Clean an Apify contact scrape into a call sheet: one row per business, best email and phone first

Updated 14 Oct 2026 · Disclosure: the contact scraper linked below is an Unadhi product on the Apify Store. This guide is free, and the sheet works with a CSV from any tool or a list you build by hand. Current Apify pricing is only on the Store page.

The finished call sheet

Example rows only, to show the layout. Not real business data.

Score: 3 = email and phone, 2 = email only, 1 = phone only, 0 = nothing usable (kept on a separate "check by hand" list).

Why a raw scrape is not a call sheet yet

• Too many columns. When Apify exports a dataset to CSV it flattens every list, so you get emails/0/email, emails/1/email, phones/0/phone, phones/0/source, socials/whatsapp/0/url and dozens more. You need about eight of them.
• Repeats. If you merge two runs, or your list had both www. and bare domains, the same company appears twice. You want one row, the better one.
• The first phone is not always the best one. Numbers are listed in number order, not by quality. A number in a clickable tel: link is the one the business wants you to call.
• Blocked and failed sites have no contacts at all and only clutter the list.

Step 1: get one contact CSV

Free, a few businesses

Open each company's contact or about page and type what it publishes into a sheet with the same column names used below (domain, status, emails/0/email, phones/0/phone, startUrl). For ten or twenty businesses that is quicker than any tool. If you already have a CSV from another contact tool, rename its columns to these names and skip to Step 2.

No code, hundreds of websites (Apify)

1. Create a free Apify account (no card needed for the free plan; see apify.com/pricing for current free monthly usage).
2. Open Website Contact Scraper: Emails, Phones & WhatsApp and click Try for free.
3. Paste your websites. In JSON view:

{
  "urls": ["example-clinic.in", "example-studio.com", "example-traders.co.in"],
  "maxPagesPerSite": 4,
  "includePersonalEmails": false,
  "respectRobotsTxt": true,
  "respectTermsOfService": true,
  "checkEmailMx": true,
  "defaultCountry": "IN"
}

Set defaultCountry to the country most of the businesses are in, so local numbers without a country code are read correctly. Leave includePersonalEmails off: the call sheet is built on shared role inboxes such as info@, sales@ and appointments@. Already have a Google Maps dataset? Pass its dataset ID instead; see enrich a Google Maps Apify dataset with emails and phones, and the business name then comes through as sourceItems/0/title.

When the run finishes, open the dataset, choose Export → CSV and download it.

Step 2: import it as text into a tab called raw

• Google Sheets: new sheet, rename the first tab to raw, then File → Import → Upload, choose Replace current sheet and untick "Convert text to numbers, dates, and formulas". Without that, +914000000001 becomes the number 914000000001.
• Excel: Data → From Text/CSV, then Transform Data and set every phones/…/phone column to Text before you load it into a sheet named raw.
• Never edit the raw tab. Next time, import the new CSV over it and the call sheet refreshes.

Step 3: add a COL helper (one-time)

The column order of a CSV can change between runs, so pick columns by their header name, not their letter.

• Google Sheets: Data → Named functions → Add new function. Name: COL. Argument: h. Definition:

=IFERROR(XLOOKUP(h, raw!$1:$1, raw!$2:$5000), "")

• Excel 365: Formulas → Name Manager → New. Name: COL. Refers to:

=LAMBDA(h, IFERROR(XLOOKUP(h, raw!$1:$1, raw!$2:$5000), ""))

COL("domain") now returns the whole domain column. If a header is missing (for example no site in the run had a second email, so there is no emails/1/email), it returns blank instead of an error.

Step 4: build the rows tab

Add a tab called rows. Type these headers in row 1 and each formula in row 2 of its column. In Google Sheets the formulas already spill down; in Excel 365 they spill without ARRAYFORMULA (you can leave it out).

F: best phone (a tel: link first, otherwise the first number)

=ARRAYFORMULA(IF(COL("phones/0/source")="tel-link", COL("phones/0/phone"),
  IF(COL("phones/1/source")="tel-link", COL("phones/1/phone"), COL("phones/0/phone"))))

I: found on (the page where the email, or else the phone, was published)

=ARRAYFORMULA(IF(COL("emails/0/foundOn")<>"", COL("emails/0/foundOn"), COL("phones/0/foundOn")))

J: score (3 email + phone, 2 email only, 1 phone only, 0 none)

=ARRAYFORMULA(IF(A2:A="", "", IF((D2:D<>"")*(F2:F<>""), 3, IF(D2:D<>"", 2, IF(F2:F<>"", 1, 0)))))

The scraper lists same-domain emails first, so emails/0/email is normally the company's own inbox. Want WhatsApp to break ties? Change the 3 to 3+(G2:G<>"")*0.5.

Step 5: one row per business (the call_sheet tab)

Add a tab called sorted. In A2, sort the usable rows best score first, then by business name:

=SORT(FILTER(rows!A2:J, rows!A2:A<>"", rows!B2:B="ok", rows!J2:J>0), 10, FALSE, 1, TRUE)

In K2 of the same tab, number each repeat of a business (1 = its best row):

=ARRAYFORMULA(IF(A2:A="", "", COUNTIFS(A2:A, A2:A, ROW(A2:A), "<="&ROW(A2:A))))

Finally add the call_sheet tab, copy the headers from rows into row 1, and in A2:

=FILTER(sorted!A2:J, sorted!K2:K=1)

That is your call sheet: one row per company, most reachable first. For the "check by hand" list (blocked, failed or empty sites), add one more tab with =FILTER(rows!A2:J, rows!A2:A<>"", (rows!B2:B<>"ok")+(rows!J2:J=0)).

Excel 365: SORT and FILTER work the same. For the de-duplication, put =UNIQUE(sorted!A2:A1000) in call_sheet!A2 and fill each other column with =XLOOKUP($A2#, sorted!$A$2:$A$1000, sorted!D$2:D$1000) (change the last range per column). XLOOKUP returns the first match, which is the best-scored row.

Step 6: check before you call

• Email on another domain? If the best email's domain differs from the website, open the "found on" page. It can be the web designer's or a booking service's inbox, not the business.
• Toll-free and switchboard numbers are fine for a first call, but ask for the right person rather than guessing direct lines.
• mxValid FALSE on an email (column emails/0/mxValid in raw) means its domain cannot receive mail. Use the backup email or the phone.
• Blocked by robots or terms (status blocked-by-robots or blocked-by-terms) means the site asked bots not to read it. Visit it yourself or leave it out.

What not to do

• Do not guess personal addresses such as firstname@company. Use what the business publishes.
• Do not load the sheet into a bulk cold-email tool. Make genuine one-to-one calls and messages, and follow your local rules on calls and marketing (for example India's DND/TRAI rules, the EU's GDPR and ePrivacy rules, or the US TCPA).
• Do not message a WhatsApp number in bulk. Only use it where the business invites chat and your first message is a real, relevant enquiry.

Partner programme opening soon

Partner programme opening soon: join the free waitlist. Planned terms are 50% of each referred sale (75% once you have 5+ paying customers), single level, paid through Gumroad only from revenue received.

Related

• Enrich a Google Maps Apify dataset with emails and phones
• Find public role emails (sales@, press@, support@) on a company website
• Find a business website's WhatsApp and social links
• Copy-paste inputs for the Contact Details and Product scrapers
• Selling to online shops? Related only: Shopify & WooCommerce Product + Price Tracker (also an Unadhi Apify product) lists a shop's public products and prices, so you can open the call knowing their range.

← All guides

Unadhi Guides are free, independent explainers. We correct errors as soon as we find them: write to mininglife7@gmail.com.