Optimize · Subject-Line & A/B Variants

Test Personalization Token Subject Lines

Test whether {first_name}, {company}, {custom_field}, or no token wins on reply rate for your sequence.

managerrepAdvanced⏱ 1-2 hours
When to use
Use this when your team defaults to {first_name} in every subject line out of habit. Different ICPs respond to different tokens — some pop on {company}, some are immune to {first_name} because every outbound tool uses it. This prompt designs a 4-arm test (no-token / first_name / company / custom_field) so you stop guessing.
The prompt
You are a sales operations lead at a digital marketing agency. You know that overusing {first_name} in 2026 is now a negative signal for sophisticated buyers.
Agency: [AGENCY_NAME] — [SERVICES] | Email body: [EMAIL_BODY] | ICP: [ICP_DESCRIPTION] | Custom personalization field you can pull: [CUSTOM_FIELD] (e.g. their top traffic page, their stack, their last hire) | Current control subject: [CURRENT_SUBJECT]
Design a 4-arm subject line test that isolates the effect of personalization tokens. Same underlying message across all 4 — only the token changes.

- Max 6 words per subject
- Arm 1: no token (baseline). Arm 2: {first_name} only. Arm 3: {company} only. Arm 4: {custom_field} ([CUSTOM_FIELD]) only
- Lowercase, no emoji, no spam triggers
- Custom field token must reference something real per prospect — note enrichment requirement
- Identify which token type is highest-risk for [ICP_DESCRIPTION] and why

Table: Arm | Subject | Token used | Hypothesis (1 line). Below the table: (a) which arm you'd bet on for this ICP, (b) the enrichment work needed to ship Arm 4 at scale, (c) one trap to avoid per arm.
Variables
  • [AGENCY_NAME] — Your agency
  • [SERVICES] — Services
  • [EMAIL_BODY] — Email body
  • [ICP_DESCRIPTION] — 1-line ICP
  • [CUSTOM_FIELD] — The deep custom field, e.g. 'their top organic landing page'
  • [CURRENT_SUBJECT] — Current control subject
Example input
Agency: NorthLoop — SEO for B2B SaaS | Email body: offering an audit of their top 3 commercial pages | ICP: Series A-B SaaS founders | Custom field: their #1 ranking commercial page URL | Control: 'quick seo question for {company}'
Example output
| Arm | Subject | Token | Hypothesis |
|---|---|---|---|
| 1 | seo audit, 3 pages | none | controls for token fatigue |
| 2 | maya, seo on 3 pages | {first_name} | classic, may feel templated |
| 3 | klario's top 3 pages | {company} | feels researched |
| 4 | /pricing is ranking #6 | {custom_field} | undeniably specific |

Bet: Arm 4 for Series A SaaS founders — they reward observable research. Enrichment: pre-pull #1 commercial URL per account via Ahrefs/Semrush API before send; ~30 sec/account. Traps: A1 looks robotic at scale, A2 burns trust on sophisticated buyers, A3 fails if {company} is awkwardly cased, A4 fails silently if URL is wrong — QA every row.
Pro tips
  • {first_name} subjects are now neutral-to-negative for senior SaaS buyers — test before defaulting.
  • Custom field tokens are the only personalization that still moves reply rate in 2026 — invest in enrichment.
  • QA your custom field rows manually for the first 50 sends — one wrong URL kills the whole campaign's trust.
Works with
ClaudeChatGPTGemini
Done with prompts? Time to install the system
Book a STAOS call
Related prompts