Marketing Automation — What's Actually Working¶
This document covers what marketing automation is proven to deliver results in B2B SaaS. For each category, we cover what works, what doesn't, and the specific mechanics that make the difference.
1. Email Marketing Automation — What Sequences Actually Convert¶
What Works: Behavioral Trigger Sequences¶
The highest-converting email sequences are triggered by specific user behavior, not time delays. They deliver value first, sell second.
The proven structure:
Trigger: User downloads an ebook, visits pricing page, or completes a specific action
↓
Sequence: 4–8 emails over 14–30 days
↓
Email mix: Educational content → case study → social proof → demo CTA
Example sequence (from multiple sources):
- Day 1: Share an industry report (no pitch)
- Day 3: Send a case study on how similar businesses improved
- Day 7: Educational content addressing a pain point
- Day 10: Product demo or trial offer
- Day 14: Testimonial or social proof
- Day 21: Demo CTA with urgency
What converts better (specific data):
- Including specific numbers and outcomes in emails ("35% increase in efficiency")
- Referencing the prospect's specific industry or role
- Sending case studies showing how similar companies benefited
- A/B testing subject lines (Notion found changing one word =20% open rate improvement)
What Doesn't Work: Long Nurture Sequences with No Segmentation¶
Generic nurture sequences that treat a Fortune 500 CMO the same as a solo consultant will underperform every time. The data consistently shows:
- Don't: Send the same sequence to all leads regardless of company size/industry
- Don't: Lead with a pitch before establishing relevance
- Don't: Send more than 2 promotional emails in a row without value in between
- Don't: Use the same subject line variants without testing
Notion's Results with Lifecycle Email Automation¶
Notion's lifecycle team achieved these metrics using behavioral-triggered, personalized email campaigns:
| Metric | Value |
|---|---|
| Open rate (feature adoption campaign) | 49–51% |
| Click-through rate | 1–1.5% |
| Conversion lift from localization | 6–7% |
| Open rate improvement from A/B testing | 20% (single word change) |
Key insight from Notion: They don't just send email — they run experiments to determine product positioning. A/B testing subject lines for email campaigns informed how they positioned the actual product for launch.
Source: Customer.io Notion case study
The "Day 1" Rule¶
The most effective sequences start with value, not a pitch. Industry reports, educational guides, and relevant content establish trust before asking for anything.
2. Lead Nurturing — What Works vs. What Doesn't¶
What Works: Multi-Touch, Segment-Specific Nurture Paths¶
The evidence: Intercom tracked a customer who read their blog for 3 years before converting when they switched jobs and brought Intercom to their new company. Single-touch attribution would have completely missed this journey.
Key components of effective nurture:
-
Segment by intent level:
- High intent (visited pricing, downloaded demo): Short, direct sequence
- Medium intent (read several articles, attended webinar): Medium sequence with case studies
- Low intent (one blog visit): Long nurture with value-first content -
Segment by company characteristics:
- Company size and industry
- Role/job title (CMO vs. individual contributor)
- Stage in buying journey -
Content upgrade offers:
- Newsletter capture to extend the relationship
- "Next Steps" prompts that offer relevant resources
- Gated content (guides, templates, tools) that solve real problems -
Behavioral triggers:
- Trigger: Visited pricing page but didn't book demo → Send: "Have questions?" email with case study
- Trigger: Watched 75%+ of a video → Send: Follow-up with softer CTA
- Trigger: Clicked "Case Study" button → Send: Full case study + product CTA
What Doesn't Work: Generic Drip Campaigns¶
Generic drip campaigns that blast the same content to everyone consistently underperform compared to behavioral-triggered, segmented sequences. The failure modes:
- Assuming traffic = intent: Most blog readers aren't ready to buy. Drip campaigns that expect immediate conversion set unrealistic expectations.
- Forcing CTAs too early: Content that immediately asks for a demo or purchase loses trust before trust is established.
- Ignoring the long tail: Intercom's3-year conversion story is common in B2B. Brands that only measure first-touch attribution miss the real value of nurture.
Localization as a Nurture Lever¶
Notion found that localized onboarding campaigns (tailored to recipient's location and language) drove 6–7% conversion rate lift for Korean and French markets. For global products, localization isn't optional — it's a direct revenue lever.
3. Content Distribution Automation — How High-Growth Companies Scale¶
The Content Supply Chain Model¶
Deloitte Digital and Adobe research both describe content creation and distribution as a supply chain — linear, systematic, and automatable at each stage.
