next week, or should I send the Sourire case study first?

That is roughly 70 words. Attention (Lyon clinic), Interest (the 6 to 9 month lag), Desire (close the gap inside 90 days, named peers), Action (a single question with two binary options). It does not try to sell the meeting twice. It does not promise the moon. It hands the prospect a small, easy decision.

PAS — Pain, Agitate, Solve

PAS is the framework you reach for when the prospect already knows they have a problem. You do not need to wake them up. You need to prove you understand the problem better than they do, and that you have a credible answer.

Pain is one sentence naming a problem the prospect feels. You are not guessing — you are reflecting back something you have evidence of, ideally something specific to their company.

Agitate is where the message earns its keep. Most reps skip this step because it feels uncomfortable. Done badly, agitation is fear-mongering. Done well, it is honesty about second-order consequences. The pain is not the bug — it is what the bug costs by the end of the quarter.

Solve is the resolution. Short, confident, with one piece of proof. Then the ask.

A full example for a rep selling field-service software to plumbing companies:

Subject: 3-day delays mentioned in your reviews

Hi Sarah — pulled a few recent Google reviews for Acme Plumbing and three of them mention waits of three days or more between the call and the technician showing up. That is the kind of feedback that does two things at once: it kills your conversion on Google Maps because new prospects read the last 5 reviews before they call, and it pushes your CSAT below the threshold where the home warranty partners renew. We help mid-size plumbing operators tighten dispatch by routing emergency tickets first and giving the dispatcher a real-time view of crew availability — Drainline cut their average response from 52 hours to 9 hours in the first quarter. Open to a 15-minute walk-through Thursday or Friday morning?

The pain is concrete (three reviews, three-day delay). The agitation names two consequences the prospect cares about (Maps conversion, warranty partner renewal). The solve is one sentence with one named proof point. The ask is binary and specific. PAS only works when the pain you cite is real — and the cleanest way to surface real pain at scale is to read the prospect's own reviews. We come back to that in the MapsLeads section.

BAB — Before, After, Bridge

BAB is the simplest framework on this list and the easiest to abuse. The structure is: here is your world today (Before), here is the world you want (After), here is how you get there (Bridge). It works because human beings buy futures, not features. The risk is that you make the After sound like a brochure.

Before is the current reality, named in the prospect's own language. Not "you are wasting time on manual entry" but "your dispatcher is rekeying every ticket from Jobber into the accounting system after 6 pm."

After is the future state, framed in outcomes that matter to the role you are writing to. An ops manager cares about throughput and cost per job. A founder cares about cash conversion and customer experience. Pick one outcome and make it vivid.

Bridge is your offer, but only the smallest credible version. You are not asking them to commit to a full implementation. You are showing them the first step.

A full example for an outbound rep selling a scheduling tool to medspas:

Subject: From a 9-tab front desk to one screen

Hi Lina — right now the front desk at most medspas the size of yours is juggling 9 tabs to take a single booking: Vagaro for the calendar, Stripe for the deposit, ManyChat for the SMS confirmation, Google Sheets for the no-show tracker, and so on. Imagine instead that the receptionist takes the call, types the client name, and a single screen handles the slot, the deposit, the confirm-by-SMS, and the rebook reminder — and the no-show rate drops below 4 percent because deposits are taken by default. The way clients move from the first picture to the second is a 30-day pilot on one location: we connect the existing tools, run side-by-side for two weeks, then switch the front desk over. Ten medspas in Paris are running on it now. Worth a quick look?

BAB earns its place when the gap between the current and the desired state is large enough to be motivating, but the bridge is small enough to be believable. If the After is utopia and the Bridge is "schedule a 60-minute discovery call with three stakeholders," the prospect will not buy the leap.

PASTOR — Problem, Amplify, Story, Transformation, Offer, Response

PASTOR is the longest framework on this list and the most complete. It comes from direct response copywriting and is built for situations where the offer is bigger, the consideration cycle is longer, or the prospect needs a story to internalize the change. In cold email it tends to live in the second or third touch, after a short PAS or AIDA opener has earned attention. It also works well in LinkedIn DMs and on landing pages that follow a cold message.

Problem is the same as PAS — name a real pain.

Amplify is the second-order consequences, but with more room than PAS allows. Two or three sentences instead of one.

Story is the move that PASTOR adds. You do not pitch — you tell a short story about another customer who was in the same place. The story is what makes the rest of the message land, because human beings believe narratives in a way they do not believe claims.

