Kampaign.ai logoKampaign.ai

The complete guide

How to build a highly targeted campaign

A highly targeted campaign reaches only the accounts that fit your profile and are showing a buying signal right now — each message personalized to that account’s context. Here is the full method, stage by stage.

LIVE — signals detected on target accounts
FUNDINGSeries B closed · revenue team forming94
HIRING5 SDR roles posted in 7 days82
REGULATORYNew FDA rule affects product line88
EXEC CHANGENew CRO · budget reallocation likely79
Definition
A highly targeted campaign is outbound outreach directed only at accounts that match an ideal customer profile and are showing a current buying signal, with each message personalized to the account’s specific context. It is the opposite of volume outbound, which sends near-identical messages to a large list regardless of timing or fit.

STAGE 00 · GET READY

Domain management and deliverability

Domain management is the prep work that decides whether your campaign reaches the inbox at all: registering a separate sending domain, creating a dedicated sender account on it, authenticating that domain, and warming it up before you send. A perfectly targeted campaign is worthless if the messages land in spam.

Targeting controls who you reach and when. Deliverability controls whether you reach them. This stage comes first because every later stage assumes your email actually arrives — and for cold outbound, that is never automatic. It has to be engineered before the first message goes out.

Why use a separate sending domain

Use a dedicated sending domain — often called a secondary or burner domain — so cold outreach never puts your primary company domain at risk. Cold sending can attract spam complaints, and complaints damage sender reputation. If that reputation lives on a throwaway domain, your real domain and its everyday business email stay protected.

  • Buy a close variant of your main domain. If you send from yourcompany.com, register yourcompany.co, getyourcompany.com, or tryyourcompany.com for outbound. Recipients still recognize the brand; your primary domain stays clean.
  • Isolate the risk. If a secondary domain’s reputation is ever harmed, you retire it and spin up another — without touching the domain your company runs on.
  • Spread volume across domains. High-volume programs use several secondary domains, each carrying a fraction of the sending, so no single domain or mailbox sends enough to look abusive.

Why a dedicated sender account

On each sending domain, create a dedicated sender email account rather than sending from a real person’s everyday mailbox. This contains outbound activity, keeps reputation and volume per mailbox controlled, and lets you authenticate and warm each account on its own.

  • One account, one job. The sender account exists only for outreach, so its sending pattern stays consistent and easy for mailbox providers to trust.
  • Contained reputation. If one account runs into trouble, it doesn’t drag down the rest of the team’s email.
  • Keep per-mailbox volume low. Many small senders outperform one large one. Splitting volume across several dedicated accounts protects deliverability.

Authenticate the domain

Before sending, set up the three authentication records that tell mailbox providers your mail is legitimate. Without them, even well-written email is likely to be filtered.

RecordWhat it does
SPFDeclares which servers are allowed to send mail for your domain.
DKIMAdds a cryptographic signature proving the message wasn’t altered and really came from your domain.
DMARCTells receivers what to do when SPF or DKIM fails, and reports on who is sending as your domain.

Warm up and test before launch

Email warm-up is the practice of gradually raising a new mailbox’s sending volume over two to four weeks — starting at a handful of messages a day and building up — so providers learn the sender is real. A brand-new domain that suddenly blasts high volume looks like spam and gets treated like it.

  1. Register and authenticate the secondary domain and its dedicated sender account (SPF, DKIM, DMARC).
  2. Warm up gradually — begin with a few sends per day per mailbox and increase steadily over 2–4 weeks, including some genuine back-and-forth replies.
  3. Run an inbox-placement test — send to seed addresses across major providers to confirm whether messages land in inbox, promotions, or spam.
  4. Fix and re-test — if messages land in spam, adjust authentication, volume, or content and test again before launching.
  5. Keep volume sustainable — hold per-mailbox daily sends at a safe level and monitor reputation continuously once live.
The rule of thumbNever launch from a cold domain. A new sending domain needs authentication plus 2–4 weeks of warm-up and a passing inbox-placement test before it carries a real campaign. Skipping this is the most common reason targeted campaigns underperform — the targeting was fine, the mail just never reached a human.

STAGE 01

Define the target

Targeting starts by writing down two things: who fits (your ideal customer profile) and what makes now the moment (the trigger conditions worth acting on). A campaign is only as targeted as the definition behind it.