The content supply chain:
1. Creation → AI-assisted drafting, templates, repurposing
2. Approval → Workflow-based review gates
3. Localization → AI translation and cultural adaptation
4. Distribution → Automated publishing to multiple channels
5. Measurement → ROI tracking per channel and piece
What Works: Systematic Repurposing¶
High-growth companies don't create content separately for each channel. They create one core piece and systematically repurpose it:
- 1 long-form article → 5–10 social posts → 1 email newsletter → 1 video script → 1 podcast episode
- AI tools automate the repurposing while maintaining brand voice
- Distribution automation pushes to LinkedIn, Twitter, email, and community channels simultaneously
Deloitte finding: Brands using a "very high level" of content automation report that their content marketing has a greater impact on annual revenue.
What Works: Owned + Earned + Paid Distribution¶
Ross Simmonds's D.R.E.A.M. framework (Distribution Rules Everything Around Me):
- Owned: Sales posts on LinkedIn, community hubs (Reddit, Quora), email list
- Earned: Organic backlinks, press, community sharing
- Paid: Targeted amplification of top-performing organic content
The automation layer: High-growth companies use tools to automate distribution scheduling, track UTM parameters, and measure assisted conversions across channels.
The Key Insight: Distribution Is the Moat¶
Foundation Inc's Ross Simmonds argues that distribution is more important than creation. You can have the best content in the world, but if no one sees it, it doesn't matter. The companies that win have systematic distribution engines, not just production engines.
4. Outbound Automation — What Books Meetings vs. What Gets Ignored¶
The Problem with Most Outbound Automation¶
Cold email volume is at an all-time high. Inboxes are flooded. Generic sequences with tokenized personalization (name, company, title) are getting 1–2% reply rates at best.
What gets ignored:
- Generic "Hi {{First Name}}" emails
- Emails with no specific context about the recipient
- Sequences that feel automated (same day/time patterns, identical subject line structures)
- Outreach that doesn't reference the prospect's actual business context
What Actually Books Meetings: AI + Intent Signal Personalization¶
Rippling's outbound engine is the most documented real-world example. Here's how it works:
The "Me Outreach" composition breakdown:
- 50% AI-generated: Each email uniquely written, referencing specific LinkedIn posts, recent news, or business context
- 30% tokenized: Standard personalization (name, company, title)
- 20% human templates: Minimal personalization, baseline messages
The key metric: Positive replies (not opens, not clicks). Rippling uses ML to analyze sentiment in email replies and track reply quality, not just volume.
The Dark TAM Engine — finding new markets:
- Aggregates net-new domains from review sites, analytics tools, enrichment vendors
- Validates websites and applies fit criteria
- Layers intent signals (clicks, visits, reviews, interactions) into account scoring
- Scores accounts with predictive model → routes → enrolls in AI-personalized sequences
Build time: 6–7 weeks for the full system
What Makes Outbound Work: Speed + Context + Persistence¶
From ColdCallMe's research (backed by Gartner, McKinsey, Salesforce research):
What books meetings:
1. Specificity over generality: "We book 30 sales meetings/month for B2B consulting firms" outperforms "We help companies grow"
2. Referencing actual business context: Recent news, LinkedIn posts, company announcements
3. Multi-channel approach: Email + LinkedIn + calling (in that order for most B2B)
4. Follow-up sequences: Most meetings are booked on the 3rd–5th contact attempt
5. Positive reply tracking: Measuring what actually matters (positive sentiment responses)
What doesn't book meetings:
- High-volume, low-relevance blasts
- Pure tokenized personalization (name + company name)
- Sequencing without behavioral triggers
- Single-touch outreach
Clay + Outreach: The Data Pipeline for Outbound at Scale¶
Rippling's specific stack:
- Clay: Lead import → enrichment from 150+ providers → waterfall approach for best data coverage
- Snowflake: Data storage and management
- Outreach: Sales engagement platform for sequence execution
Result: 3x enrichment rate vs. previous solution. Clay is described internally as "Rippling marketing team's secret weapon."
5. PLG Growth Automation — Activating and Expanding Users¶
The PLG Automation Framework¶
Product-led growth automation has three distinct phases, each requiring different automation:
Phase 1: Activation (getting users to the "aha moment")
- Define the core action that indicates activation (e.g., "created first project," "automated first workflow")
- Track 5–10 core events that indicate depth of usage
- Automate targeted onboarding messages when users haven't completed key actions
Phase 2: Conversion (free → paid)
- Identify PQL (Product Qualified Lead) thresholds — the combination of actions that predict conversion
- Track account-level rollups (not just user-level)
- Automate upgrade prompts triggered by expansion signals
Phase 3: Expansion (existing customers growing)
- Track breadth signals: invited3+ teammates, connected an integration, exceeded usage limits
- Automate upgrade prompts when usage signals suggest need for higher tier
- Behavioral triggers for expansion (e.g., team size growth, new use cases)
The70% Rule for Free Tiers¶
From Optif.ai's PLG playbook:
- Free tier should provide enough value to activate users (reach the aha moment)
- But have clear limits that push toward conversion
- Sufficient to validate product-market fit, not so generous that conversion is never needed
NRR Before Scaling¶
Critical rule: Net Revenue Retention (NRR) must exceed 110% before aggressive growth scaling. If NRR < 100%, you're losing customers faster than acquiring them — scaling traffic to a leaky funnel amplifies the problem.
