How to Run an Email Campaign Targeting Biotech Researchers for AI Training in 2026

Quick Answer: You already used Origami to find biotech researchers who need high-quality AI training data. But a list doesn’t book meetings — Origami’s built-in email sequencer does. It turns your prospect list into a hands-off outreach campaign. Paste your own 3-touch templates or let Origami’s AI write personalized sequences for every lead, then send everything without leaving the dashboard. Here’s exactly how to run that campaign from list to reply.

If you haven’t built your list yet, read how to build a list of Biotech Researchers for AI Training first. This guide assumes you have a clean, enriched list ready inside Origami.

---

Step 1: Build (or Recap) Your List in Origami

Even if you already built the list in the parent post, it’s worth a quick sanity check. The prompt that works for finding biotech researchers active in AI training is:

“Find biotech researchers with a title like Principal Scientist, Senior Computational Biologist, or AI/ML Research Lead, working at companies or institutes that publish in areas such as drug discovery, genomics, protein folding, or medical imaging. Include only people who mention AI model training, data annotation, or custom dataset curation in their recent work or publications. Focus on commercial labs and well-funded research institutes in the US and EU.”

Origami returns a list with verified names, email addresses, phone numbers, job titles, company details, and source links. You can run this on the free plan (1,000 credits, no credit card) to test it, then upgrade when you’re ready to scale. I always run a few prompts before committing to a full list build — Origami’s live-web search surfaces leads that typical databases miss.

But the raw list isn’t ready for email yet.

---

Step 2: Refine and Qualify Your List for Email

Cold email to researchers fails fast when you treat every name the same. Use Origami’s table view to scan and filter. Here’s what I do for this specific audience:

Remove clearly bad fits. Drop pure academic postdocs with no industry affiliation unless your goal is to reach a PI and they’re listed. Remove anyone whose role is solely administrative (lab manager, program coordinator) — they won’t be the one deciding on training data.

Segment by company type. In Origami, add a tag column and mark leads as:

• Pharma/Biotech enterprise (Pfizer, Roche, Novartis, etc.)
• AI-first biotech startup (Recursion, Insilico Medicine, etc.)
• Specialized contract research org (CRO) that builds models for clients
• Academic research group with commercial spin-off

Segment by role influence. A Principal Scientist building an internal deep learning model for toxicity prediction has a different buying trigger than a Head of Data Science evaluating external datasets. I look at the enriched profile data Origami pulls — recent papers, patents, tool stacks — and tag leads as “model builder”, “data evaluator”, or “both”.

What “qualified” looks like for this audience:

• They have a current position in a commercial or well-funded translational research org.
• Their work or publications show hands-on work with ML models (PyTorch, TensorFlow, protein language models, etc.) or they mention the need for labeled microscopy/genomics data.
• They hold a senior enough title to influence a budget line for external data or annotation services (Senior Scientist and above, or a Lab Head with procurement authority).

After tagging, you can clone your refined list into separate campaigns inside Origami — one for enterprise pharma, one for startups, etc. This lets you tailor messaging without manually re-sorting later.

---

Step 3: Create the Email Sequence

Origami gives you two ways to build your sequence.

Option 1: Paste your own templates. If you already have cold email copy that works, just drop it into Origami’s sequencer. Set the delay between touches (we use Day 1, Day 3, Day 7), upload your customized templates, and hit “Launch”. The sequencer personalizes , , automatically from the enriched lead data.

Option 2: Let the Origami agent write it. This is where it gets interesting. In the sequencer, you can prompt the AI agent to generate a 3-day personalized sequence for all leads. It writes emails based on each prospect’s actual profile — their research area, company, tools mentioned — so every message reads like you did the homework. The agent even suggests the cadence based on what works for similar campaigns.

Below is a full, copy-paste-ready 3-touch sequence I’ve used when reaching biotech researchers about curated AI training data. The angle: you’re selling high-quality, labeled biological datasets that accelerate model training. Adjust the value prop to your product, but keep the tone and brevity.