Transformation is the After state, but described as the result of the story. You are showing, not telling.

Offer is a clean description of what you are proposing.

Response is the ask, narrowed to one option.

An example for a rep selling a B2B onboarding platform to SaaS founders:

Subject: When trial-to-paid stalls at 12 percent

Hi Tom — most B2B SaaS products I see sit somewhere around 12 percent trial-to-paid conversion, and that number does not move no matter how much they tune the trial length or the pricing. The real problem is that 70 percent of trial users never reach the second activation moment in the first 7 days, which is the only window where habits form. When that window closes, no email sequence will reopen it. Earlier this year a workflow tool called Loopstack came to us at 11 percent. We rebuilt their onboarding into a six-step Figma-style guided flow that triggered the second activation moment inside the first session, and we ran a 30-day cohort. Their trial-to-paid moved to 21 percent, and their CAC payback dropped from 14 months to 6. The transformation was not new content — it was sequencing what already existed so that the user finished the second step before they closed the tab. We package this as a 4-week onboarding sprint: audit, redesign, ship, measure. Open to a 20-minute walk-through next Tuesday?

PASTOR is dense by design. If you write it well, the prospect feels they have read a small case study, not a sales pitch. If you write it badly, it reads as a wall of text. The discipline is to keep each section to its job — Problem is one line, Amplify is two, Story is three, Transformation is one, Offer is one, Response is one. Roughly 180 to 220 words total.

STAR — Situation, Task, Action, Result

STAR is borrowed from behavioral interviewing and is the framework most people forget about, even though it is the cleanest way to deliver a case study inside a cold message. STAR works particularly well when you are reaching out to a senior buyer who has seen every other framework a hundred times. It feels less like a pitch and more like a peer telling them what happened on another account.

Situation is the context — which company, what stage, what they were trying to do.

Task is the specific objective.

Action is what was done. Concrete steps, in order.

Result is the outcome, with numbers.

A full example, used as the body of a cold email to a VP of Sales:

Subject: How we took Aerospike from 1.4 to 3.7 SQLs per rep

Hi Priya — quick story rather than a pitch. Aerospike came to us last year with a 1.4 SQL per rep per month average across an 18-person SDR team — solid pipeline, but not enough to support the new ARR target. The task was clear: double SQL output without doubling headcount, inside two quarters. We built a prospecting layer on top of their existing tools that pulled local commercial accounts within 50 km of every named-account headquarters, ran a Reputation enrichment pass to surface accounts with public service complaints, and routed those into a dedicated PAS sequence written off the actual review keywords. The result, by month four: 3.7 SQLs per rep per month, no new headcount, and a 22 percent close rate on the new lane versus 14 percent on the named-account lane. Would 20 minutes next week be useful to walk through how the lane was built? Happy to share the full breakdown either way.

STAR works because it is honest about the structure. The reader knows from sentence two that they are reading a case study, and they will read it anyway, because the situation is specific enough to be real. There is no manipulation, just a story with numbers attached.

The Value Proposition Canvas adapted for SDRs

The Value Proposition Canvas was built for product teams, but it is the single best diagnostic tool for cold copy. The canvas asks two questions: what does the customer want, and what do you offer that delivers it. In SDR work the canvas is simpler — three columns instead of two halves.

The first column is the prospect's jobs to be done. Not their job title — the actual outcomes they are measured on. A VP of Sales is hired to grow pipeline, hit quota, and reduce ramp time. A franchise owner wants more bookings, fewer no-shows, and higher repeat rate.

The second column is their pains. The friction, costs, and risks that block those jobs. Pains are concrete: "the SDR team is missing quota by 18 percent," "no-show rate is 22 percent and rising," "the warranty partner is threatening non-renewal." Generic pains do not move anyone.

The third column is their gains. The wins beyond just removing pain. Status, optionality, growth headroom, time back. Pains and gains are not opposites — pains are about avoiding loss, gains are about creating upside.

Once you have the three columns, your value proposition writes itself: a single line that names the most acute pain, promises the most relevant gain, and points at how you deliver both. The canvas is not the email — it is the worksheet that produces the email. Spend 20 minutes filling it in for a single ICP and the next 50 emails write themselves.

A common mistake: confusing your features with the prospect's gains. A feature is what the product does. A gain is what the prospect gets when the product does that thing. "Real-time dispatch routing" is a feature. "Crews arrive 4 hours faster on emergency calls, which means the warranty partner renews" is a gain.

