Local Business TAM from Directories

The workflow

Play 32

Local Business TAM from Directories

1Size the market before paying for it

Directory totalsFree and exact

One search per region and category. Read the result count, scrape nothing.

Fetch

Browserbase

Maps pilot30 per term

Shows coverage, which categories are noise, and whether a term runs dry early.

Run

ApifyGoogle Maps

Hard dollar cap agreed before the full run. The cap is per search term.

2Scrape two kinds of source

Google MapsWidest coverage

PhoneWebsiteAddressRating

Search the whole region, exclude cities afterwards. Maps lists locations, not companies.

With

ApifyGoogle Maps

Industry directoriesCategory detail

CategoryAddressPaid listings

Confirm page 1 and page 2 differ before crawling. Resumable on region, category and offset.

With

Browserbase

3De-duplicate across sources

Match onThen

Same website domain→Merged, richest copy kept

Same name and street address→Merged across sources

Directory or link-in-bio domain→Domain blanked, never matched

One business, several cities→One row, every city kept

4Enrich, then check it is still true

Website and owner

Find a site for address-only rows, then read the team page for the owner.

Match

Clay

Read

Claude

Email waterfall

Only the misses move to the next provider. Nothing goes out unverified.

Find

ProspeoFindymail

Verify

BounceBan

Then

Freshness checkListings go stale

Before anyone calls: the listing is not marked closed, the phone still connects, and the email is re-verified on the day it is sent.

Status

Google Maps

Verify

ZeroBounce

5Tier, then load

Tier 1Several sites, strong reviews

Call, then emailNow

The phone came with the listing. A rep calls first, then the sequence starts.

Call

Nooks

Log

HubSpot

Send

Smartlead

Tier 2 and 3Single site or thin data

Email onlyAuto

The copy names the category and the city the business was found under.

Send

Smartlead

A deduped local TAM, tiered and reachable

The stackBrowserbaseApifyGoogle MapsClayClaudeProspeoFindymailBounceBanZeroBounceNooksHubSpotSmartlead

Oriel · GTM plays library

Open full size

The diagram as text

1. 1

   Size the market before paying for it

   Directory totals (free and exact): One search per region and category. Read the result count, scrape nothing. Fetch: Browserbase.

   Maps pilot (30 per term): Shows coverage, which categories are noise, and whether a term runs dry early. Run: Apify and Google Maps.

   Hard dollar cap agreed before the full run. The cap is per search term.

2. 2

   Scrape two kinds of source

   Google Maps (widest coverage): Phone, Website, Address and Rating. Search the whole region, exclude cities afterwards. Maps lists locations, not companies. With: Apify and Google Maps.

   Industry directories (category detail): Category, Address and Paid listings. Confirm page 1 and page 2 differ before crawling. Resumable on region, category and offset. With: Browserbase.

3. 3

   De-duplicate across sources

   Match on → Then

   • Same website domain → Merged, richest copy kept
   • Same name and street address → Merged across sources
   • Directory or link-in-bio domain → Domain blanked, never matched
   • One business, several cities → One row, every city kept
4. 4

   Enrich, then check it is still true

   Website and owner: Find a site for address-only rows, then read the team page for the owner. Match: Clay. Read: Claude.

   Email waterfall: Only the misses move to the next provider. Nothing goes out unverified. Find: Prospeo and Findymail. Verify: BounceBan.

   Labeled connection: Then.

   Freshness check (listings go stale): Before anyone calls: the listing is not marked closed, the phone still connects, and the email is re-verified on the day it is sent. Status: Google Maps. Verify: ZeroBounce.

5. 5

   Tier, then load

   Tier 1: Several sites, strong reviews

   Call, then email (now): The phone came with the listing. A rep calls first, then the sequence starts. Call: Nooks. Log: HubSpot. Send: Smartlead. Tool: Nooks.

   Tier 2 and 3: Single site or thin data

   Email only (auto): The copy names the category and the city the business was found under. Send: Smartlead. Tool: Smartlead.

   Outcome: A deduped local TAM, tiered and reachable. Tools: HubSpot.

Feedback loop: The raw scrape is kept, so new filters and new tiers re-run on it at no cost.

The stack: Browserbase, Apify, Google Maps, Clay, Claude, Prospeo, Findymail, BounceBan, ZeroBounce, Nooks, HubSpot and Smartlead.

