\nWhich AI tools use variables<\/a>
\nAdd variables to prompts<\/a>
\nHow column type affects variables<\/a><\/p>\n
- \n
- Single columns: direct reference or personalization<\/a><\/li>\n
- Bulk columns: AI analysis and new insights<\/a><\/li>\n<\/ul>\n
\n![\"learn\"](\"https:\/\/snov.io\/knowledgebase\/wp-content\/themes\/knowledgebase-n\/assets\/img\/fluent_chat-warning-16-regular.svg\")
<\/div>\n\nBefore you start<\/strong><\/p>\n
To follow this guide, make sure you have:
\n1. Existing prospect data<\/a>.<\/strong> Enrichment variables reference data already added to your list<\/p>\n2. Tokens for Enrichment & AI tools<\/a>.<\/strong> Variables are added inside a prompt for AI-based tools<\/p>\n<\/div>\n<\/div>\n
What are enrichment variables for<\/strong><\/h2>\n
Variables reference existing data (columns\/fields) and pull their value into the prompt for an AI agent.<\/p>\n
Using variables, existing data can act as search context or input for generating new insights and improving your own data.<\/p>\n
\n![\"tip\"](\"https:\/\/snov.io\/knowledgebase\/wp-content\/themes\/knowledgebase-n\/assets\/img\/academicons_ideas-repec.svg\")
<\/div>\nCampaign variables vs. enrichment variables:
\n<\/strong>Campaign variables<\/a> are for personalizing email and LinkedIn messages using existing data. Enrichment variables are for using existing data across other enrichments.<\/div>\n<\/div>\nWhich AI tools use variables<\/strong><\/h2>\n
Variables are used across AI-powered enrichment tools that rely on your prompts to generate results.<\/p>\n
1) Fill out from web (Web Search)<\/strong><\/h4>\n
Use data from existing fields as a search context or as a reference to generate new insights from it or as input to qualify prospects better.<\/p>\n
2) Clean and edit<\/strong><\/h4>\n
Specify existing fields as the data you want to update or improve.<\/p>\n
![\"\"](\"https:\/\/snov.io\/knowledgebase\/wp-content\/uploads\/2026\/08\/SCR-20260819-jlro.png\")
<\/a><\/p>\nAdd variables to prompts<\/strong><\/h2>\n
To add a variable to your enrichment prompt, type {{ }}<\/strong>\u00a0or click the {}<\/strong> icon, then select a field.<\/p>\n
Correctly added variables appear as a colored tag.<\/p>\n
\n<\/div>\nTo start any AI enrichment, your prompt must include at least one variable. Fields of any type (native, custom, or enrichment) can be used as variables.<\/div>\n<\/div>\n![\"\"](\"https:\/\/snov.io\/knowledgebase\/wp-content\/uploads\/2026\/08\/SCR-20260819-jmwj.png\")
<\/a><\/div>\nHow column type affects variables<\/strong><\/h2>\n
Depending on the data stored, there are single <\/strong>(one value)\u00a0and bulk <\/strong>(multiple values)\u00a0columns in your prospect list.<\/p>\n
\n<\/div>\nThe column type you reference in a variable determines what data you receive. Learn more about data fields and column types in this guide<\/a>.<\/div>\n<\/div>\nSingle columns: direct reference or personalization<\/strong><\/h3>\n
Single columns work best when you need a single value. They return one value that you can use directly in a prompt or personalization.<\/p>\n
In AI enrichment prompts<\/strong><\/h4>\n
Pass one data point to give the AI context:<\/p>\n
\n\n\"The prospect is a {{Job Position}} at {{Company Name}}. Write a one-sentence icebreaker.\"<\/p>\n
\"Find recent events and interviews where {{Prospect Name}} discussed their industry.\"<\/p>\n<\/div>\n<\/div>\n
In emails or LinkedIn messages<\/strong><\/h4>\n
Insert the value into your campaign content:<\/p>\n
\n\"Hi {{First Name}}, I noticed {{Company Name}} is expanding its sales team\u2026\"<\/div>\n<\/div>\nBulk columns: AI analysis and new insights<\/strong><\/h3>\n
Bulk columns contain rich data. They work best when you want the AI to analyze or summarize multiple values, not just insert text.<\/p>\n
When your variable references a bulk column in an AI prompt, it passes the full dataset (with multiple values). The AI receives the complete data, then uses your prompt to generate results.<\/p>\n
\n![\"note\"](\"https:\/\/snov.io\/knowledgebase\/wp-content\/themes\/knowledgebase-n\/assets\/img\/ph_circle-wavy-warning.svg\")
<\/div>\n\nUsing bulk columns in variables may not always be necessary or efficient for best results. Extract a value you need<\/a> into a separate column, then reference that column in your prompt.<\/p>\n<\/div>\n<\/div>\n
In AI enrichment prompts<\/strong><\/h4>\n
Pass the full dataset and let the AI extract insights:<\/p>\n
\n<\/div>\nIf you reference a bulk column in your prompt, add instructions to pull what you need first.<\/div>\n<\/div>\nPrompts for Clean and edit:<\/strong><\/p>\n
\n\n\"Here's the prospect's full work history: {{LinkedIn Experience}}. Summarize their career and identify their top three areas of expertise. Based on experience, identify the prospect's current company and job title.\"<\/p>\n<\/div>\n<\/div>\n
In emails or LinkedIn messages<\/strong><\/h4>\n
Bulk columns aren't meant for personalization variables.<\/p>\n
Run an AI prompt that extracts the specific detail you need into a single column, then use that single column as a variable.<\/p>\n
\n\nRelated articles<\/strong><\/h3>\n
Enrichment tools in Snov.io<\/a>
\nHow to enrich data in your prospect list<\/a>
\nProspect data fields and column types<\/a><\/p>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"This guide explains what enrichment variables are and how to use the data that already exists in your prospect lists in AI enrichment prompts to get a custom result. What are enrichment variables for Which AI tools use variables Add variables to prompts How column type affects variables Single columns: direct reference or personalization Bulk […]<\/p>\n","protected":false},"author":16,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[53],"tags":[],"_links":{"self":[{"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/posts\/48252"}],"collection":[{"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/users\/16"}],"replies":[{"embeddable":true,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/comments?post=48252"}],"version-history":[{"count":14,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/posts\/48252\/revisions"}],"predecessor-version":[{"id":48338,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/posts\/48252\/revisions\/48338"}],"wp:attachment":[{"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/media?parent=48252"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/categories?post=48252"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/snov.io\/knowledgebase\/wp-json\/wp\/v2\/tags?post=48252"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
- Bulk columns: AI analysis and new insights<\/a><\/li>\n<\/ul>\n