Most outbound is untargeted because the list is built once and frozen. A targeted campaign treats the target as two filters applied together: a stable fit filter and a live timing filter. An account has to pass both to enter the campaign.

What to specify

  1. Ideal customer profile (ICP) — industry, company size, business model, and the role you sell to. Keep it narrow enough that a message can be specific.
  2. Trigger conditions — the events that say an account is worth contacting today: a funding round, a hiring spike, a regulatory change. These become the signals you monitor in stage two.
  3. Disqualifiers — conditions that remove an account even with a strong signal, such as an existing competitor contract or a recent loss.
Why it mattersThe narrower and more event-specific the definition, the more personalized every later stage can be. Vague targets force generic messages, which is what makes most outbound fail.

STAGE 02

Detect buying signals

A buying signal is a public or observable event indicating an account may be ready to buy — funding, hiring, an executive change, a regulatory shift, or a tech-stack change. Signals open a short window, usually 48–72 hours, when outreach is most likely to land before competitors arrive.

Targeting is timing. The same account that ignores you in a quiet quarter will reply within a week of closing a funding round or hiring a new revenue leader, because a real problem just became urgent and budget just appeared. Detection is the work of catching that moment automatically rather than checking by hand.

The main signal types

SignalWhat it indicatesTypical play
FundingNew budget and a mandate to grow; teams forming fast.Reach the new buyer before the org fills its stack.
Hiring spikeA function is scaling — e.g. five SDR roles means the revenue team is expanding.Contact the VP who owns the function being built.
Executive changeA new leader, often within 90 days, reallocating budget and tools.Multi-contact reach to the new exec and their directs.
Regulatory changeA new rule creates an immediate, dated problem for affected companies.Map the affected accounts and contact them the same day.
Tech-stack changeAdoption or removal of a tool signals an active project or a gap.Position against the adjacent tool or the newly exposed gap.
The windowSignals decay. After 48–72 hours the account is usually saturated with competing outreach and response rates drop sharply. Detection only pays off if it feeds action the same day.

STAGE 03

Score and prioritize accounts

Scoring assigns each account a number from 0 to 100 based on the strength and recency of its signals combined with how well it fits the ICP. Accounts are then ranked so reps always work the hottest first, while weaker accounts enter nurture rather than being discarded.

Detection without ranking just produces a longer list. A score turns signals into a queue: it tells a rep what to do next and why, and it gives them something defensible in a deal review. The two inputs — signal strength and fit — are multiplied, not added, so a strong signal on a poor-fit account doesn’t jump the queue.

How a score is built

  1. Fit component — how closely the account matches the ICP from stage one.
  2. Signal component — the type, strength, recency, and number of signals firing at once. Several signals together score higher than any one alone.
  3. Threshold bands — set action tiers: high (act now), medium (nurture and watch), low (hold). A common cut is 75+ for immediate, hands-on outreach.
Make it visibleA useful score carries its reasoning. “82/100” means little; “82 — hiring spike plus exec change in the last 14 days, exact ICP fit” tells the rep what to say.

STAGE 04

Research account context

Research gathers the specific facts about each account and contact that make outreach relevant instead of generic: why now, what changed, who to reach, and what they likely care about. This is the raw material personalization is built from.

The difference between targeted and generic outreach is almost entirely here. A message that references the actual event — the round that closed, the rule that just passed, the role being hired — reads as written for that account, because it was. Research turns a detected signal into a reason to reach out that the recipient recognizes as true.

What to gather per account

  • Why now — the signal in plain language: what happened and why it creates a need.
  • The stakeholders — the three to seven people whose work the signal touches, and how each relates to it.
  • The angle — what each role likely cares about: executives want strategic framing, managers want operational detail, product leaders want validation.
  • The next move — the single most relevant thing to propose, not a generic ask for a meeting.

STAGE 05

Draft personalized multi-channel outreach

A targeted campaign coordinates email, LinkedIn, and phone across the relevant stakeholders, sequenced over several days rather than fired all at once. Each channel and each role gets a tone suited to it, all drawn from the research in stage four.

One channel is a touch; a coordinated sequence is a campaign. The point of multi-channel is presence without nagging: an email, then a LinkedIn connection, then a call script, then a follow-up, spaced so the account feels a consistent, relevant approach rather than a single ignorable message.

