TL;DR: Smart Columns are most powerful when stacked. These recipes show common stacks, the columns to use, the order to run them in, and the per-row credit cost.
Smart Columns Recipes
Each recipe is a stack of Smart Columns you can add to a lead list. The columns are listed in the order you should run them — earlier columns often produce data that later columns rely on.
Recipe 1 — Clean a messy import
Goal: Tidy up a CSV import before you start outbound.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | clean_first_name | 0.1 | Normalize capitalization, accents, suffixes |
| 2 | format_phone | 0.1 | Convert to E.164 |
| 3 | data_quality_name_check | 0.5 | Flag rows with garbage names |
| 4 | data_quality_domain_check | 0.5 | Flag personal / disposable / catch-all domains |
| 5 | email_validation | 0.2 | Re-verify deliverability |
Total: ~1.4 credits/row.
After running, filter to data_quality_name_check = ok AND email_validation = reliable and you have a clean working list.
Recipe 2 — Research a target account list
Goal: Build context on every account before SDRs reach out.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | enrich_company | 0.5 | Industry, headcount, founded year, social links |
| 2 | find_competitors | 0.5 | Top competitors (useful for differentiation in outreach) |
| 3 | find_similar_companies | 0.5 | Lookalike accounts to add to the list |
| 4 | website_analysis | 0.5 | Summarize positioning, products, ICP signals |
Total: 2.0 credits/row.
Layer the optional linkedin_research column (0.5/row) if you want a contact-level summary too.
Recipe 3 — ICP scoring + outreach prep
Goal: Score every lead against your ICP, then draft a personalized opener for the high-fit ones.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | enrich_company | 0.5 | Provide context for ICP scoring + email drafting |
| 2 | icp_fit_analysis | 0.5 | Score against your ICP profile (see ICP Profiles) |
| 3 | Filter: keep only rows with high fit | — | Save credits on the next column |
| 4 | linkedin_research | 0.5 | Personal context for the email |
| 5 | cold_intro_email | 3.0 | Generate the actual opener |
Total: 1.0 credits/row for the first pass, +3.5 credits/row on the filtered subset.
The filter step is critical — if you skip it, you'll spend 3 credits per row on the most expensive column for leads you'd never send to anyway.
Recipe 4 — Find more accounts like your best customers
Goal: Expand your TAM by finding lookalikes of your best customers.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | Generate ICP from the list of best customers | — | One-time setup (see ICP Profiles) |
| 2 | find_similar_companies | 0.5 | Each row produces a list of lookalikes |
| 3 | Add the lookalikes to a new list | — | Use Add to list action |
| 4 | enrich_company on the new list | 0.5 | Get firmographics |
| 5 | icp_fit_analysis on the new list | 0.5 | Score the lookalikes against your ICP |
Total: 1.5 credits/row across the two lists.
Recipe 5 — Tag leads by buying intent
Goal: Sort leads into buckets like "ready to buy", "researching", "not now."
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | enrich_company | 0.5 | Context |
| 2 | website_analysis | 0.5 | Positioning + recent activity |
| 3 | custom_classification | 1.0 | Bucket into your categories |
For step 3, use a custom classification prompt like:
Given the company background and website summary below, classify the buying readiness into one of:
- ready_to_buy
- evaluating
- researching
- not_a_fit
Company: {{company.name}}
Background: {{enrich_company.summary}}
Website: {{website_analysis.summary}}Total: 2.0 credits/row.
Recipe 6 — Timezone-friendly outreach
Goal: Schedule outreach so it lands during the recipient's working hours.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | contact_timezone | 0.5 | Infer timezone from city / state / country |
| 2 | Sort the list by timezone | — | Now your SDRs can dial geographically |
Total: 0.5 credits/row.
Pair this with a Custom AI column that drafts a time-of-day-appropriate opener if you want to go further.
Recipe 7 — Custom field extraction
Goal: Pull a specific fact from each company that doesn't have a built-in column.
Example: extract the company's CRM (HubSpot, Salesforce, Pipedrive, etc.) from their website.
| Order | Column | Cost / row | Why |
|---|---|---|---|
| 1 | website_analysis | 0.5 | Get a summary of the website |
| 2 | custom_ai (prompt to extract CRM) | 1.0 | Pull the specific fact |
Custom AI prompt:
Based on the website summary below, identify which CRM the company appears to use.
Reply with one of: hubspot, salesforce, pipedrive, zoho, other, unknown.
Only output the value, no explanation.
Website summary: {{website_analysis.summary}}Total: 1.5 credits/row.
This pattern works for any extracted field — pricing tier, hiring status, tech stack, etc.
Tips that apply to every recipe
- Filter before expensive columns. The cheapest credits are the ones you don't spend.
- Re-run failed only. If 30% of rows failed because a third-party API hiccupped, re-run just those rows from the column header menu.
- Test on a sample first. Right-click → Run on selection → pick 5 rows and see the output before committing.
- Combine with Action Columns. After scoring with
icp_fit_analysis, use an Action Column to push only high-fit rows to your CRM.