Social proof in cold outreach — what works, what backfires

Social proof is the most under-used and over-used element in cold copy at the same time. Reps under-use it when they list logos no one cares about. They over-use it when they pile three case studies into one message and the email starts to read as a brochure.

The version that works is narrow, comparable, and recent. Narrow means the proof is from a company in the same segment, size, and geography as the prospect. A 12-person plumbing operator does not care that you closed Vodafone. They care that you helped Drainline, who they can google and recognize. Comparable means the metric you cite matches a metric the prospect tracks. Cycle time, conversion rate, average ticket size — yes. Vague claims like "increased ROI" — no. Recent means inside the last 12 months. Older proof reads as you living in the past.

Three patterns that work:

• The named peer. "Two operators on the same street as you — Sourire Plus and Dental One — moved from 6 to 23 reviews a month inside the first quarter."
• The numbered result. "Cut average dispatch time from 52 hours to 9 hours in 90 days." One number, in context.
• The specific quote. A single sentence pulled from a real customer, with a name and a role. Not five bullet points.

Three patterns that backfire:

• The logo wall. Listing eight customer names with no context. Reads as desperate.
• The unverifiable stat. "300 percent ROI." If the prospect cannot picture how it was measured, it lowers your credibility.
• The borrowed authority. "As featured in Forbes." No buyer has bought because of a Forbes mention since 2009.

A useful test: read your social proof line out loud. If it sounds like something a rep would say, cut it. If it sounds like something a customer would say to a peer over coffee, keep it.

Subject lines — 8 tested patterns

The subject line has one job: get the email opened. Not summarize the body, not pitch the offer, not be clever. Open. Short, lower-case, specific, and slightly mysterious tends to outperform every other style in 2026 inboxes.

Eight patterns that consistently outperform:

• The one-word hook. "Lyon?" or "dispatch?" Short enough to feel like a colleague.
• The named question. "quick question about Acme dispatch" — names the company, signals specificity.
• The recent-event reference. "Saw the Lyon opening" — works only when true.
• The number prompt. "From 1.4 to 3.7 SQLs" — invites curiosity.
• The peer reference. "How Drainline cut dispatch time" — pulls in social proof at the open.
• The friendly correction. "small thing on your reviews" — warm, low-key, opens because it sounds personal.
• The introduction frame. "intro from a fellow operator" — works for executive-level outreach.
• The blunt offer. "30 mins for an extra 10 SQLs" — works in mature ICPs that respond to directness.

Avoid all-caps, double exclamation marks, the word "exclusive," and emoji. Avoid mismatch between subject and body — if the subject promises a question and the body is a pitch, the prospect will mark you as spam the next time too.

CTAs that convert (interest-CTA vs ask-CTA)

The single most common mistake in cold copy is asking for the meeting too soon. There are two CTA styles, and most reps default to the wrong one.

The ask-CTA is direct: "Open to a 15-minute call Thursday at 10 or 11?" It works when the prospect already has a reason to say yes — late in a sequence, with a warm referral, or after they have engaged with content.

The interest-CTA is softer: "Worth a look, or not relevant?" or "Want me to send the case study first?" It works in the first or second touch, when the prospect has no reason to give up calendar time yet. The interest-CTA asks for permission, not commitment. Permission is cheap. Commitment is expensive.

A good rule: in touches 1 and 2, use interest-CTAs. In touch 3 onward, switch to ask-CTAs. The interest-CTA opens a thread. The ask-CTA closes one. Mixing them up costs reply rates in both directions.

A final detail on CTAs: never offer more than one option. "Want a call, a deck, or a case study?" gets ignored because choice is friction. Pick the most likely next step for that prospect at that stage and ask only for that.

Tone of voice — what fits your ICP

Tone is not a personality trait. It is a function of who you are writing to. The same offer, written for a 12-person plumbing operator, will land flat if you use the tone you would use for a CFO at a 2,000-person SaaS company.

Three axes to think about:

• Formality. Operators, trades, and franchise owners respond to plain, direct language. Senior corporate buyers expect a slightly more measured register, but never stiff.
• Density. Technical buyers — engineers, RevOps, IT — accept and reward density. Non-technical buyers will skim. Match the length of your sentences to the reader's tolerance.
• Confidence. Cold copy tolerates a surprising amount of confidence, as long as it is backed by specifics. Hedging language ("just wanted to," "if it might be possible") signals weakness and gets cut.

