How to Send 300 Personalized Cold Emails Per Day (Without a Sales Team)

The Personalization Paradox

There's a tension at the core of cold email. Better personalization leads to higher reply rates. Higher volume leads to more opportunities. But more volume usually means less personalization.

The way out is AI research. The bottleneck in cold email personalization isn't writing - it's research. Finding the one specific detail about this person's company that makes the first line feel like it was written for them takes 2-5 minutes per lead manually. With Strawberry, it takes about 15 seconds.

Once research is fast, the paradox disappears. You can have both.

Step 1: Build a Verified Lead List

Bad data is the single biggest cause of cold email failure - not bad copy.

A good lead list has:

• Verified email addresses (use Hunter, Apollo, or ZoomInfo - not guessed patterns)
• Current job titles (people change jobs; a 6-month-old list is often 20% wrong)
• Company context (industry, size, recent news)
• No obvious disqualifiers (competitor employees, people who've unsubscribed)

For 300 emails per day, you need at least 2,000-3,000 verified leads queued up.

Step 2: Research at Scale with AI

For each lead, you want to find:

1. One specific company signal - a recent funding round, a new hire, an expansion, a product launch
2. One role-specific angle - something about what their job actually involves
3. Anything recent and public - a LinkedIn post, a press release, a job listing

With Strawberry, the AI opens each lead's profile and company site, pulls the most relevant signals, and formats them for the personalization step.

Step 3: Email Structure That Gets Replies

Line 1 (personalized hook): One sentence referencing something specific to this person or company.

Lines 2-3 (credibility + value): Who you are and what you've helped similar companies do. "We helped a 15-person Danish agency cut reporting time from 6 hours to 40 minutes" is better than "we help agencies save time."

Line 4 (ask): "Would it be useful to see a quick example?" beats "do you have 30 minutes for a call?"

Subject line: Short, no clickbait. "10 min next week?" still performs because it sets accurate expectations.

Keep the whole email under 120 words. Longer emails get lower reply rates.

Step 4: Domain Health

Sending 300 emails per day from a single domain will get you flagged if you don't set it up correctly:

• SPF, DKIM, and DMARC: Non-negotiable. Check with mail-tester.com.
• Warm-up period: A new domain needs 4-6 weeks. Start at 20/day, increase by 20/day each week.
• Bounce rate: Keep under 3%. At 5%+, Google starts throttling your domain.
• Reply rate: A 5% reply rate on cold emails is good. Below 2% and you should rethink the offer, not just the copy.

Step 5: Follow-Up Sequences

• Day 0: Personalized cold email
• Day 5-7 (no reply): One-line bump. "Wanted to make sure this didn't get buried."
• Day 12-14 (still no reply): Breakup email. "I'll stop reaching out after this - happy to pick this back up whenever it makes sense."

Three touches is the maximum. After that, move on and re-engage in 90 days with a fresh angle.

Step 6: CRM Sync and Reply Handling

Every sent email should create a CRM record automatically. With Strawberry, for each sent email a person and opportunity record is created in your CRM with a follow-up task scheduled for 6 days out. When a reply comes in, the workflow routes it to a draft response, updates the CRM status, and cancels the automated follow-up.

A 300-Email-Per-Day Operation

• Morning (09:00): Batch send fires automatically. 300 verified leads, each with a personalized first line generated the night before.
• Afternoon (15:00): Follow-up sweep runs. Sends second touches to non-responders at day 6, breakup emails at day 13.
• Evening: Research queue refills. The AI pulls 300 new leads, researches each one, writes personalized first lines, and queues them for the next morning.

With Strawberry, this entire operation runs in the background. The only thing requiring your attention is reading and responding to replies.

What "Personalized at Scale" Actually Looks Like

When the first line says "Saw your team just hired a Head of AI last month - curious whether you're building internal tooling or looking for external solutions," that's a real observation drawn from a real LinkedIn job posting. The person reading it can't tell whether a human or an AI pulled that detail.

The standard: would this first line work if you'd written it yourself? If yes, send it. If it reads like a template with a name swapped in, fix it before it goes.

That quality bar is achievable at 300 emails per day.