Loading tutorials…
Loading tutorials…
Lemlist's flagship feature is personalization that actually scales — not just {{firstName}} hacks. Done correctly, it's worth 2-3x reply rate. Done lazily, it's worth nothing. Here's the line.
Who this is forOutbound operators who already have basic Lemlist campaigns running and reply rate is 2-5%. Personalization is the lever that takes you from 5% to 12%+. If your foundation (deliverability, list quality) is broken, fix that first — personalization can't rescue a bad list.
What you'll need
Step 1
The variables you can use in Lemlist are limited by what's in your CSV. Spend the time enriching the list — it's where personalization lives or dies.
Open your lead CSV. Standard columns (firstName, lastName, email, companyName, jobTitle) only enable shallow personalization.
Add 3-5 'depth' variables per lead. Examples: recentFundingAmount, recentBlogPostTitle, recentLinkedinPostURL, techStackTool, competitorTheyUse, recentJobChange, podcastAppearance.
Sources: Apollo enrichment, BuiltWith (tech stack), Clearbit, ZoomInfo, Crunchbase (funding), LinkedIn manual review for top 50 leads.
For high-value campaigns (50-100 leads to senior buyers), spend 5-10 minutes per lead on manual research. It's expensive per-lead but the reply rate justifies it.
For broader campaigns (300+ leads), use Apollo + Clearbit auto-enrichment for industry/funding data. Don't pretend to deeper personalization than your data supports.
Step 2
Lemlist's dynamic image generator inserts the prospect's name, company logo, or screenshot into a personalized image. Worth 2-3x click rate on the right campaigns.
Lemlist → Images → Create dynamic image. Choose a template (notebook, billboard, whiteboard) or upload a custom background.
Add text layers with personalization tokens: '{{firstName}}, let\'s talk' on a whiteboard image.
For more advanced: insert the prospect's company logo (Lemlist pulls from Clearbit) or a screenshot of their website with a custom annotation overlay.
Best uses: cold campaigns to senior buyers (VPs, founders) where the image makes the email feel hand-crafted. Avoid on transactional or low-stakes campaigns — images can hurt deliverability marginally.
Test deliverability with mail-tester.com after adding images. Image-heavy emails sometimes hit spam filters; lighter images (under 100KB) are safer.
Step 3
Video personalization (a 30-second loom recorded for each prospect) is the most powerful personalization in cold outbound. Worth 5-10x reply rate on enterprise campaigns.
Connect Vidyard or Loom in Lemlist → Settings → Integrations.
Record a 20-40 second video per prospect: state their name, reference something specific (their post, their company milestone), then the ask. Total: under 60 seconds.
Lemlist embeds a thumbnail with the prospect's name overlaid. The thumbnail itself is a personalized image.
Practical scale: 10-20 personalized videos per day max. This is high-effort, high-yield work — reserve for top-tier prospects.
For mid-tier prospects: use one 'evergreen' Vidyard video and personalize via Lemlist variables in the surrounding email copy.
Step 4
Use Lemlist's Liquid syntax (if/else) to swap email content based on lead attributes. One campaign, multiple personalities.
In any email step, use Liquid: `{% if industry == 'SaaS' %} I saw {{companyName}} just hit Series B. {% elsif industry == 'Ecom' %} Black Friday season is brutal — hope your team is prepping. {% else %} Hope your quarter is closing well. {% endif %}`
Common patterns: industry-specific opening lines, role-specific value props (CMO gets brand language, RevOps gets pipeline language), funding-stage-specific CTAs.
Variables you can branch on: any custom variable in the CSV. Test the branches by importing a small sample with different values and previewing the rendered email per lead.
Don't go too deep. 2-3 branch points per email is the sweet spot. More than that and the email reads as fragmented.
Step 5
Lemlist's AI generates a personalized opening line per lead based on their LinkedIn or company data. Pre-generates 100+ snippets at once.
Lemlist → AI → Generate first lines. Connect leads from a campaign. Lemlist scrapes their LinkedIn profile, recent posts, or company page.
AI generates a 1-2 sentence opener referencing something specific: 'Loved your post on agent-native UX — the part on tool composability really clicked.'
Critically — review AI-generated snippets before sending. AI hallucinates 5-15% of the time, especially for less-public prospects. A hallucinated reference (wrong post, wrong topic) tanks credibility instantly.
Use AI snippets for mid-tier prospects (100-300 lead campaigns). For top-tier (under 50 leads), write the opener manually — the time investment is worth it.
Save high-performing AI snippet patterns to a reusable library for future campaigns.
Step 6
Run A/B tests where Variant A is shallow personalization (firstName + companyName only) and Variant B is deep (custom variable + AI snippet). Measure the lift to justify the effort.
Lemlist → Campaigns → A/B test settings. Split 50/50.