Open the diagram full size (1180 px wide)

The play in six steps

1. 01

   Step 1: Size the market before paying for it

   Local businesses are thin or missing in B2B databases, but most directories will tell you how many listings exist before you scrape a single page.

   • Fetch one result page per region and category with Browserbase and read the total. It costs nothing and it is exact.
   • Run a Google Maps pilot on Apify at 30 places per search term. If every term fills all 30, there is more supply. If terms come back short, the full run will stop early.
   • The cap is per term, not a total: multiply it by the number of terms to get the real size of the run.
   • Agree a hard dollar budget before the full run starts.
2. 02

   Step 2: Scrape Google Maps

   Maps has the widest coverage of local businesses and carries a phone, a website and an address on most rows.

   • Use the Apify Google Maps actor with two or three phrasings of each search term.
   • Search a whole state or region, then drop cities afterwards instead of narrowing the query.
   • Pass a max spend so Apify stops charging at the budget.
   • Filter on category. Broad terms pull in neighbouring trades.
   • Maps lists locations, not companies. A business with five offices shows up five times.
3. 03

   Step 3: Add the industry directories

   Trade and professional directories hold businesses Maps misses, plus category detail Maps never had.

   • Fetch result pages with Browserbase. Server-rendered listings need no browser session and no LLM.
   • Find the parameter that really changes the page: fetch page 1 and page 2 and confirm they differ before crawling.
   • Category labels are exact strings. Test each one on a single fetch first.
   • Make the crawl resumable on region, category and offset, with retries.
   • Expect contact data only on paid listings. Free listings usually carry a name and an address.
4. 04

   Step 4: De-duplicate across sources and cities

   The same business turns up in several sources, several categories and several neighbouring cities, and every duplicate is a wasted call.

   • Match on website domain first, then on normalised name plus street address.
   • Blank directory and link-in-bio domains before matching, or unrelated businesses collapse into one row.
   • Keep the richest copy of each record, plus the full set of categories and cities it appeared under.
   • Count rows with a CSV parser, not a line count. Descriptions carry line breaks.
5. 05

   Step 5: Enrich the website, the owner and the email

   Directories rarely publish emails, so the contact layer is built downstream, and a listing that closed last month is worse than no listing.

   • Resolve a website in Clay for rows that arrived with only a name and an address.
   • Have Claude read the about or team page and name the owner or principal.
   • Run the email waterfall in Clay: Prospeo first, Findymail on the misses, BounceBan on everything found.
   • Run a freshness check before anything is sent:
     • Drop listings Google Maps marks as permanently or temporarily closed
     • Drop phone numbers that no longer connect
     • Re-verify the email with ZeroBounce on send day, not the day it was found
6. 06

   Step 6: Tier and load

   A local TAM is large and uneven, so the tier decides where a rep's time goes.

   • Score on what the sources actually hold: number of locations, review count and rating, website quality, category fit.
   • Tier 1 gets a call through Nooks and then an email. The number is a published business line, so a call is fair.
   • Tier 2 and 3 get email only through Smartlead, naming the category and city they were found under.
   • Load companies and contacts into HubSpot before the first send.
   • Keep the raw scrape so filters can be re-run later at no cost.

Tools used

The twelve tools this play names. Each links to the other plays that use it.

BrowserbaseApifyGoogle MapsClayClaudeProspeoFindymailBounceBanZeroBounceNooksHubSpotSmartlead

Related plays

Play 30

Public Records Signals

1Pull public records on a schedule

Government APIsFree, read-only

Disaster declarationsWeather eventsSEC filings

Daily for declarations, weekly for filing phrase searches.

Pull

Plain GET on a schedule

Portals with no APIReal browser

Court docketsPermits

Throttled, sessions recycled. Portals behind a CAPTCHA are out.

Drive

BrowserbaseFirecrawl

One row per recordDate, source URL, verbatim excerpt

2Match each record to an account

Entity resolutionThe hard part

A record names a legal entity or a place, not your CRM account.

Name

Strip Inc, Corp, LLC, The

Place

County or address to known sites

Against

HubSpotSalesforce

MatchConfidence

Name and location agree→High: routes on its own

Name only→Medium: review queue

Fuzzy or alias→Low: review queue

3Qualify in two gates

Gate1

Not suppressed

Customers, open deals and competitors come out first.