A typical coordinated sequence

  1. Email #1 — sends immediately. References the signal directly, 90–120 words, one soft call to action.
  2. LinkedIn connection — 24 hours later. Conversational, peer-to-peer, question-based, no links.
  3. Call script — 48 hours later. Talking points built from the signal context, not a cold pitch.
  4. Follow-up email — 72 hours later. Adds a second angle or a relevant proof point.
  5. Optional SMS or final touch — 96 hours later, where appropriate.

Match the message to the role

  • Executives — strategic framing tied to the business outcome the signal implies.
  • Managers — operational detail on how the change affects their team’s work.
  • Product / technical leaders — specifics and validation rather than positioning.

Worked example — regulatory signal

When a new FDA food-dye rule was announced, the affected companies could be identified within a day, scored by exposure, and contacted the same afternoon — each message referencing the specific rule and its impact on that company’s products. Reaching them inside the window, while the problem was fresh and competitors were still building lists, is what produced an open rate far above the outbound norm.

Signal detected 127 accounts mapped Scored by exposure Sequenced same day Meeting booked

STAGE 06

Measure signal-to-revenue

Measure the full chain: signal → touch → meeting → pipeline → revenue. Tracking the whole path shows which signals and which messages actually produce revenue, and feeds that back so the scoring model in stage three keeps improving.

Most outbound is measured at the touch — opens and replies — which tells you almost nothing about revenue. Targeting closes the loop: by tying each closed deal back to the signal that started it, you learn which signal types are worth acting on and which messaging converts, then route more effort there.

What to track

  • Signal → meeting rate — which signal types reliably turn into conversations.
  • Message → reply rate by role and channel — which framings land with whom.
  • Meeting → pipeline → revenue — the part most teams never connect back to the original signal.
  • Model feedback — outcomes update the scoring weights, so the next campaign targets better than the last.
The compounding partThis is what makes a targeting system improve over time rather than stay flat: every campaign teaches the score which signals and messages actually convert, so each subsequent campaign starts sharper.

Frequently asked questions

What is a highly targeted campaign?
A highly targeted campaign is outbound outreach directed only at accounts that match an ideal customer profile and are showing a current buying signal, with each message personalized to the account’s specific context. It contrasts with volume outbound, which sends similar messages to a large list regardless of timing or fit.
Why do you need a separate domain for cold outbound?
You use a separate sending domain — sometimes called a secondary or burner domain — so cold outbound never risks the deliverability and reputation of your primary company domain. If a secondary domain’s sender reputation is damaged by spam complaints, your main domain and its day-to-day business email stay protected. A common pattern is buying a close variant of your main domain and sending all cold outreach from there.
What is email warm-up and why is it needed?
Email warm-up is the practice of gradually increasing sending volume from a new mailbox over two to four weeks, starting with a few messages a day and building up, so mailbox providers learn the sender is legitimate. Without warm-up, a brand-new domain that suddenly sends high volume is treated as suspicious and its messages are routed to spam.
How do you keep cold emails out of spam?
Authenticate the domain with SPF, DKIM, and DMARC, warm up each mailbox gradually, keep per-mailbox volume low, write relevant personalized messages, and run inbox-placement tests to confirm where messages land before launching. Deliverability is engineered before the campaign, not fixed after it.
What is a buying signal?
A buying signal is a public or observable event that indicates an account may be ready to buy — such as new funding, a hiring spike, an executive change, a regulatory shift, or a tech-stack change. Signals create a short window, often 48 to 72 hours, when outreach is most likely to land.
How do you score and prioritize accounts?
Assign each account a score from 0 to 100 based on the strength and recency of its buying signals combined with how well it fits the ideal customer profile. Rank accounts by score so reps contact the highest-scoring accounts first, while lower-scoring accounts enter nurture.
How long is the window to act on a buying signal?
The effective window is usually 48 to 72 hours after a signal becomes public. After that, the account is typically saturated with competing outreach and response rates fall sharply, so detection only pays off when it feeds same-day action.
What channels should a targeted campaign use?
A targeted campaign coordinates email, LinkedIn, and phone across three to seven relevant stakeholders per account, sequenced over several days rather than sent all at once. Each channel uses a tone suited to it: strategic framing for executives, operational detail for managers.
How do you measure a targeted campaign?
Measure the full chain from signal to revenue: which signals were detected, which touches were sent, which produced meetings, and which meetings became pipeline and closed revenue. Feed these outcomes back into the scoring model so future targeting improves.

Last updated 23 June 2026 · Kampaign.ai