---

Day 1 – Initial cold email

Subject: Thoughts on [Company]’s AI training data bottlenecks?
Preview: A quick idea for scaling your model validation

Hi ,

I’ve been following [Company/Institute]’s work in [specific area — Origami fills this from their profile, e.g., “protein-ligand interaction prediction”].

Many teams in this space tell me they spend months annotating internal data before they can even start training. We specialize in delivering pre-annotated, domain-specific datasets — for example, 50k labeled protein-ligand pairs with full structural metadata.

If you’re trying to shorten your model development cycle, this could shave weeks off your validation phase.

Worth a look?

Best,

---

Day 3 – Follow-up (different angle)

Subject: re: training data for [topic from recent paper] Preview: Saw your recent work — a follow-up idea

Hey ,

I came across your recent publication on [Topic, sourced by Origami], and one thing stood out: the mention of limited training data for cell segmentation models.

We just released a validated library of 120k annotated microscopy images across multiple cell lines, specifically built for training segmentation networks. It’s designed to plug directly into common frameworks like Cellpose and DeepLab.

Could something like that accelerate your current project?

No pressure — just thought the timing might be right.

---

Day 7 – Final breakup email

Subject: Closing the loop on AI training data Preview: One last thought

,

I’ll keep this quick. If external training data isn’t a priority right now, I totally understand.

But if you ever need annotated genomics, proteomics, or imaging datasets to train or validate your models, our catalog is designed specifically for research teams like yours. We even offer custom curation if you have a niche target.

I’ll leave this here. If there’s someone else on your team who’s closer to this work, a warm intro would be hugely appreciated.

Thanks,

---

The personalization tokens — [Company], specific topic, recent paper — all get filled by Origami because the platform enriched those details when you built the list. That’s a big reason you can run this sequence and sound like you did hours of research, without actually doing it.

---

Step 4: Send the Sequence Directly from Origami

This is where most tools force you to export a CSV, sync with a separate email platform, and pray the connections don’t break. Origami doesn’t require any of that.

You launch the entire multi-step sequence from the same dashboard where you built and refined the list. Origami’s built-in email sequencer sends your Day 1 email, waits the delay you set (Day 3, Day 7 — or whatever cadence you choose), then sends the follow-ups automatically. No exporting, no syncing.

What happens while it sends:

• Opens, clicks, replies appear in real time on your campaign dashboard. You see exactly which contacts engaged and when.
• Prospect context stays visible. While reviewing a reply, you can still see the enriched profile — title, company, tools used, published work — so you remember why you reached out in the first place.
• Automatic un-enrollment removes anyone who replies. If a researcher writes back on Day 2, the Day 3 and Day 7 emails never fire. No accidental breakup emails after a booked meeting.

Pricing reality: The sequencer itself is included on all paid plans — you’re only paying for the credits you use to enrich leads. Plans start at $29/month, and there’s a free plan (1,000 credits, no credit card) so you can test the entire workflow before committing.

What response rate to expect for biotech researchers: On a list I refined using the tags above and this exact 3-touch sequence, I consistently see 5–10% reply rates. Enterprise pharma researchers tend to be a bit lower (3–5%) unless you time the outreach around a conference or a recent paper. AI-first startups often hit the higher end because they move faster and are actively looking for data partners. If you’re below 2% across the board, the messaging or list quality needs work.

When to iterate on messaging vs. iterate on the list:

• If open rates are strong (above 45–50%) but reply rates are weak, your subject lines and preview text are working, but the body isn’t hitting the right pain point. Test different angles — one angle around data shortages, another around cost of in-house annotation.
• If open rates are low (under 30%), suspect deliverability or list quality. Check that the emails Origami enriched are valid — the platform validates them, but always spot-check a few. Also, segment by domain to see if certain company email gateways are throttling.
• If replies are all polite “not interested,” your “who” might still be right but your “when” is off. Switch from a generic value prop to a trigger-based opener (recent grant, new preprint, hiring announcement) — Origami’s AI agent can generate those triggered variants on the fly.

---