The 4 levels of personalization
Not all personalization is equal. The reply-rate impact compounds at each level:
Level 0: No personalization
"Hi there, I wanted to introduce our company..." Reply rate: 0.5-1%. Don't even send.
Level 1: Token personalization
"Hi {FirstName}, I noticed {Company} is doing great work in {Industry}." Reply rate: 1-3%. Slightly better than nothing, recognized as automated within 2 seconds.
Level 2: Surface research personalization
"Hi Sarah, I saw on your LinkedIn you recently joined Acme as VP Marketing." Reply rate: 3-6%. Better — proves you looked. But "joined Acme" is one click of research; recipients know this.
Level 3: Deep contextual personalization
"Hi Sarah, just read your blog post on the customer retention reframe — the bit about 'activation as a moment, not a metric' is genuinely the most useful framing I've seen on it this year." Reply rate: 8-15%. This is where real outreach lives. Requires reading their actual content.
Level 4: Insight + observation personalization
Level 3 + a specific insight or pattern from your own work that connects to what they're doing. Reply rate: 12-22%. The ceiling. Almost no one does this consistently.
What to actually personalize (the research checklist)
Here's the exact list to scan for each prospect. Allocate ~10 minutes per high-value prospect, 2-3 minutes per medium-value:
- Recent funding or hiring announcements (Crunchbase, LinkedIn) — signals priorities
- Their last 3-5 LinkedIn posts — signals what they think about
- Their company's last blog post or product launch — signals what the team is shipping
- Recent press, podcast appearances, or interviews — signals how they talk about their business
- Open job listings on their careers page — signals what's broken
- Their actual product or service (sign up, try the demo, browse the pricing page) — signals their UX, positioning, gaps
- Their competitive context — who they're winning/losing to, what's changing in their market
You're looking for one specific, recent, falsifiable thing to mention. Not "I see you're growing fast" — that's generic. "Saw the Series B announcement and the new mid-market hire — that combination usually means..."
The 'one specific thing' rule
Every cold email needs exactly one specific thing — a sentence the recipient knows you couldn't have written without doing real research. Not three things. One.
Why one and not three? Three is performative. The reader sees you trying. One feels natural — like a colleague who happened to notice.
The "one specific thing" can be:
- A recent quote from their interview or podcast
- A specific design choice in their product
- A line from their last blog post or LinkedIn essay
- A pattern in their hiring or expansion
- A friction point in their UX you actually hit while using the product
- A specific number from their last earnings call or press release
The acid test: would the recipient know this came from real attention, not a script? If yes, you've nailed personalization.
How to do this at scale (the only honest answer)
You have three real options for scaling personalization. There is no fourth.
Option A: Manual at small scale (10-30 prospects/week)
Spend 10 minutes per prospect. Best quality, doesn't scale. Right answer if your average deal size is over $20K.
Option B: AI-augmented research (50-200 prospects/week)
Use a tool that automates the research step — fetches the prospect's site, parses their content, surfaces specific things to mention. Mail Doneky does exactly this: paste a prospect, get a researched, personalized 4-email sequence in 30 seconds.
Critical: AI-augmented research only works if the AI actually reads the content, not just merges variables. Most "AI personalization" tools just template tokens with fancier inputs. The test: does the email mention something that wasn't in the prospect's name, company, or title? If no, it's not real personalization.
Option C: Hybrid — AI research + human edit (10-50 prospects/week, high-value)
AI does the research and drafts; you spend 2 minutes editing the most important emails. Best ratio of quality to volume for high-stakes deals.
The personalization mistakes that backfire
- Personal-life personalization — "Loved your hiking pic from last weekend!" reads creepy at best, sycophantic at worst. Stick to business context.
- Stale references — citing a 2-year-old blog post or an outdated company description. Recency matters; old references signal sloppy research.
- Generic compliments — "I love what your team is doing." Means nothing. Either say what specifically, or don't.
- Forced personalization in the subject line — "Quick question for you, {FirstName}!" The exclamation point + name combo screams template.
- Overcompensating with multiple specifics — three personal references in 4 lines reads as performance, not attention.
- Wrong-industry personalization — referencing an industry term incorrectly is worse than not referencing it. Confirm before using jargon.
What good personalization looks like — annotated example
Hi Lena,
Saw the new pricing page — moving from monthly-default to annual-default is a smart bet given that your free trial is 14 days and your typical activation curve is probably 21+ days. (← researched: trial length is on their pricing page; activation curve is industry pattern)
The thing I'd watch: when you make annual the default, the trial-to-paid email sequence usually has to be rewritten too. We've seen teams 2x conversion when they reframe the trial-end email from "your trial expires" to "lock in your annual rate." Different message, same data. (← original insight from real work, falsifiable but credible)
I help SaaS teams audit their pricing-to-trial flow when they make changes like this — usually 1-2 weeks, $4-8K depending on scope. (← concrete offer, transparent pricing)
Worth a look? Happy to send the audit framework first if useful. (← soft CTA, deliverable-based)
Sam
Total time to write this email manually: ~20 minutes. Total time with Mail Doneky: ~30 seconds. Same depth of research, same structure, same reply rate.
Personalize at scale without spending all week
Mail Doneky researches each prospect, writes a personalized 4-email sequence, and sends from your Gmail in 30 seconds per prospect. Free to start.
Try Mail Doneky Free →