Sponsored by
How Jennifer Anniston’s LolaVie brand grew sales 40% with CTV ads
The DTC beauty category is crowded. To break through, Jennifer Anniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.
Most cold outreach fails for one of two reasons.
The first is obvious: it's generic. The message could have been sent to anyone. "Hi [FIRST NAME], I came across your profile and thought there might be a great synergy between what we're doing and what you're doing..." Nobody responds to this because nobody feels like it's actually for them.
The second reason is sneakier, and it's the one I see more now that everyone has access to AI. The message is personalized in a way that feels mechanical. It references something real about the prospect but it's clearly been assembled from a template with their information plugged in. The personalization is technically present but emotionally absent.
Real personalization, the kind that actually gets responses, requires you to understand the person well enough to say something they'd be surprised a stranger knew. Or cared about enough to mention.
AI makes that level of research viable at scale. Here's the system.
The Personalization Paradox
There's always been a tension in outreach between quality and quantity.
Manual research produces great personalization but takes 20 to 30 minutes per prospect. You can't run a real pipeline on that.
Mass outreach lets you send volume but produces garbage personalization. You get response rates in the 1% to 2% range and spend half your time talking to people who aren't remotely qualified.
The AI research stack breaks this tradeoff. You get meaningful personalization at the speed of templates. Done right, you can run full prospect research in under five minutes per contact and produce outreach messages that genuinely feel like they were written for that specific person.
The Four-Layer Research Stack
Layer 1: Company Context
What does the company do, how big are they, who do they sell to, and what's their current trajectory? Have they made any significant announcements recently?
"Give me a concise company brief for [COMPANY NAME]. Include: what they do, their approximate size and stage, who their customers are, and any notable recent news, announcements, or changes. Keep it under 150 words. Bullet points are fine."
Layer 2: Individual Context
Who is this specific person? What's their role, how long have they been there, what's their professional background? If you have access to LinkedIn, pull their recent posts. Feed them to Claude:
"Here are recent LinkedIn posts from [PERSON'S NAME], who is the [TITLE] at [COMPANY]. Summarize: what topics they're focused on, what challenges or priorities they seem to be grappling with, and what kind of content they engage with or create. What do they clearly care about?"
Layer 3: The Connection Point
This is the specific thing that makes your outreach relevant to this person right now. It might be a recent company initiative your service could support, a post they made about a challenge you solve, or a mutual connection.
"Given what I now know about [PERSON] and [COMPANY], and given that my business [DESCRIBE WHAT YOU DO], what is the most relevant, specific, non-generic connection point I could reference in an outreach message? Give me three options ranked by likely relevance."
Layer 4: The Offer Frame
Before you write the message, get clear on what you're actually offering in a way that's relevant to this specific person.
"I want to reach out to [PERSON] about [YOUR OFFER]. Based on what I know about their situation, how should I frame the value of what I'm offering to make it most relevant to them specifically? What problem are they likely experiencing that my offer addresses?"
Now you have everything you need.
Writing the Outreach Message
With your four layers of research done, feed it all to Claude with this prompt:
"Using the research below, write a cold outreach email to [PERSON NAME] at [COMPANY]. The message should: open with a genuine, specific observation about them or their company (not generic flattery), briefly explain who I am in one sentence, connect what I do to something specific about their situation, and make a low-friction ask (not a demo request, just a conversation starter). Target 120 to 150 words. No buzzwords. No 'I hope this finds you well.' Sound like a smart peer, not a vendor."
Review the draft. The research layers mean Claude will produce something that references real specifics about this person. Your job is to make sure the voice is right and nothing feels off or overly presumptuous. One more edit pass. Send.
Scaling It With Clay
For building your prospect list and running this research at scale, Clay is the tool I recommend. Clay lets you build prospect lists that auto-enrich with LinkedIn data, company news, funding information, and other signals you'd have to manually research otherwise. You can connect it to Claude or other AI models to run research prompts automatically across your entire list.
The result: you build a list of 50 targeted prospects, Clay runs the research workflow on all of them overnight, and you wake up to 50 personalized outreach drafts waiting for your review and final edit. The research leg work, which was the bottleneck, is done.
The Response Rate Math
If generic outreach gets a 1% to 2% response rate and you send 500 emails a month, you get 5 to 10 responses. Of those, maybe 3 to 5 turn into real conversations.
If AI-personalized outreach gets a 6% to 10% response rate, and you send 100 targeted, well-researched emails a month, you get 6 to 10 responses. From a list one-fifth the size. Same pipeline, better prospects, more time for actual selling.
The quantity play never wins long-term anyway. Quality compounds.
This Week in Review
We covered a lot of ground this week: the Prompt Stack, meeting prep, content repurposing, inbox management, client reporting, and today's prospecting system. Every one of these is a real system you can implement this week. Not someday. This week.
Pick one. Build it. See what changes.
If you want the complete toolkit, including templates, automations, and full system documentation, reply with BLUEPRINT for the $47 starter or ACCELERATOR for the full $97 program.
Tomorrow: the Sunday Debrief. What's actually moving in AI right now, and what it means for your business. No hype. Just signal.
Jordan Hale | The AI Newsroom | ainewsroomdaily.com