The outbound engine for small teams

1Choose a market your reps can prospect

Pick the market before you hire or buy tools. A good outbound market is one your reps can find, reach and understand. If the buyer has no clear title or no reachable phone, outbound stalls, however good the rep.

Test each segment against 4 questions:

• List: can you build it from filters such as industry, size, location and title?
• Reach: do buyers have work emails, direct dials and active LinkedIn profiles?
• Trigger: can you see a reason to call from outside, such as hiring, funding or a new leader?
• Proof: do you have a customer or result in this segment you can name?

Start narrow: 1 industry, 1 headcount band and 2 or 3 buyer titles. A narrow ICP lets you write messages that sound like you know the buyer’s week.

Write the ICP as a list rule. A description says mid market software companies. A list rule says US software companies in your best headcount band, hiring SDRs, with a new VP of Sales.

Then check volume. Pull the list and count the accounts and contacts it returns. Too few means the list runs out early. Too many means you can’t personalize, so add a trigger.

2Fix data quality before you dial

Data quality is one of the biggest levers on connect rates. Gong’s research finds top reps prioritize direct dials and mark bad numbers.

The gap is wide. In Gong’s analysis of 300 million cold calls, the average rep connects on 5.4% of dials. The top quartile connects on 13.3%.

Set these data rules for the team:

• Put direct dials and mobiles first, switchboards last
• Mark wrong numbers and disconnects in the CRM the same day
• Replace contacts who left with the person now in the role
• Verify every email address before a sequence starts
• Watch bounce rate per mailbox, and pause any mailbox where it climbs

Track connect rate by list source every week. When 1 source connects far below the others, stop buying from that source. Every wrong number is a dial your rep could have spent on a buyer.

3Write messages for each buyer role

Write 1 message per buyer role. The economic buyer, the team lead and the daily user care about different problems, even inside the same company.

• Economic buyer, such as a VP or founder: cost, risk and the number they report
• Team lead: team output, hiring and time lost to manual work
• Daily user: the task they repeat and the tool slowing them down

For each role, write 3 lines before any template: the problem in their words, the trigger you saw, and the proof you can cite. If you can’t fill all 3, you don’t know the role well enough to write yet.

Example email to a VP of Sales who just posted SDR roles:

• Trigger: Saw you’re hiring SDRs this quarter.
• Problem: New SDRs often spend their first weeks on LinkedIn messages instead of calls.
• Proof: name a customer with the same problem and what changed.
• Ask: Worth a 15 minute call next week?

Use the same structure on the phone. In Gong’s data, reps who state the reason for the call have a 2.1 times higher success rate. Say the trigger and the problem in your first 2 sentences.

4Build the multichannel cadence

Vendor data suggests coordinated channels work better than a single one, and Gong’s voicemail data shows it for email and phone. In Gong’s data, email reply rates were 2.73% without a voicemail in the sequence and 5.87% after a voicemail.

Email has hard rules. Google and Yahoo hold bulk senders to a spam rate ceiling of 0.3%, and Google recommends staying under 0.1%. Gmail treats senders of 5,000 messages a day to personal accounts as bulk senders. Authenticate every sending domain with SPF, DKIM and DMARC. Since late 2025, noncompliant bulk mail draws hard bounces.

LinkedIn has softer limits. Vendors report a practical cap near 100 connection requests a week, set by account trust and acceptance rate. LinkedIn publishes no number. Leadriver puts a good acceptance rate at 30 to 45% and says accounts below 20% get throttled. Notes run to 300 characters on Premium and 200 on free accounts. Withdraw invites still pending after 30 days.

A 14 day cadence for 1 buyer role:

• Day 1: call, leave a voicemail, then send email 1
• Day 2: connection request with a note tied to the trigger
• Day 4: call, then email 2 with 1 new piece of proof
• Day 7: LinkedIn message if connected, a call if not
• Day 10: email 3, 1 short question
• Day 14: final call and a closing email

Run the cadence in the prospect’s time zone. Call timing data conflicts. In ZoomInfo’s dataset of 1.4 million calls, Tuesday and Wednesday hold 44% of demos booked, and other datasets disagree. Protect midweek call blocks and test your own days.

5Choose tools by category

This playbook names categories, not vendors. Start with these 5:

• CRM: the record of every account, contact, activity and meeting
• Data and enrichment: contacts, direct dials and work emails for your ICP
• Email verification: checks addresses before a sequence starts
• Sequencer: sends email steps with a daily cap per mailbox
• Dialer: calls from the CRM and logs every outcome

Add call recording once a manager coaches each week, and deliverability monitoring as soon as you send at volume. Both feed the weekly review.

Send from a custom domain. In Hunter’s data, cold email from free webmail addresses drew a 2.5% reply rate, against 5.2% from a custom domain.

Connect every tool to the CRM before launch. If activity lives in 5 places, your weekly review measures the tools instead of the pipeline. Pick each tool by fit. Does it write to your CRM and respect your caps? Can a new rep learn the tool within a week?

6Review the pipeline every week

Activity is easy to count. Dials, emails and connection requests tell you about effort. Run the weekly review on meetings held and qualified opportunities, the 2 numbers closest to revenue.

A 30 minute agenda:

• Meetings held last week, by rep, segment and channel
• Qualified opportunities created, confirmed by the AE
• No shows, and the reason for each
• Connect rate by list source
• Spam rate, bounce rate and LinkedIn acceptance rate
• 1 change to test next week, with an owner

Read activity only to explain results. When meetings drop, check the funnel in order: list, connects, conversations, meetings booked, meetings held. Fix the first step that fell. In Gong’s data, the average rep books a meeting from 4.6% of conversations. The top quartile books from 16.7%.

Templates fatigue, so refresh sequences on a schedule instead of waiting for replies to fall:

• Every week: swap the weakest subject line or opener for 1 new test
• Every month: rewrite the step with the lowest reply rate
• Every quarter: rebuild the sequence around new triggers, proof and objections
• Any week spam rate climbs toward 0.1%: pause, clean the list, then resume