A useful exercise: pick five recent customer emails (real ones, from your actual customers) and underline the tone markers. Use those markers in your cold copy to the same ICP. Your prospects will sound more like your customers than like the rest of your inbox.

Personalization at scale — what's worth personalizing

Personalization is the single most over-promised and under-delivered idea in cold outbound. The honest answer in 2026: personalize the opener, mostly do not personalize the body, and sometimes personalize the CTA.

Opener: yes. The first 1 to 2 sentences must be specific to the prospect. This is the single highest-ROI piece of personalization you can do. A relevant opener doubles or triples reply rates compared to a generic one.

Body: mostly no. The body is your value proposition, your social proof, your offer. Those are framework-level decisions, not prospect-level. If you find yourself rewriting the body for every prospect, your ICP is too broad. Tighten the segment and the body becomes shared.

CTA: sometimes. If the prospect's role or company stage changes the most likely next step, the CTA should reflect that. A founder gets "worth a quick look?" An IT director gets "happy to send the security one-pager first." But in most ICPs the CTA is shared.

The way most reps fail at personalization is by trying to do it everywhere. The way the best reps win is by doing it in one place — the opener — and doing it well, with real signal pulled from real public data.

Adapting frameworks to LinkedIn, voicemail, video

The five frameworks above were written for cold email, but they translate cleanly to other channels with small adjustments.

LinkedIn DMs punish length and reward warmth. Cut the framework in half. PAS becomes one line of pain, one line of agitation, one line of soft solve. AIDA becomes one sentence per beat, total of four. The CTA shifts from "book a call" to "open to a quick chat?" because LinkedIn is a relationship channel before it is a transaction channel.

Voicemail is brutal. You have 20 to 30 seconds before the prospect deletes. PAS works best because it grabs by stating a real pain in the first 5 seconds. AIDA works if the attention line is razor-sharp. Always end with your phone number twice — once at the start, once at the end. Always reference one specific thing about their business so they know it is not a robocall.

Video is its own discipline. Use BAB or STAR. The first 3 seconds must include their first name and one visible signal that the message is for them — a screenshot of their website, a map of their location, the inside of their store. Keep videos under 60 seconds. End with a single ask, then drop a written summary in the body for skimmers.

How MapsLeads data fuels personalization

Frameworks tell you what shape to write. The specifics that fill those shapes have to come from somewhere, and in the local-business and SMB world, the highest-signal source is the prospect's own Google Maps presence. MapsLeads is built around exactly this workflow.

You start with Search: a query plus a city. "Plumbers in Lyon" or "dental clinics in Marseille" returns the local universe of businesses with their Maps data. That is your raw list. From there, you decide which enrichment modules to enable.

The two modules that matter most for copywriting are Reputation and Photos. Reputation pulls the keywords that come up most often in the prospect's reviews, the average rating, and a sample of recent review snippets. Photos pulls the visual context — interior, exterior, team — which is gold for video openers and for understanding what the business actually does.

Concrete workflow: you enable Reputation on a list of 200 plumbing operators, and the export comes back with a column listing the most-mentioned terms in their last 30 reviews. You filter for accounts where "delay" or "late" or "rescheduled" appears two or more times. That is your PAS list. Saw a recent review of Acme Plumbing mentioning 3-day delays — the BAB framework writes itself: Before is "your customers are waiting three days," After is "same-day dispatch on emergency tickets," Bridge is your offer. The opener writes itself too: "Saw three of your last 20 reviews mention waits over three days." That single line is more personalized than anything ChatGPT will produce, because it is grounded in something the prospect already knows is true.

Credits callout: 1 cr Base, +1 Contact Pro, +1 Reputation, +2 Photos. You only pay for the modules you use, so most teams run a Reputation-only enrichment for copywriting prep and skip Photos until they move to video. Groups, dedup across cities, and CSV/Excel/Google Sheets export keep the workflow clean. Wallet and billing are the same shared credit pool, so you can scale a campaign up or down without re-buying lists. Real review keywords beat ChatGPT-flavored guesses every time.

