§0 The Unify 25M-Email Playbook · 10 Rules We Just Adopted
The findings (Unify Anatomy Report · May 2026)
1. Connector subjects win. "DXDT <> {{company_name}}" = top reply rate of any format tested.
2. Short subjects (under 4 words) come second. Especially with a competitive or product hook.
3. One variable = +30% reply lift. Stacking variables drops performance below zero-variable.
4. Lowercase subjects = +6% lift. Across millions of sends.
5. Emoji in subject = +38% lift. Only 5% of teams use it. Biggest whitespace.
6. Body length is U-shaped. Under 25 words OR over 150 words win. 25-75 = dead zone (where 59% of all emails sit). v1 sits at 50-90 = dead zone.
7. Signal-driven opener = 2x reply rate. Reference a real event (funding, hire, launch). Only 3.8% adoption.
8. $ figures in cold = -15% drop. Sounds like an ad. Lead with relevance, save numbers for reply.
9. One question > three. Multi-question emails feel like surveys.
10. Calendar link = 1.6x pipeline opportunity. Frame as convenience ("if easier"), not directive.
Bonus: "Hey {{first_name}}" beats "{{first_name}}," · P.S. lines = -10% · AI personalization = +57% · Deep research = 4x baseline.
§0.5 v1 vs v2 · What Changed
| Lever | v1 (Tier-A only) | v2 (Unify-applied) |
|---|---|---|
| Subject A | "extend runway" (lowercase ✓) | "dxdt <> {{company_name}}" (connector, top format) |
| Subject B | "$50-100k+ credits" | "👋 cloud question" (emoji, +38%) |
| Body length | 50-90 words (dead zone) | Bimodal: A = under 25w · B = over 150w |
| Greeting | "{{first_name}}," | "Hey {{first_name}}," (+casual) |
| Opener | "I know you get a lot of these" | Signal: "Saw {{company}}'s Series B" / "Saw you're hiring {{role}}" |
| $ figures | "$50-100k+ in credits" | Removed. Reframed as "programs you qualify for" (saves dollar talk for reply) |
| CTA | "Worth a look?" | "Worth comparing notes? Calendar if easier: [link]" |
| Calendar link | None | Included on every E1 (+1.6x opp) |
| Questions | Mostly single ✓ | Single, enforced |
| P.S. | None ✓ | None ✓ |
Trade-off to test
v2 drops the "$50-100k+ credits" Tier-A winner (GG W-rate 57%) per Unify's -15% $-figure finding. This is a real conflict in the data. Avishay's GG corpus says credits-number works; Unify's 25M says $-figures hurt. Possible reconciliation: GG audience is more financially-literate founders who treat the number as proof; Unify general B2B sample doesn't. A/B test will resolve.
A Cohort A · Funding Map
dxdt <> {{company_name}}Hey {{first_name}},
Saw {{company_name}}'s Series {{round}}. Most scaleups your stage qualify for two or three hyperscaler funding programs the AWS rep never surfaces.
Worth comparing notes? Calendar if easier: [calendar_link]
👋 funding questionHey {{first_name}},
Saw {{company_name}} closed Series {{round}} in {{quarter}}. Quick context on why I'm reaching out.
We work with Series A and B SaaS scaleups across {{geo}} on hyperscaler funding programs. MAP, RAMP, Founders Hub, ISV Accelerate. Most founders at this stage qualify for two or three of these and never hear about them, because the AWS rep is comped on consumption and the Google side is structured the same way.
We're a certified AWS and Google partner. The way we work: we map the programs that apply to your stage and cloud commit, then walk you through the claim. We're paid by AWS and Google on the co-sell side, so it doesn't cost you anything.
MAP alone usually covers a meaningful slice of first-year cloud spend. The rest layer on depending on commit and product stage.
Worth comparing notes on what similar Series {{round}} teams in {{geo}} are claiming? Calendar if easier: [calendar_link]
Eligibility rules for these programs contradict each other depending on which page you land on. We mapped the overlap. Want the one-pager?
🚧 before you migrateHey {{first_name}},
Most founders find out about AWS MAP after migrating. By then the window has closed.
If your migration is still ahead, worth a quick look?
