The SDR playbook: a complete template for outbound teams

The playbook on one page

Use this as the template. Each row is a section of your playbook, the decision it has to record, and who owns it.

1. ICP and account tiers

Build the ICP from your last twenty closed-won and closed-lost deals, not from a brainstorm. Write the disqualifying criteria as clearly as the ideal ones: below the size where budget exists, nobody who owns the problem, a competing tool bought recently, a region you cannot serve. Then tier the list:

• Tier 1. Matches the profile and shows a trigger event now. Research deeply, personalise heavily.
• Tier 2. Matches the profile without a trigger. Sequence with light personalisation.
• Tier 3. Plausible but unproven. Test in volume and promote what responds.

The best ICPs contain one number. For Easyship, it was eCommerce businesses shipping a thousand parcels a month or more. In logistics software generally, throughput predicts value better than headcount, which is why the logistics ICP is built on a volume floor. End this section with a named account list, not a description. Method: how to build a B2B ICP sales will use.

2. Personas

For each tier, name the roles to contact, what each one is measured on, and who signs. Most B2B deals involve several people who need different reasons to agree. A logistics software deal, for example, needs operations for the pain, finance for the budget and IT for the integration, and outbound that only reaches operations stalls after a good first meeting. The worked example is in the logistics buying committee.

Write down who to contact first and who to bring in before a pilot or proposal, so reps do not discover the second buyer at the end of the cycle.

3. Messaging

Write messaging per segment and per persona, not per campaign. Each message needs three things the playbook should spell out:

• The problem, in the buyer’s language. If an opening line would work unchanged on a different company, it is not specific enough.
• One proof point mapped to each claim: a named result, a number, or a customer the buyer will recognise.
• A second angle, a different problem for later touches, so touch five does not read like touch two.

Low reply rates are usually targeting problems before they are copy problems, so this section is only as good as section 1. More in ICP, persona and messaging.

4. Lead sources and priority

List every source of leads and the order in which they get worked. A sound default:

• Inbound requests first, with a first response inside one business hour, measured in the buyer’s working day. One hour is the fastest standard most teams can hold every day. See speed to lead.
• Then Tier 1 outbound accounts showing a trigger event.
• Then the rest of the sequence queue.

Assign every lead to a named person, not a shared queue. A lead nobody owns is answered by nobody. Segment first on the one or two attributes that change who should handle it, then assign. Detail in inbound lead routing.

5. The sequence

Seven touches is a reasonable starting shape, but the count matters less than whether each touch says something new. Record four positions:

• The hook. Why this account, specifically.
• The proof point. One claim, with evidence.
• The angle change. A different problem, not the same pitch louder.
• The variations. Short, direct, and willing to close the loop.

Run email, LinkedIn and phone as one motion. Email carries the argument, LinkedIn carries recognition so the name is familiar before the call, and phone carries the conversation. Then write down the rules that hold it together: a fixed touch pattern and interval, a written exit rule for when to stop, one variable changed at a time, and reply analysis across the whole campaign. Full structure: how to structure a B2B outbound sequence. None of this works without the sending setup in the cold email deliverability checklist.

6. Calls and objections

Write down the five questions a good rep asks on every first call. If the founder or best rep cannot list them, the process is not ready to hand to a new hire. Then list the five objections your market actually raises, with an answer for each that a rep can say out loud without sounding scripted, and rehearse them in recorded mock calls.

Add a rule for the end of every call: the next step, agreed with a date. A conversation that ends with “send me some information” and no date is not pipeline.

7. Qualification and handoff

This is the section that moves numbers. Write the criteria a lead must meet to become an SQL, as evidence rather than impressions: not “prospect is interested” but “prospect has named the problem in their own words and confirmed who signs”. Add the disqualifiers. Test the standard by having two reps grade five recent calls independently; any disagreement means it is not finished.

At Locus, writing this down and enforcing it when a lead changed stage moved MQL to SQL conversion from 10% to 25% with no additional leads. At WorkStore, every inbound lead was qualified on four criteria: budget, genuine need, timeline and units required.

Then define the handover. The note to the account executive carries the reasoning against each criterion, not just a name and a number. And every booked meeting gets three confirmations: a human message on booking, a useful note the day before, and a one-line check the morning of. At ConvertCart, fixing follow-up and that confirmation flow cut no-shows from 44% to 15%. Detail in MQL to SQL conversion, qualification frameworks and meeting no-shows.

8. Tools and data rules

Name the tool for each job and nothing else. For most teams that is a CRM, one data and prospecting tool, a sequencer and an AI assistant for research. Then write the data rules: which fields must be filled before a lead can move stage, and nothing more, because a CRM that asks for more than a rep will complete mid-call stops being updated.

The stack by stage is in best sales tools for startups and the B2B sales tool stack.

9. Metrics

Report each number at the speed it can actually move:

• Daily, for the team: new meetings booked, replies needing an answer, inbound leads still unworked.
• Weekly: conversion between the stages you are fixing, by rep and by source, plus deals older than two cycles.
• Monthly: pipeline velocity by segment, SQL to close rate, and cost per opportunity by source.

Judge outbound on replies, not opens, which have been inflated by image pre-fetching since 2021. If a metric can rise without anyone outside the company changing their behaviour, it is an activity measure, not a result. More in sales metrics that predict revenue.

10. Coaching and onboarding

Write the coaching rhythm into the playbook so it survives a change of manager. Each SDR gets a weekly one-to-one, separate from the pipeline review, covering one recorded call, one sequence with rewrites attached, and every meeting the account executive rejected. Each session ends with one change for the next week. One manager can properly coach around six to eight reps.

New SDRs follow a 30 day plan: product, ICP and the qualification standard in week one; shadowed calls and supervised sends in week two; their own cadence and first call reviews by week four. Full productivity typically takes three to six months. Detail in SDR coaching and how to hire your first SDR.

How to write it, and in what order

Write sections 1, 3 and 7 first: the ICP, the messaging and the qualification standard. They decide whether anything else works, and they are the three a new hire cannot invent. Then the sequence and the sending setup, then tools, metrics and coaching. In a new outbound build that takes most of the first month, and nothing should be sent in the first two weeks. The week-by-week version is the first 90 days of outbound.

Keep it short enough to use. Each section should fit on a page, with links to the detail. A playbook nobody opens is a document, not a playbook. Review it every quarter against closed-won: if SQLs that meet every criterion still do not close, a criterion is wrong.

Where this playbook has run

Locus. Shobhit was hired to qualify inbound on BANT and spent seven to eight months doing it call by call before the role expanded into global SDR and revenue operations. The written qualification standard is where the 10% to 25% MQL to SQL result came from, and the US and Europe were later opened through outbound with a distributed team of twelve.

GoComet. The SDR team grew from 10 to 23 across the US, India and South East Asia, with two SDR managers hired along the way and weekly call reviews against the written standard. Output rose from 2 to 6 SQLs per rep per month. How the scale was managed is in how to scale an SDR team.

Landmark Group’s Logistiq division. The ICPs, personas, CRM workflows and sequences were the foundation the founding inside sales team worked from, and the division generated $500K ARR in its first quarter.

NuVista AI. A client playbook written from zero: founder sessions to define markets, ICPs and personas, then messaging, Apollo set up for campaigns and sequences, and the SDR onboarded onto it. More in the case studies. If you want the playbook written with your team, that is sales playbook development, and hiring into it is the SDR team build engagement.