Common copywriting mistakes

A short list of things that kill reply rates, in rough order of damage done:

• Starting with "I". The first word should be about them, not you.
• Using "Hope this finds you well". It signals template. Cut it without replacement.
• Pitching in the first sentence. Earn the next line first.
• Listing more than one social proof. Two or more reads as desperate.
• Asking for a 30-minute call in the first touch. Calendar time is expensive. Ask for permission first.
• Jargon mismatch. Using your internal product terms instead of the prospect's industry terms.
• Three-paragraph wall. If the body is over 130 words on a first touch, it loses.
• Generic close. "Looking forward to hearing from you" is not a CTA. It is filler.
• Mismatched subject and body. Subject promises a question, body is a pitch — instant trust loss.
• AI tone. Em-dashes everywhere, "delve," "navigating the landscape," "in today's fast-paced world." All immediate deletes.

Copywriting checklist

Before you send, run the message through these checks:

• Is the first sentence about the prospect, not me?
• Is the pain or insight specific enough that it could not be sent to 100 other companies?
• Is there exactly one social proof, and is it comparable to the prospect?
• Does the body fit in under 130 words for first touch, under 200 for follow-ups?
• Is there one CTA, and is it interest-style on touch 1, ask-style by touch 3?
• Does the subject line match the body?
• Would you open this email if you received it?
• Is there anything in the message that ChatGPT would have written by default? Cut it.

FAQ

What is the AIDA framework?

AIDA stands for Attention, Interest, Desire, Action. It is a four-step copywriting framework borrowed from late-1800s print advertising, used to move a reader from not thinking about a problem to taking a specific next step. In cold email, AIDA works best when the prospect is not yet aware of the problem you solve, because it is built to grab attention first and earn interest before introducing the offer. Each step is short — typically one to two sentences — and the action at the end is a single, concrete next step.

What is PAS in copywriting?

PAS stands for Pain, Agitate, Solve. It is a three-beat framework used when the prospect already knows they have a problem, so the message does not need to wake them up — it needs to prove you understand the problem better than they do and offer a credible answer. The Pain is one sentence naming a real, specific issue. The Agitate is two sentences naming the second-order consequences. The Solve is one sentence introducing your answer with one piece of proof, followed by a single ask.

What is the best cold email opener?

The best cold email opener is one specific, recent, and prospect-relevant observation in 8 to 14 words. "Saw the Lyon opening last week" beats "Hope you are doing well" every time. The opener should not contain your company name, your offer, or a question yet — its only job is to earn the next sentence. The fastest way to write good openers at scale is to pull a real signal from public data — a recent review, a new location, a hiring spike — and reflect it back in one line.

How long should a cold email be?

A first-touch cold email should land between 70 and 130 words. Anything shorter feels thin. Anything longer reads as a pitch deck and gets skimmed or skipped. Follow-ups can stretch to 180 to 220 words when you are using a denser framework like PASTOR, but only if every sentence is doing real work. The single best length test: count the sentences. Eight or fewer for a first touch. If you have more than that, cut until you do.

How do I personalize cold emails at scale?

Personalize the opener, mostly do not personalize the body, and sometimes personalize the CTA. The opener is the highest-ROI piece of personalization — it should be specific, recent, and grounded in real public data about the prospect. The body is your shared value proposition and social proof, which should not change per prospect; if it does, your segment is too broad. The CTA can shift based on role or stage but is shared inside a tight ICP. Tools like MapsLeads make scaled personalization realistic by pulling real review keywords, ratings, and snippets you can paste into the opener.

How many follow-ups should a cold sequence have?

Most modern cold sequences run 4 to 7 touches across 14 to 21 days, blending email with LinkedIn and an occasional voicemail. The first touch uses an interest-CTA. The second adds a different angle on the same pain. The third introduces social proof. The fourth narrows to an ask-CTA. Anything past touch 5 should either change channel or stop. Sending a tenth email to a prospect who has not replied is not persistence — it is noise.

Next steps

Frameworks are how you stop writing from scratch and start writing from a structure that converts. The other half of the equation is the data you fill them with — and that is where most teams cut corners. Good frameworks plus generic data still produce generic emails. Good frameworks plus real review snippets, accurate locations, and verified contacts produce messages prospects actually reply to.

If you want to see how this works end-to-end, the related guides cover the rest of the stack: cold email Google Maps leads walks through running cold campaigns off Maps data, sales prospecting with Google Maps covers the prospecting workflow, and Google Maps reviews signal buying intent explains how to turn review data into a qualification signal. When you are ready to run it on your own ICP, pricing lays out the credit model and get started opens a wallet so you can pull your first list today. Pick one framework, pull one ICP, write one sequence — then let the replies tell you what to tune next.