Outbound Sales Playbook
Outbound works when the list is small, right and explained. Our playbook starts from a signal, cuts it down to the people who can actually buy, and checks every message before it goes.
What it is and when to use it
Outbound sales is reaching out first to buyers who have not come to you. The volume approach, a big list and a template, mostly produces unsubscribes. The approach we run starts from a reason to contact an account, qualifies the person and the company by what they actually do, and routes each lead to the one channel most likely to reach them.
Every stage writes to the client database, so at the end each person on the list has a final flag and a sentence explaining why they are there.
Use this playbook when
- A conference or event attendee feed that needs to become a send-ready list.
- A job-posting or technology signal that defines a target account universe.
- Target accounts that have no contacts yet.
- An attendee list with names and companies but no LinkedIn profiles.
- A qualified dataset that has to land in the CRM without creating duplicates.
Tools we use
Only tools we have run this playbook with, and what each one does in it.
- ClearCue: LinkedIn signal feeds, synced hourly into the database
- Blitz: Job searches that define the account universe
- Apify: LinkedIn profiles, company pages and employee lists
- Prospeo: People search on a company domain and verified emails
- Claude Code: Title grading, company qualification and adversarial review by subagents
- Parallel: Prospect research for personal hooks
- Supabase: The system of record for every stage
- HeyReach and Smartlead: LinkedIn and email sending
- HubSpot: CRM sync and the duplicate audit
The steps
The order we work in. Each step feeds the next, and every result is written to the database as it happens.
Take the qualifying question and personas
Two inputs only: the question that decides fit, in the client's words, and the buyer and champion personas.
Ingest and de-duplicate
Every source lands in the client database, de-duplicated on the normalized LinkedIn URL, never on name and company.
Grade titles by meaning
Distinct titles are graded once against a department and seniority grid that includes every department that can buy. If an event names a role, people are graded against that role, and a company with nobody who clears the bar is reported as uncovered.
Enrich and qualify the company
Reuse what is stored, pull the company page, and judge from the description. The industry tag never decides.
Apply the employment and evidence gates
People who recently left are flagged, not silently dropped. Evidence such as a speaker slot or a self-posted attendance is recorded.
Choose how many people per account
Under about 100 employees we keep one person, between 100 and 500 two, and above 500 two or three.
Route each lead to one channel
The mail gateway check decides email or LinkedIn, and a query proves no lead is in both. On LinkedIn the sequence checks for an existing connection first.
Write hooks only from the person's own posts
If there is a real hook in something they wrote, use it. If not, use the clean generic opener.
Cross-examine and read back
The pre-launch checklist runs, and every lead is read back from the platform before launch.
Sync the CRM with zero duplicates
Companies first, a domain to id map, then contacts, and a duplicate audit after every push.
Read and classify every reply
Replies are captured into the database and classified by intent, and the hot ones are checked against the raw thread.
What good output looks like
From recorded runs:
A conference list went from 2,432 attendees to 534 real personas to 27 people at 22 qualified companies.
A job-posting universe went from 913 companies to a send list of 102 after adversarial review.
Of 70 people held back for an employment mismatch, 46 were rescued after checking their live profiles.
On a 219-person LinkedIn build, 78 people (35%) had a real hook, and a read-back of all 219 found 0 stray braces, 0 em-dashes, 0 wrong-sender signatures and 0 notes over the length limit.
Across 79 replied conversations, every thread was read and classified: 15 declined, 11 agreed to stay in touch, 10 said not now, 9 had met the team at the event, 8 asked for a meeting and 6 asked product questions.
Common mistakes, and the fix
Each of these happened on a real run. The fix is now part of how we work.
Name-first identity searches returned namesakes.
- Cause
- The company term does nothing in a Google search of LinkedIn.
- Fix
- Pull people by company and title, then match names.
Most "verified" identities failed an audit.
- Cause
- The gate tested plausibility instead of contradiction.
- Fix
- Reject when the live employer contradicts the source.
Real buyers were graded too low.
- Cause
- The rubric left out a department that buys.
- Fix
- Check that every department has a path to the top grade.
One industry tag held 220 of 534 people, and the "Financial Services" tag caught only 17 and missed a top account.
- Cause
- Qualifying on the tag.
- Fix
- Judge every company from its description.
A company qualified on a misleading line in its description.
- Cause
- A single-pass judgment.
- Fix
- Add an adversarial auditor that cross-checks similar companies.
A global security chief was graded "entry level".
- Cause
- The provider's seniority field.
- Fix
- Grade from the person's real title and profile.
Every contact got the wrong profile URL.
- Cause
- The scraper returned results in a different order.
- Fix
- Map results by name, never by position.
Two different people with the same name nearly got merged.
- Cause
- Name and company matched.
- Fix
- Compare LinkedIn URL and title before any merge; merges cannot be undone.
"The campaign filtered everyone out."
- Cause
- It was simply paused.
- Fix
- Read the platform state before diagnosing anything.
Questions people ask about outbound sales
What is outbound sales?
Outbound sales is reaching out first to buyers who have not come to you, by email, LinkedIn or phone. We run it as a system that starts from a reason to contact an account, qualifies the person and the company, and routes each lead to the one channel most likely to reach them.
What is the difference between inbound and outbound sales?
Who starts the conversation. Inbound begins when the buyer comes to you, for example through content or a form. Outbound begins when you contact them first. The two work together: people who engage with content can be qualified and contacted as warm outbound.
What is signal-based outbound?
Outbound that starts from an event at the account, such as a relevant job posting, event attendance, or engagement with a topic or competitor, instead of from a static list. The signal decides who to contact and when; qualification still decides whether they are a buyer.
Want this playbook running on your market?
Tell us what you sell and who you sell to. We will show you how this playbook would run for you, and what we would check first.