Variant A (control): shallow personalization. Subject line uses {{firstName}}, body uses {{companyName}}. That's it.
Variant B (deep): all of A plus a custom variable opener, plus AI-generated first line, plus conditional industry block.
Run for 14 days, minimum 100 leads per variant.
Measure: open rate (subject line effect), reply rate (body effect), meeting-booked rate (full sequence effect).
If Variant B doesn't beat A by 30%+ on reply rate, the personalization isn't generating real lift — the issue is your data quality, your offer, or your targeting. Personalization can't fix those.
Common mistakes
Fake personalization — {{firstName}} + nothing else
What goes wrong: Prospects detect templated emails instantly. {{firstName}} in subject line plus generic body actually performs *worse* than a clean non-personalized template, because it raises expectations and then dashes them.
How to avoid: Either go deep (custom variables, AI snippets, dynamic images for top campaigns) or go non-personalized (clean templated email). The middle ground is the worst place.
No fallback for empty variables
What goes wrong: Email reads ' caught my eye' or 'I saw {{companyName}} just' — broken templates land in inbox and immediately get deleted. Reply rate craters.
How to avoid: Set fallback values for every custom variable: `{{recentPost | default: "your recent work"}}`. Or filter out leads missing critical variables before sending.
Hallucinated AI snippets that reference nothing real
What goes wrong: AI generates 'Loved your post on Web3 governance' but the prospect has never posted about Web3. Prospect sees through it immediately and your credibility is gone for the rest of the sequence.
How to avoid: Review every AI-generated snippet before sending. Reject any snippet you can't verify against the source. Build a "patterns to avoid" list.
Personalizing the wrong thing
What goes wrong: You spend hours referencing their company logo or recent funding round, but the *offer* is generic and doesn't fit their stage. Personalization without a personalized offer is just decoration.
How to avoid: Personalize the *insight* and the *offer*, not just the opener. A series-B founder needs different language than a bootstrapped solopreneur even if you reference the same company milestone.
Heavy images and embedded video that tank deliverability
What goes wrong: Image-heavy emails get filtered to spam by ~10-15% of Outlook setups. You gain on the people who see it; you lose on the people who don't.
How to avoid: Keep images under 100KB. Use one image per email max. Test every variant via mail-tester.com before launching at scale.
Not measuring lift from personalization investment
What goes wrong: You spend 4 hrs/week on personalization and never A/B test against shallow versions. You can't tell if it's actually working or you're burning time.
How to avoid: Run an A/B test on every new personalization technique. Measure reply rate and meeting-booked rate, not just open rate. If lift is under 30%, simplify.
Recap
Done — what's next
How to use Lemlist AI for cold email copy that doesn't sound like AI
Read the next tutorial
Hand it off
Personalization at scale is one of the hardest skills in outbound — it's why specialists charge for it. Building reusable enrichment pipelines, snippet libraries, and variant tests takes 3-6 months to dial in solo. A demand-gen specialist on EverestX has done it 50+ times — $14-16/hr, typically $600-1,500/mo for ongoing personalization-led campaigns.
See specialist rates
Real personalization (deep variables + dynamic content) lifts reply rate 2-3x in tight ICP campaigns. Shallow personalization (firstName-only) doesn't move it at all — sometimes hurts. The technology works; most implementations don't.
Apollo (firmographics, recent funding), Clearbit (industry, tech stack), BuiltWith (specific tools they use), LinkedIn Sales Navigator (recent posts/activity), Crunchbase (funding history). Most operators stack 2-3 sources.
Snippets only. Whole-email AI generation tends to produce templated-feeling copy with subtle hallucinations. Use AI for the personalized opener; write the value prop and CTA yourself.
10-20 quality 30-second videos per day if it's your primary task. Most operators do 5-10 before quality drops. This is why video personalization is reserved for top-tier prospect lists, not broad sends.
Marginally — 5-10% deliverability hit on the most aggressive Outlook setups. The reply-rate lift from a well-done image usually outweighs the deliverability loss for cold outbound. Test via mail-tester.com before scaling.
Lemlist
Lemlist AI can write a full sequence in 60 seconds. The output is also instantly recognizable as AI to anyone who reads cold email all day. Here's how to actually use AI effectively — for ideation and personalization, not for finished copy.
Lemlist
Single-channel cold email is dead. Multichannel sequences — email + LinkedIn + calls — typically lift reply rate 2-3x over email-only. Here's how to build a real multichannel cadence in the 2026 Lemlist UI.
Lemlist
Reply rate under 3% on Lemlist? You're not alone — most DIY cold outbound stalls there. But the cause is usually one of 5 specific issues, not 'cold email doesn't work.' Here's the diagnostic sequence specialists run.
Lemlist
DIY cold outbound works — until it doesn't. This is the honest framework: when the cost of self-managing exceeds the cost of hiring, and how to tell which side you're on right now.