Marketing automation in B2B has a reputation problem. Most teams that implement it spend three months configuring workflows, six months fighting data quality issues, and eventually settle for using it as an email sending tool that costs £800 per month more than Mailchimp. The gap between what marketing automation can do and what most B2B teams actually do with it is enormous — and the gap is almost always a strategy and process problem, not a technology problem. This guide covers what good looks like, how to build it, and what to fix if your automation investment is not producing pipeline.
What marketing automation actually does
Before strategy: a clear-eyed definition. Marketing automation is software that executes marketing processes based on rules, data conditions, and time-based triggers — without manual intervention at the point of execution. It does not replace marketing strategy, content quality, or human judgment. It executes, at scale and with consistency, the processes that a skilled marketer would perform manually if they had unlimited time.
Consistent execution
Send the right email to the right segment at the right time, every time, without manual scheduling errors.
Scalable personalisation
Personalise content, timing, and offers based on contact data and behaviour without individual attention to each contact.
Lead scoring and qualification
Automatically score contacts on ICP fit and engagement behaviour, promoting to MQL when threshold is reached.
Fix bad data
Automation amplifies data quality — good data produces good results, bad data produces automated bad results at scale.
Replace strategic content
Automated sequences only work if the content they deliver is genuinely useful. Automation of low-quality content accelerates unsubscribes.
Build relationships
Automation creates touches. Human conversations create relationships. The handoff from automation to sales must be genuine.
The core automation workflows every B2B programme needs
Most B2B marketing automation programmes benefit from five core workflows before adding complexity. Build these first, make them work well, and then expand.
A 6-12 email sequence delivered over 90-180 days to contacts who have shown early interest but are not ready to buy. Content follows the buyer journey: problem education (emails 1-2), solution awareness (3-4), differentiation and social proof (5-7), conversion offer (8-10). Triggered by a specific action (content download, webinar registration) not by time alone. The goal is to keep your brand present and valuable during the long period between first interest and buying window.
When a contact reaches MQL threshold (lead score crosses a defined point), automatically: (1) notify the relevant sales rep or SDR via email and Slack, (2) create a CRM task with the lead context package, (3) enrol the contact in a short "bridge" email sequence that maintains engagement while sales prepares for outreach, (4) suppress the contact from generic nurture sequences. The entire workflow should complete within 5 minutes of the MQL trigger.
Contacts who have been inactive for 90+ days (no email opens, no website visits, no form submissions) are at risk of becoming permanently stale. A re-engagement sequence of 3-4 emails sent over 2 weeks attempts to reactivate them: a direct "are you still interested?" message, a high-value content offer, and a final unsubscribe option. Those who don't engage should be suppressed to protect deliverability.
Event attendees are warm prospects who have demonstrated interest by giving time. The follow-up sequence should begin within 24 hours of the event, deliver the promised resource (recording, slides, summary), include a specific next-step offer (a related guide, a strategy call), and be segmented between attendees and no-shows — with no-shows receiving a slightly different first email that references what they missed.
When a contact becomes an active sales opportunity in your CRM, automatically enrol them in a background content sequence delivering relevant case studies, ROI calculations, and competitor comparison content — not as marketing emails, but as resources the sales rep can reference in conversations. Simultaneously, trigger internal alerts to marketing when the opportunity reaches specific pipeline stages, enabling targeted support like custom content creation or executive outreach.
Lead scoring: building a model that works
Most lead scoring models reward what is easy to measure — email opens, page views, form submissions — rather than what actually predicts buying intent. A scoring model that is not validated against closed-won data will systematically promote the wrong contacts.
Platform selection: what to use at each stage
HubSpot Marketing Hub Starter (£45-50/month) covers basic email automation, forms, and simple workflows. For more sophisticated automation at lower cost, ActiveCampaign's B2B plan (£200-400/month) offers excellent workflow logic, lead scoring, and CRM integration. Both are appropriate until you need advanced attribution reporting or complex multi-object data models.
HubSpot Marketing Hub Professional (£720-900/month) unlocks advanced workflows, smart content, A/B testing, and multi-touch revenue attribution. Marketo Engage is appropriate if you're running Salesforce as your CRM and need deep Salesforce integration — but expect 3-4 months of implementation time and dedicated admin resource.
At scale, Marketo Engage and Salesforce Marketing Cloud (Pardot) handle complex data models, multi-entity programmes, and enterprise integration requirements. HubSpot Enterprise remains competitive for companies with simpler sales motions. The primary differentiator at this stage is Salesforce CRM integration depth — if your CRM is Salesforce, Marketo or Pardot have significant native advantages.
Most mature B2B automation stacks include: Intercom or Drift (in-app and web chat automation), Outreach or Salesloft (sales sequence automation, not marketing), Vidyard (personalised video in automated sequences), and Loom (screen recording for high-touch follow-ups). These extend your MAP's capabilities without replacing it.
What goes wrong and how to fix it
- Automation without audience definition: Workflows triggered by "any new contact" rather than "contacts matching ICP criteria" send irrelevant emails to irrelevant people and train your audience to ignore your brand. Every workflow should have an explicit enrolment criteria that enforces ICP filtering before automation begins.
- Email sequences that prioritise volume over value: Sending 12 emails in 90 days because "more touches equals more pipeline" is a deliverability-destroying, unsubscribe-generating mistake. Each email in a sequence should deliver something genuinely useful — an insight, a framework, a case study — not just maintain frequency. If you cannot justify each email with its specific value proposition, cut it.
- Lead scoring that has never been validated against revenue: Run this test: take your last 20 closed-won deals and check what their lead score was at the point of MQL. If your closed-won customers' lead scores were not consistently higher than the leads sales rejected, your model is not predicting buying intent. Rebuild it using closed-won behaviour as the training data.
- Automation that replaces human judgment at the wrong moments: The moment a contact becomes a hot lead and needs immediate human contact is exactly the wrong moment to rely on automation. Automated sequences should reduce the manual work in low-signal situations — not replace human judgment in high-signal ones.
B2B Marketing Automation Playbook
The complete frameworks, checklists, and templates from this guide in a single PDF.
We are preparing a downloadable version of this guide as a complete reference document — including all frameworks, scoring tools, and checklists in a single printable format. Book a strategy call to register your interest and we will send it directly when available.