Clay logo, go to homepage

Search Marketplace

Browse all Skills
Verify & cleanSkill

Clean Company Names For Copy

Cleans raw company-name strings into short, natural versions for outreach copy, running a free normalizer first and escalating to AI only when a 20-row review shows the normalizer falls short.

Eric Nowoslawski
Eric Nowoslawski

Founder at Growth Engine X

About this Skill

When to use it

Use this Skill when you need to clean messy company names before dropping them into email copy, stripping suffixes, taglines, and dba entities so the name reads like a person wrote it. It runs a free normalizer first and only escalates to an AI pass when a 20-row review of your own list proves the normalizer cannot fix what is wrong. Avoid using it for finding company facts like funding, hiring, or pricing, since those need separate skills.

What it delivers

You get a cleaned company-name column ready for campaign copy, plus a run summary showing which path ran, how many rows shipped, and where names disagree with domains. The Skill excludes rows with no reliable name from campaigns instead of filling them with a generic placeholder.

SKILL.md preview

Sign in to install and view the complete SKILL.md

Illustrative preview. Sign in for the actual file.

How someone would prompt this

Clean up the company names in my outreach list so they read naturally in email copy, only running the AI pass if my 20-row review shows the free normalizer cannot fix them.

Inputs & outputs

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

Inputs

  • 01Lead data & domain
    • Table, CSV, or audience with the raw company name per row
    • optional bare domain (no www) to catch name/domain mismatches
  • 02Campaign copy details
    • Tells the Skill which sentences name the company (subject line, opener, possessive)
    • decides whether to run at all
  • 03Normalizer review decision
    • Your verdict after reviewing 20 rows of the free normalizer's output
    • decides whether the AI pass runs
  • 04Output & routing settings
    • Column name for the cleaned value (defaults to company_clean)
    • where abstained rows get routed for review
    • whether to keep low-confidence rows by dropping the company clause instead of excluding them
  • 05Model choice
    • The AI model configured in your workspace for the escalation pass
    • sets its token cap based on model class

Outputs

  • 01Cleaned company name columnA row-by-row column with the cleaned name beside the raw one, plus changed and confidence flags, ready for your campaign copy to reference.
  • 02Run summary reportA summary telling you which path ran, how many rows got a clean name, how many rows the Skill excluded, and which rows have a name that does not match the domain.

Representative output

The cleaned column, row by row

company_name_rawcompany_cleanchangedconfidencewhat happened
Ajax Turner Company, Inc.Ajax Turnertruehighlegal suffix stripped — the free path does this
ABN TECH CORPABN Techtruehighcasing, with the acronym protected narrowly
318, Inc dba Hamiltons Bud and BloomHamiltons Bud and Bloomtruehighresolved to the operating brand
AlaMark Technologies \| FileMaker ConsultantsAlaMark Technologiestruehightagline after the pipe dropped
(319) Auto Body(319) Auto Bodyfalsehighthe parentheses ARE the brand
Northstar Example (a Vantage company)Northstar Exampletruehighparenthetical descriptor dropped
accounting business solutionsAccounting Business Solutionstruelowa generic string that is a real name — flagged, not abstained
Self-employedtruehighguard — whole-string placeholder match
Private Practicetruehighguard — mini would have shipped this as a brand
Retired - BTH Banktruehighguard on the OUTPUT — the regex path would have sent Retired

The run summary

path: AI column (20-row read found 6 edits, all model-shaped)
  free normalizer alone scored 14/20 acceptable on the same read

1,000 rows processed
  974 shipped a copy-ready value
   26 abstained and routed to review (placeholder guard: 19 on input, 7 on output)
    0 recorded as an abstain that was really a truncation

  11 rows flagged: cleaned name does not match the domain -- review before sending

copy should reference: company_clean
abstained rows routed to: <the review view the installer named>

How the Skill runs

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

01
Check workspace access

Confirms the table exists and states exactly which columns this run will add; the run stops and names the failing step if the check fails.

Step
02
Decide whether AI cleanup helps

Runs the free normalizer on 20 sample rows and tests three conditions to see if AI is worth it; the run stops and waits for you to confirm the decision.

Step
03
Run the free normalizer

Applies the free deterministic cleaner to every row at no cost; if the review found this enough, this is the final result and the run moves to delivery.

Step
04
Run the AI cleanup pass

Calls an AI model on each row with a locked prompt that strips suffixes, resolves dba names, drops taglines, and abstains on junk entries, with a guard to catch placeholder answers.

Step
05
Review 20 cleaned rows

Shows 20 cleaned names next to their raw originals; the run stops and waits for your review, and gets reworked if more than one needs editing.

Step
06
Exclude rows missing a clean name

Removes rows with an empty or placeholder company name from any campaign that names the company, routing them to your review list instead of guessing a substitute.

Step
07
Deliver results and summary

Reports which path ran, how many rows got a value, how many rows the Skill excluded, and which rows have a name that does not match the domain.

Step

Connections required

No connections listed.

Install Clay and run it

Sign in once, install the Skill, then run the prompt.

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

Version history

Version 2From submitted SKILL.mdCurrent

Related Skills

Clean First Names For GreetingsCleans the raw first_name_raw field into a proper greeting name in a new first_name_clean column, withholding rows that fail six guard checks instead of guessing.View Skill →Icp Outbound CampaignTurns your ideal customer profile into a verified lead list saved in Clay Audiences and a drafted outbound email sequence, without sending or launching anything.View Skill →10k Value Prop MatchCompares a seller's value proposition against a target company's own 10-K language and returns matched initiatives with quotes and need status.View Skill →

Related searches

Verify & clean skills