What this playbook is for

Automated outbound is not the act of sending more emails. It is the system that decides which accounts to target, which people to contact, what to say, which channels to use, and how to learn from the results.

The order matters. If the ICP is vague, the list will be poor. If the account data is poor, personalization will be inaccurate. If the contacts are not mapped to buying roles, the message will reach the wrong person. If performance is not reviewed, the team will repeat the same mistakes at higher volume.

This playbook creates a complete operating sequence: define the ICP, choose the campaign strategy, source and qualify accounts, map contacts, build copy, launch multi-channel outreach, and review the results every week.

“Define the ICP, build and qualify account lists, personalize messaging, and deploy multi-channel campaigns with tracking and optimization built in.” — Workflows.io,

The system at a glance

Step 1: Create the ICP model

Define the targeting framework before sourcing accounts. Use three groups of criteria.

Firmographics include industry, headcount, revenue, and geography. Technographics include the technology stack, tools used, and infrastructure signals. Account-fit signals include hiring activity, funding stage, website traffic, and engagement.

The ICP should be specific enough to make a decision. “Mid-market technology companies” is a starting category, not an operating rule. A usable model might define an industry range, headcount band, geographic market, relevant systems, and a business signal that indicates the account has a reason to act.

Separate hard filters from prioritization signals. A company may be excluded because it is outside the market or too small to support the product. Another company may qualify but receive a higher priority because it is hiring for the function your solution supports.

Step 2: Choose the campaign strategy

The source playbook identifies three campaign structures.

Do not use one campaign structure for every objective. A broad campaign can help you learn the market, while a micro campaign may produce a clearer message test. A signal-based campaign should use the signal in the timing and prioritization, not merely as a list label.

Step 3: Source and enrich accounts

Build the account list from multiple sources. The original playbook names Apollo, AI Ark, Ocean.io, Discolike, Sales Navigator, and Store Leads as database sources; Apify, Octoparse, Serper.dev, and Python for scraping and discovery; and Clay, RB2B, Jungler, Fibbler, Sumble, and TheirStack for signals and enrichment.

Do not treat source quantity as data quality. Aggregate the records into Clay, deduplicate them, normalize company names and domains, enrich the required fields, and apply the qualification logic. Remove non-qualified accounts before you begin contact sourcing.

Every account should have a clear reason for inclusion. That reason may be fit, an active signal, or both. Keep source, date, and campaign fields so results can be compared later.

Step 4: Source and segment contacts

Pull contacts from qualified accounts and categorize them as decision makers, champions, or users.

A decision maker controls or strongly influences the purchase. A champion experiences the problem, benefits from the change, or can help build internal support. A user works directly with the process or product and may provide valuable context even without budget authority.

The categories are not job-title shortcuts. A VP may be a decision maker in one company and a distant executive in another. Use department, seniority, company size, and the buying problem to determine the role.

Step 5: Build copy and personalization

Structure the message around positioning, offer, and a personalization layer. Positioning explains why the problem matters and why your approach is relevant. The offer states what the prospect can get or do next. Personalization connects the message to the account, contact, role, or signal.

Create multiple variants and run A/B tests, but change one meaningful variable at a time. If the audience, offer, subject line, message, and sending domain all change together, the result will not teach you much.

Personalization should improve relevance, not display research. Use a detail that changes the reason for contacting the person. Do not add random facts simply to make the message look customized.

Step 6: Launch multi-channel outreach

Use Instantly for email outreach and HeyReach for LinkedIn outreach. Configure sending infrastructure and daily limits before launch. Sending limits protect deliverability and prevent the team from creating a volume problem that the copy cannot solve.

Log activity into the CRM and update contact records, deal stages, and reply status. The CRM should show what happened without requiring a rep to reconstruct the campaign from separate tools.

Send Slack notifications when replies arrive. Include the contact name, company, campaign, and reply text so the owner can act without searching for context. Define who owns the reply and how quickly positive responses must be handled.

Multi-channel does not mean every prospect receives every possible touch. It means the campaign can use the channel that fits the audience, role, and level of priority.

Step 7: Report and optimize weekly

Track four core outcomes: reply rate, positive reply rate, meetings booked, and pipeline created. Reply rate tells you whether the market is responding. Positive reply rate tells you whether the response is commercially useful. Meetings show conversion. Pipeline shows whether the campaign creates business value.

Review performance by account segment, contact role, source, signal, message variant, and tier. Then change one of four levers:

•Targeting: Are the right accounts being selected?

•Copy: Does the message match the problem and role?

•Segmentation: Are contacts being grouped by buying context?

•Signal filters: Are the triggers meaningful enough to justify outreach?

Do not optimize toward opens or raw activity if they do not lead to useful conversations. A smaller campaign that creates qualified pipeline is better than a larger campaign that produces noise.

Launch checklist

Outbound scales when the system becomes more disciplined as volume increases. Automation should remove repetitive work, not remove judgment from targeting, qualification, or follow-up.