B Cohort B · Hiring Signal
dxdt <> {{company_name}}Hey {{first_name}},
Saw you're hiring {{hiring_role}}. That role usually inherits cloud waste no one flagged. We map it before they start.
Worth a look? [calendar_link]
🛠 before {{hiring_role}} startsHey {{first_name}},
Saw {{company_name}} is hiring {{hiring_role}}. Quick context.
When we've gone back to founders six months after a new SRE or Platform hire started, most say the first quarter went to cleanup work nobody had flagged. Old commitments, idle resources, oversized storage tiers, the kind of thing that's invisible until someone goes looking.
We're a certified AWS and Google partner. Our AI runs on funded scaleups before the new hire starts and surfaces the same artifacts an SRE team would produce over a couple of months. They walk into a cleaner environment and their first quarter goes to the roadmap you actually hired them for.
We're paid by AWS and Google on the co-sell side, so it doesn't cost you anything.
Worth comparing notes on what we usually find right before someone like {{hiring_role}} starts? Calendar if easier: [calendar_link]
Most new SRE hires spend their first month finding cloud waste instead of shipping roadmap. Want what we typically find on similar scaleups?
⏱ {{hiring_role}} start dateHey {{first_name}},
When {{hiring_role}} starts, cloud is exactly as the last setup left it.
Worth mapping the inefficiencies first? [calendar_link]
C Cohort C · DXDT Audit
dxdt <> {{company_name}}Hey {{first_name}},
Saw {{company_name}} runs {{detected_stack}}. That stack is the #1 source of unflagged cloud waste at scale.
Worth a scan? [calendar_link]
👋 {{detected_stack}} questionHey {{first_name}},
Saw {{company_name}} is running {{detected_stack}}. Quick context on why that caught my attention.
In our scans, that stack at your headcount is consistently where the largest unflagged cloud spend sits. Idle warehouses, oversized k8s clusters, storage tier mismatches, retention windows nobody touched in two years. The kind of thing that doesn't show up on a monthly report because everything's "working".
We're a certified AWS and Google partner. We've built an AI that runs cost optimization and security checks in one pass on funded scaleups. Same kind of work an SRE team would normally do over a couple of months, surfaced much faster.
We're paid by AWS and Google on the co-sell side, so it doesn't cost you anything.
Worth comparing notes on what we typically find on similar {{detected_stack}} stacks? Calendar if easier: [calendar_link]
Ran the same scan on a similar {{detected_stack}} scaleup last month. About a fifth of their cloud was unused.
Want what we found?
🔍 one passHey {{first_name}},
Cost optimization and security checks in one pass. Same artifacts an SRE team would produce in months.
Worth a look? [calendar_link]
§4 v2 Locked Rules (Unify-derived)
Hard rules
• Bimodal body length: Variant A under 25 words. Variant B over 150 words. Nothing in between.
• One variable per subject. Don't stack.
• Connector OR emoji subject. Lowercase. Under 4 words if not connector.
• "Hey {{first_name}}," greeting. Comma after, not period.
• Signal-driven opener. Reference {{round}} / {{hiring_role}} / {{detected_stack}} in line 1.
• One question only. Never two, never three.
• No $ figures in cold E1. Save credits-numbers for reply or auto-reply intake.
• Calendar link on every E1. Framed as "if easier" not "book here".
• No P.S. Anything important goes in the body.
• 2-touch + 1 new thread. Day 0 + Day 3-4 threaded + Day 7-8 new subject.
Variables required (signal data must be enriched BEFORE upload)
{{first_name}} · {{company_name}} · {{round}} (A/B) · {{quarter}} (A) · {{geo}} (A) · {{hiring_role}} (B) · {{detected_stack}} (C, from BuiltWith)
If any of these is null for a lead, fall back to v1 copy or suppress that lead from v2 test.
The big bet
v2 drops two GG Tier-A winners (extend runway subject, $50-100k+ credits phrase) in favor of Unify's 25M-email general findings. Risk: Avishay's audience may be different from Unify's sample. Mitigation: Run v1 and v2 in parallel on equal-sized cohort splits, compare 14-day reply rates, kill loser.