Swap Test

Check if a cold email is template or truly personalized

Checks how much of a cold email is template versus real fact, labeling each sentence and rewriting the generic ones with sourced details about the prospect or sender. Running it on a batch also catches sentences that read the same across multiple prospects.

Matteo Fois

Co-Founder at Kinetyca

About this Skill

When to use it

Use this Skill when you want to know if a cold email or ABM email is actually personalized or just a template with names swapped in, and get it rewritten using real, sourced facts about the prospect and sender. Avoid using it for building prospect research briefs, picking case studies, checking deliverability wording, or writing a sequence from scratch.

What it delivers

You get a labeled sentence-by-sentence breakdown, a pass or fail verdict, and a rewritten email that spends only facts tracing back to a real source. The Skill flags any sentence that would read the same for a different prospect, and on a batch it also catches sentences that repeat across prospects after it masks their names.

SKILL.md preview

# SKILL.md

The full Skill file is available after sign-in.
This is an illustrative preview, not the source.

## Inputs and setup

Required inputs and setup details appear here.
Sign in to read the complete instructions.

## Workflow instructions

The Skill's execution steps appear here.
This placeholder contains no actual Skill content.

## Outputs and examples

The full file includes its source documentation.
Sign in to view and copy the original file.

## Notes and requirements

Additional source details appear after sign-in.
The actual SKILL.md is not loaded in this preview.

Illustrative preview. Sign in for the actual file.

How someone would prompt this

Check if this cold email to a prospect is actually personalized or just a template with their name swapped in, and rewrite any generic lines using the facts from their lead data.

Inputs & outputs

What someone provides before the Skill runs, and the reviewable artifacts it returns.

Inputs

Email to testThe cold email or batch of rendered emails you want checked, subject line included.
Lead and sender facts
- Prospect enrichment data from a Clay table, CSV, database export, or sheet, with facts cited by column or note
- sender's checkable claims such as results, named clients, delivery mechanics, and guarantees, each with a source.
Test settings
- Choose account mode (email is for one prospect) or segment mode (email is for a group)
- set the maximum allowed percentage of template sentences or use the author's defaults.
Banned phrase lists
- Default lists of asks, openers, hedges, noun-swap phrasings, and decoration lines that count as template on sight
- you may add to these lists but not remove entries silently.
Batch prospect list
- Each prospect's name, company, and industry for a batch run
- used to mask those words and compare rewritten sentences across prospects for hidden repetition.

Outputs

Pass or fail verdictTells you whether the email passes or fails the personalization test and states the exact reason, such as unsourced claims or too many generic sentences.
Sentence-by-sentence breakdownShows every sentence with its label, the fact and source behind it, and the before-and-after wording so you can see exactly what changed and why.
Batch template checkReports any sentence that reads identical across multiple prospects after masking names, since a repeated sentence is template no matter what it says.
Coverage summary lineGives a quick count of sentences, exemptions, template share before and after rewrite, and facts spent, so you can judge the rewrite at a glance.

Representative output

Sentence table

#	Before	Label	Why	After	Fact
1	Hey Dana - we help EdTech companies hire sales talent with district relationships.	slot	noun-swap frame: "We help [title] [outcome]"	Contoso posted two district-sales openings in March, both asking for K-12 procurement experience.	B-2 (careers page, 2026-03-14)
2	EdTech is the only space we recruit for.	slot	by the question: true for any prospect	Contoso's 2026 roadmap names Texas and Florida as the two expansion states.	B-4 (investor update, 2026-02)
3	We've placed a lot of people in this space.	slot	hedge where a number should be	Fabrikam Talent placed 28 contractors at a K-12 vendor in under two weeks.	S-1 (case study, named)
4	Our clients see results within 3 weeks.	unsourced	no proof row carries "3 weeks"	(cut)
5	Open to learning more?	slot	banned ask	Want the two-page version with the district-procurement screen we use? Reply yes.	S-3 (deliverable confirmed)
6	PS A few companies we've worked with: Alpha, Beta, Gamma, Delta.	slot	decoration list	PS The shortlist arrives in 5 business days.	S-2 (mechanics)
6b	(from the same PS)			If a hire is not K-12-fluent in week one, we replace them at no cost.	S-4 (guarantee)

Batch check (12 prospects)

2 sentences came out identical for 5 prospects after masking ("Your roadmap names two expansion states") and were relabelled slot; 5 emails returned to Step 5 and spent a different lead data column. Opener frame shared across the batch after masking: 0 of 12.

Coverage line

6 sentences counted · 2 exempt · slot before 6 · slot after 0 · facts spent 6 · facts unspent 2 · unsourced claims cut 1.

How the Skill runs

Follow the steps in order. Each row includes the full instruction.

Confirm setup and explain scope

Confirms the email and any data files open, checks Clay access if lead data lives in a table, and states plainly that this Skill only reads and analyzes, making no changes outside two output files.

Collect all inputs first

Gathers the email, mode, lead data, sender proof, slot line, and banned phrase lists before reading a single sentence, since setting these after seeing a score would defeat the test.

Split and mark sentences

Breaks the email into numbered sentences and marks greetings, sign-offs, and plain opt-out lines as exempt from scoring.

Label every sentence

Labels each sentence as template, prospect fact, sender fact, or unsourced, then stops and waits for you to confirm or correct the labels before continuing.

Score the pass or fail verdict

Applies fixed rules in order: unsourced claims fail first, then too many template sentences, then zero prospect facts in account mode; anything else passes.

Rewrite failed sentences

Replaces or deletes template and unsourced sentences using only sourced facts, then stops and shows the rewrite for you to review before continuing.

Check for repeated sentences

Masks prospect names and industries across a batch, flags any sentence that reads identical for two or more prospects as template, and sends failing emails back for another rewrite.

Deliver the final files

Writes the sentence table and rewritten email or batch to two files in your working directory, ready to take to your sending tool.

Connections required

No connections listed.

Install Clay and run it

1Install ClayUse the connector or official plugin in your coding-agent environment.
2Sign up or log inAuthorize Clay access and return when the connection is ready.
3Install the SkillRun the one-line install command from the Add this Skill card.
4Run the SkillPaste the example prompt and review the result.

Version history

Version 2From submitted SKILL.mdCurrent