Loom's PLG Growth Loops¶
Loom's growth (42X in 5 years) was driven by:
- Acquisition loops: Users create content → invite friends → pay for product → money reinvested into marketing → content attracts more users
- Retention loops: Value deepens as teams adopt Loom for more use cases
- Activation triggers: Onboarding sequences guiding users to their first "aha moment"
Linear's Zero-Ad-Spend PLG¶
Linear is the extreme PLG case:
- Lifetime ad spend: ~$35,000
- More cash in the bank than raised in venture capital
- $400M valuation
- Product quality as the only marketing that matters
Key insight: PLG automation only works when the product is genuinely excellent. Automation amplifies good products and accelerates bad ones.
The Activation Rate Metric¶
Track the percentage of users who fire the core action event within their first 7 days. This is the single most critical PLG metric — if users don't activate quickly, nothing else matters.
6. Cross-Category Patterns: What Automation Actually Works¶
Pattern: Event-Driven Triggers Beat Time-Based Delays¶
Every effective automation system described here is event-driven, not time-based:
- Notion: Triggered when users engage with new features
- Rippling: Triggered by intent signals and account behaviors
- Intercom: Triggered by content consumption patterns
- Loom: Triggered by product usage events
Time-based drip campaigns are a fallback, not a best practice.
Pattern: Personalization Stack (Token → AI → Human)¶
The evolution of personalization in automation:
1. Token era: {{First Name}}, {{Company}} — now expected, not impressive
2. AI era: Contextual awareness, referencing LinkedIn posts, recent news, specific business context
3. Human touch era: For high-value accounts, human-written personalized messages outperform AI
Most companies should be in the AI era. High-value enterprise motion may need the human era.
Pattern: Data Infrastructure Is the Prerequisite¶
Every automation system that works has clean, connected data:
- Rippling: Openprise as the data orchestration layer
- Notion: Customer.io for event-driven automation
- Clay:150+ data providers with waterfall enrichment → Snowflake → Outreach
Without data infrastructure, automation amplifies messiness rather than scaling effectiveness.
Pattern: Measure Outcomes, Not Activities¶
| What to measure | What NOT to measure |
|---|---|
| Positive replies (outbound) | Open rates alone |
| Revenue impact (email) | Click rates alone |
| NRR (PLG) | Signup volume alone |
| Conversion rate lift (localization) | Emails sent |
| Multi-touch attributed revenue | First-touch conversions |
Pattern: Speed of Execution Matters¶
Rippling built their entire outbound engine in 6–7 weeks. Notion runs hundreds of experiments per year. The companies winning with automation move fast, test constantly, and iterate based on data — not months of planning followed by perfect execution.
Pattern: Define Exit Criteria for Every Sequence¶
Every automated sequence needs explicit stop conditions — don't keep emailing people who already converted or explicitly opted out. Sequences without exit criteria erode trust and waste spend. Treat "when to stop" as a first-class part of the workflow design, not an afterthought.
Pattern: Tools Are Commoditized — Strategy Is the Differentiator¶
Automation without personalization is wasted spend. The tooling (ESPs, sequencers, enrichment) is commoditized and broadly available; the durable advantage is the segmentation, triggers, and message strategy layered on top. Buying the stack is table stakes; deciding what it does is the moat.
Sources¶
- Customer.io — Notion case study (January 2025)
- Openprise — Rippling outbound engine webinar recap (April 2025)
- Clay.com — Rippling customer case study
- ColdCallMe — B2B cold calling research (2026)
- Optif.ai — PLG playbook guide (November 2025)
- Deloitte Digital — Content supply chain automation (December 2024)
- Adobe — How enterprises scale content creation workflows (May 2026)
- Foundation Inc / Ross Simmonds — D.R.E.A.M. framework
- Sequenzy — Email nurture sequence examples
- Sendr.ai — B2B cold email templates (2026)
- Userpilot — Product-led growth playbook (April 2026)
Concepts¶
Related concepts: cold-email-sequence · email-deliverability · activation · product-led-growth · workflow-vs-agent · human-review-gate