How to write a B2B email nurture sequence with AI
- Level: Intermediate
- A few days to set up
The short answer
Define the trigger, the reader and the one step each email should move them towards, then give a model that plan, your proof and your voice guide. Ask it to critique the draft as a sceptical buyer, edit the result yourself, and build it with an exit for anyone who replies or books.
On this page
At a glance
- The marketing problem
- Many nurture sequences are a series of “just checking in” emails written in one sitting. A sequence built around the questions buyers have at each stage is more useful to them, and AI makes drafting and critiquing each email quick.
- Who this is for
- Demand-generation, lifecycle and marketing-operations teams in B2B.
What you need
Before you start
- A clear trigger, such as a guide download or a webinar registration
- Buyer stages and the questions buyers have at each one
- Approved proof: case studies, figures with sources, customer quotes
- A marketing automation tool (for example HubSpot, Marketo or Customer.io)
What goes in
- The asset or event that triggers the sequence
- Common objections from sales
- Your voice guide
- Existing emails that performed well
How it works
- 01 · You decidePlan the sequence
- 02 · AI stepDraft the emails
- 03 · AI stepCritique as a sceptical buyer
- 04 · You decideEdit and approve
- 05 · ToolBuild it in your automation tool
- 06 · OutputReview monthly
- Tool
- AI step
- You decide
- Output
Step by step
01 You decide
Plan the sequence
Write down the trigger, who receives it, their stage, and the one question or objection each email should answer. Decide the exit: a reply, a booking or a sales conversation.
02 AI step
Draft the emails
Give the model the plan email by email, so it cannot drift into generic follow-ups.
Prompt to copy
Draft a [number]-email nurture sequence for [audience] who [trigger]. Plan, email by email: [goal and question answered for each]. Each email: under 150 words, one idea, one call to action, a subject line of under seven words. Use only this proof: [paste]. Follow this voice guide: [paste]. Suggest the delay before each email.
03 AI step
Critique as a sceptical buyer
Ask the model to read the draft as the recipient would, and to be specific about what fails.
Prompt to copy
Read this sequence as a busy [role] who downloaded [asset] but is not ready to buy. For each email, say whether you would open it, read it and act on it, and why not. Point out generic lines, repeated points and any claim that needs proof. Sequence: [paste]
04 You decide
Edit and approve
Rewrite based on the critique, check every claim and name, and approve the final copy and timing.
05 Tool
Build it in your automation tool
Set the trigger, the delays and the exit conditions, and suppress contacts who are already in a sales conversation.
06 Output
Review monthly
Check replies, meetings and unsubscribes per email, and replace the weakest email first.
Where a person decides
A marketer edits and approves every email and the exit rules before the sequence goes live.
What to watch for
- Models default to generic follow-ups; each email needs one specific idea.
- Too many emails, too close together, damage deliverability and trust.
- Figures and customer names must be approved; do not let the model invent proof.
How to tell it is working
- Replies, meetings and pipeline from the sequence are measured, not only opens.
- Unsubscribe and spam-complaint rates are tracked per email.
- Each email is tested once against a simpler version.
An illustrative example
Input
Illustrative: people who downloaded a fictional guide to choosing an ERP system, most of them finance managers at mid-sized firms.
Output
Five emails, each answering one question (cost, migration risk, timeline, internal buy-in, next step), with delays, subject lines and an exit when someone replies or books a call.
Want this running in your team?
AI workflow implementation. Turn a repetitive marketing task into a workflow your team can use with confidence.
Related workflows
How to personalise cold emails at scale with AI
Write outreach from verified research on each account, with one relevant reason to talk and a human edit before anything is sent.
How to research a prospect with AI before a sales call
Turn a new enquiry into a sourced one-page brief with questions for the first call, checked by a person before sales sees it.
How to analyse customer interviews and reviews with AI to sharpen your messaging
Find the pains, outcomes and exact phrases customers use, so headlines and ads say what buyers already think.