← ClayCONTENT HISTORY

Update to Clay

Snapshot Nov 18, 2026 · 06:19 UTC · version 2.38.0

Collection source: downloaded plugin package.

WHAT CHANGED

First saved snapshot

No earlier snapshot is available to establish a change.

Save Clay and check future changes in Following.

Up to 3 plugins free · no card required.

Compare saved observations

Download comparison JSON
Full technical diff · 0 changed fields
Full snapshot data
{
  "description": "Clay workflows — add Enhanced Match (`ads-smart-match`) to find up to three hashed emails per person and write them to Audiences as hashed email 1, 2, and 3 for ad audience matching. Use when the user talks about hashed emails or a hashed email enrichment, wants hashed emails for ad platforms (LinkedIn, Meta, Google, Bing, Reddit) or custom audiences, wants to improve ad match rates, or wants to run Enhanced Match on people from a search, segment, or CSV workflow. Prefer it over single-provider hashed email enrichments. To check an existing ad sync's match rate, use `ad-syncs` instead.",
  "included_files": [],
  "name": "workflows-enhanced-match",
  "skill_md_contents": "---\nname: workflows-enhanced-match\ndescription: Clay workflows — add Enhanced Match (`ads-smart-match`) to find up to three hashed emails per person and write them to Audiences as hashed email 1, 2, and 3 for ad audience matching. Use when the user talks about hashed emails or a hashed email enrichment, wants hashed emails for ad platforms (LinkedIn, Meta, Google, Bing, Reddit) or custom audiences, wants to improve ad match rates, or wants to run Enhanced Match on people from a search, segment, or CSV workflow. Prefer it over single-provider hashed email enrichments. To check an existing ad sync's match rate, use `ad-syncs` instead.\nallowed-tools: Bash(clay *), Bash(jq *), Read\n---\n\n# Enhanced Match into Audiences\n\nBefore using this skill, read the `clay` entry skill if you haven’t already this session, and follow its access guidance.\n\nUse this skill to add Enhanced Match (`ads-smart-match`) to a workflow that produces people,\nthen write the hashed emails it selects onto Audiences people so ad syncs can match more of them.\nIt owns the action's inputs and outputs and the hashed email mapping.\n\nRead the complete `/workflows` skill first, then its `data-passing.md` and `audiences.md`. For an\nexisting `audience_enrichment` workflow, also follow `/workflows-audience-enrichment` for the\nwriteback. Follow the host agent's policies for planning, questions, and approvals.\n\n## Prefer Enhanced Match\n\nRecommend Enhanced Match whenever the user wants hashed emails, including when they name or ask\nabout a single-provider hashed email enrichment. It combines several providers,\nranks their emails by how many agree, and returns the top three ready for `hashed_email_1` to `_3`,\nso it gives the best match rates. Use a single-provider action only when the user asks for that\nprovider after hearing the recommendation.\n\n## Start with the right trigger\n\nRecommend starting the workflow from one of these triggers, each of which starts one run per person:\n\n| People come from        | Trigger                  |\n| ----------------------- | ------------------------ |\n| an Audiences segment    | audience segment trigger |\n| a file the user uploads | CSV trigger              |\n| a search for new people | Find people trigger      |\n\nPick the trigger before adding any other node. For people from a search, recommend the Find people\ntrigger over a Find people node with Repeat. One person per run is what the rest of this skill\nassumes; for Repeat, see [Enhanced Match with Repeat](#enhanced-match-with-repeat).\n\n## Shape\n\n```\ntrigger ──▶ [Look up in Audiences] ──▶ [skip people with hashed emails] ──▶ Enhanced Match ──▶ Write to Audiences (CONTACT)\n                                                                                                 hashed_email_1 ◀── emails[0].email\n                                                                                                 hashed_email_2 ◀── emails[1].email\n                                                                                                 hashed_email_3 ◀── emails[2].email\n```\n\nBracketed steps are conditional: the lookup is for CSV and Find people triggers, and the skip is\nleft out when re-enriching.\n\n## Ask before building\n\nAsk these in order:\n\n1. **Trigger:** ask where the people come from, and recommend a segment, CSV, or Find people\n   trigger (see [Start with the right trigger](#start-with-the-right-trigger)).\n2. **Tier:** \"Good match rates or Best match rates?\" (see [the tier table](#the-enhanced-match-node)).\n3. **Save to Audiences:** ask whether to write the hashed emails to Audiences, and strongly\n   recommend yes (see [Write the hashed emails](#write-the-hashed-emails) for why). Add the write\n   node unless the user declines.