× Customer or open deal

Gate2

Actually relevant

Claude reads the excerpt and gives a one-line reason.

× Not relevant

4Tier what the record allows

Act aloneThe event is the reason

Owner alertThis week

Record link, excerpt and matched sites, straight to the owner.

Alert

Slack

Log

HubSpot

One more factPerson or site first

Find the person

The right contact, then a checked email.

Find

AI Ark

Email

ProspeoFindymail

Verify

BounceBan

Rank onlyContext, no message

Score only

Adds weight to the account score. Never opens a message.

On

HubSpot

5Reach out with the record as evidence

State the public fact plainly

Something the buyer can check for themselves, with the source link at hand.

Email

Smartlead

Call

Nooks

A dated reason to call this week

The stackSEC EDGARBrowserbaseFirecrawlClayHubSpotSalesforceClaudeSlackAI ArkProspeoFindymailBounceBanSmartleadNooks

Oriel · GTM plays library

Play 30Outbound · ABM

Public Records Signals

5 steps · Email, Call, CRM

+8 more

Play 08

Competitor Post Engagers

1Watch the competitor voices

Company pagesCompetitors

LaunchesWebinarsCustomer posts

Every competitor page that posts.

With

LinkedIn

Founders and execsTheir people

FoundersExecsCreators

The people who post for them, reviewed monthly.

With

LinkedIn

New posts from the last 24 hoursDeduped by post id

2Capture who engages

Post engagersScraped once per post

ReactionsCommentsPost text

A comment counts for more than a reaction. The post travels with every engager.

Scrape

Apify

Pool

Clay

3Enrich, then qualify in three gates

Person and company

RoleSeniorityHeadcountVerified email

Find

Explorium

Email

ProspeoFindymail

Verify

BounceBan

Gate1

Company fit

ICP industry, headcount and region. Competitor staff and partners come out.

× Not ICP

Gate2

Persona

The titles that buy, not everyone who reacts.

× Wrong persona

Gate3

Not already yours

Customers and open deals are removed.

× Customer or open deal

4Pitch the switch on the post topic

They engaged withSwitch angle

A launch post→How it compares to yours

A how-to post→The same job, done simpler

A pricing post→The cost of staying

Name the topic, never the act of watching.

5Multi-thread the account

“An opener written from the post topic”

The engagerDay 0

A LinkedIn message, on the platform where they saw the post.

Send

HeyReach

Their colleaguesDay 3

The buying committee hears the same angle by email.

Send

Smartlead

Verify

BounceBanon send day

3-day delay

Switch conversation booked

The stackLinkedInApifyClayExploriumHubSpotProspeoFindymailBounceBanClaudeHeyReachSmartleadSlack

Oriel · GTM plays library

Play 08Outbound

Competitor Post Engagers

6 steps · LinkedIn, Email, CRM

+6 more

Play 12

Contact Data Refresh Before Every Send

1Pull everyone about to be contacted

Contacts queued for next week's sendsLists and sequences

What can have gone stale

JobCompanyWork emailMobileLinkedIn URLTitle

2Check each field against a live source

Still at the company?Job change

Compare the current LinkedIn role with the CRM record.

Check

LinkedInApify

Email still delivers?Verify

Re-verify on send week, not on the day it was found.

Verify

BounceBanZeroBounce

Number still live?Line type

Look up carrier and line type, drop disconnected numbers.

Check

Twilio

3Decide what each record needs

FindingAction

Left the company→Find the replacement, follow the person

Email risky or invalid→Find a new email, else LinkedIn only

Title changed→Re-score, maybe a new angle

Number dead→Remove it from call lists

All still true→Send as planned

4Fill the gaps

Replacement at the account

The new person in the same seat inherits the account context.

Find

ApolloClay

Email

FindymailProspeo

The mover at the new company

A past champion who moved is often the warmest lead you have.

Track

LinkedIn

Mobile

FullEnrich

Gate1

Only verified records go out

Anything still unverified is held for the next cycle.

× Unverifiedheld, rechecked next week

Every send goes to a real person in the right seat

The stackHubSpotLinkedInApifyBounceBanZeroBounceTwilioClayApolloFindymailProspeoFullEnrichSmartlead

Oriel · GTM plays library

Play 12Outbound · ABM

Contact Data Refresh Before Every Send

4 steps · Email, LinkedIn, Call, CRM

+6 more

All GTM plays