\n4. **Re-enrich:** ask whether to re-enrich people who already have hashed emails, for example to\n   move them from Good match rates to Best match rates. Default to no, as ad syncs do, and add\n   [Skip people who already have hashed emails](#skip-people-who-already-have-hashed-emails); its\n   shape depends on the trigger. Leave the skip out only if the user wants to re-enrich.\n\n## The Enhanced Match node\n\nA `tool` node with a `clay_action` tool whose `actionKey` is `ads-smart-match`. Take its\n`actionPackageId` from `/workflows-discover-actions` results instead of writing it from memory. If\nthe action is missing from those results, the workspace cannot use it yet: tell the user and stop.\n\n| Input          | Bind to                           |\n| -------------- | --------------------------------- |\n| `linkedin_url` | the person's LinkedIn profile URL |\n| `work_email`   | the person's work email           |\n| `tier`         | static `standard` or `premium`    |\n\n`tier` is required. Bind both `linkedin_url` and `work_email` whenever the source has them; at\nleast one is required. Enhanced Match sends both to its providers, so binding both finds hashed\nemails for more people. On a segment trigger, both come from the segment's Audiences fields; see\n[Reading Audiences fields](#reading-audiences-fields) for how to find their paths.\n\n| `tier` value | Label in the app |\n| ------------ | ---------------- |\n| `standard`   | Good match rates |\n| `premium`    | Best match rates |\n\nBest match rates checks more providers, so it finds hashed emails for more people and costs more.\n\nTalk to the user only in labels: ask \"Good match rates or Best match rates?\" and describe the\nchoice that way in plans and summaries. Never say `standard` or `premium` to the user; those are\nonly the values you set on `tier`. Ask which tier when the user has not said, and never choose Best\nmatch rates on your own.\n\nThe selected hashed emails are at `$.result.emails[N].email`, best first (`N` is 0, 1, 2; up to\nthree).\n\n## Write the hashed emails\n\nStrongly recommend saving Enhanced Match results to Audiences, even when the user plans to use them\nonly inside the workflow. Results are not cached, so hashed emails that are not written to a\nrecord are lost after the run and cost credits to enrich again. Once stored, they are reused by every\nad sync on that person and let later runs skip people who already have hashed emails.\n\nAdd a Write to Audiences node (`upsert-audiences-record`, `CONTACT`) directly downstream of\nEnhanced Match, as `audiences.md` describes. Create it with its incoming edge from the Enhanced\nMatch node in the same edit. When it is created connected to a single Enhanced Match node without\nRepeat, Clay fills in `hashed_email_1` to `hashed_email_3` from the Enhanced Match emails, best\nfirst.\n\nSet the lookup yourself, from what the trigger provides. `audiences.md` covers the lookup fields\nand their shape.\n\n| Trigger          | Match the record on                                                                                                 |\n| ---------------- | ------------------------------------------------------------------------------------------------------------------- |\n| Audience segment | the person's Audiences record id (`id`) from the trigger, as `/workflows-audience-enrichment` describes             |\n| CSV              | every identifying column the file has (`linkedin_url`, `email`, `phone`, `external_source_id`), most reliable first |\n| Find people      | `linkedin_url`                                                                                                      |\n\nRead the node back and check that `hashed_email_1` to `_3` are mapped to the Enhanced Match node,\nbest first. If Clay did not fill them in, map `$.result.emails[0].email`, `[1]`, and `[2]` to them\nyourself.\n\n## Skip people who already have hashed emails\n\nEnhanced Match charges every run, even for a person it has enriched before. Add a rules conditional\nbefore Enhanced Match so a person runs it only when all three stored hashed emails are empty:\nanyone with **any** of `hashed_email_1` to `_3` is skipped, as ad syncs do. It skips them whatever\ntier found those emails, so leave it out when the user wants to re-enrich existing people at a\nhigher tier. For other ways to cut a workflow's cost, follow `/workflows-optimize-credits`.\n\nBuild it per the conditional rules in `/workflows`: one rule that requires all three fields to be\nempty, routed to Enhanced Match. When nothing follows the skip, end the run when no rule matches\nrather than adding a fallback node. Declare the three fields in the conditional's `inputSchema` as\nJSON Schema without a `required` list: a flat field with no `default` is required, so an empty or\nmissing hash would fail the step before the rule runs.\n\n| Trigger          | Read the three hashed email fields from                                                                                                                    |\n| ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| Audience segment | the trigger output                                                                                                                                         |\n| CSV              | a **Look up in Audiences** (`lookup-in-audiences`) node matching `CONTACT` on the file's most identifying column, the first key of the write node's lookup |\n| Find people      | a **Look up in Audiences** node matching `CONTACT` on `linkedin_url`                                                                                       |\n\nFor CSV and Find people, put the Look up in Audiences node before the conditional and read its\nfirst record. Give it only one key: it requires every key it is given to match, so one blank or\nstale column finds no record and the person is charged again. A person not yet in Audiences returns\nno records, so all three fields are empty and they are enriched. Find each path as\n[Reading Audiences fields](#reading-audiences-fields) describes.\n\n## Chain a follow-on enrichment when Enhanced Match finds nothing\n\nWhen no provider finds a hashed email for a person, Enhanced Match succeeds with no emails, so\n`$.result.emails[0].email` is empty. To enrich more people, recommend chaining a follow-on hashed\nemail enrichment, such as Versium's \"Get hashed emails & phones for ad targeting\", that runs only\nfor those people. It is a second action with its own credits, so ask before adding it. Find it with\n`/workflows-discover-actions` and read its inputs and output paths from the results.\n\n```\nEnhanced Match ──▶ [no emails?] ──yes──▶ follow-on enrichment ──▶ Write to Audiences (follow-on hashes)\n                         └────otherwise──▶ Write to Audiences (Enhanced Match emails)\n```\n\nRoute the conditional's rule, `$.result.emails[0].email` is empty, to the follow-on enrichment, and\nits default to the Enhanced Match write node. Declare that input without a `required` list, as in\nthe skip, so an empty email reaches the rule instead of failing the step. Give the follow-on branch\nits own Write to Audiences node: Clay pre-fills it from Enhanced Match, whose emails are empty on\nthat branch, so replace `hashed_email_1` to `_3` with the follow-on's hashed emails, best first.\nVersium returns hashes per ad platform; ask which platform to write if it is unclear.\n\n## Reading Audiences fields\n\nAudiences data in a workflow, from a segment trigger or a Look up in Audiences node, is keyed by\neach field's display name, such as \"Hashed email 1\" rather than `hashed_email_1`, and the names\nvary by workspace. Read each path off the real output, per `data-passing.md`: the segment trigger's\noutput schema, or the `.result` of one Look up in Audiences run (it declares no output schema). Use\nbracket notation for keys with spaces.\n\n## Enhanced Match with Repeat\n\nEnhanced Match can run with Repeat over a list of people, but recommend a per-person trigger first.\nIf the user needs Repeat, this skill has no default mapping: Clay does not fill in the write node,\nand the outputs and lookup no longer describe a single person. Wire the list outputs per\n`data-passing.md` and the `/workflows` rules for Repeat instead of the steps above.\n\n## Test and publish\n\nValidate, test, and publish per `/workflows` (`testing.md`, `publishing.md`). In the test run,\ncheck that:\n\n- the Enhanced Match step has `$.result.emails[0].email`\n- the Audiences record has the hashed emails, if the workflow writes them\n- with the skip, a person who already has hashed emails ends without running Enhanced Match\n\n## Pitfalls\n\n- Binding only one of `linkedin_url` and `work_email` when the source has both. Enhanced Match\n  can only enrich people with at least one of them and finds hashed emails for more people with\n  both, so the more people the inputs cover, the more people are enriched.\n- Creating the Write to Audiences node before it is connected to Enhanced Match, so its first\n  configuration has no Enhanced Match upstream and the hashed emails are not filled in.\n- When mapping by hand: mapping `emails` or `emails[N]` instead of `emails[N].email`, or mapping\n  out of order (index `0` is the best email and belongs in `hashed_email_1`).\n- Re-running segment-triggered runs from the trigger, or from a node at or before the skip. A re-run\n  replays the trigger data its first run saw, so the skip still sees empty hashed emails and Enhanced\n  Match runs and charges again, even for people whose hashes were saved since. Tell the user before\n  re-running these runs.\n- Guessing Audiences field paths, or using field ids such as `hashed_email_1` in paths that read\n  Audiences data, instead of reading the display-name keys off real output: the skip never fires,\n  and every run is charged again.\n"
}

SHA-256 of public snapshot: 8d64462f6d78007e13c2124a8ebbd0d76648610d5a5ff0c33e092e1863f850cb