Skip to content
Lead discovery

Find the companies, prove the addresses, then decide.

Scout searches Google Places for B2B companies matching a niche and a location, finds email addresses and confirms each one with an SMTP handshake, then scores every lead from 0 to 100. Discovery mode stops there. The pool waits for a person.

No payment method. No recurring charge. Your own SMTP, your own data.

Discovery source
Google Places Text Search (New)
Address check
SMTP handshake, EHLO and MAIL FROM
Scoring
0-100 before the pipeline
CSV import
Up to 20,000 rows
Scout - discovery results5 of 5 shown
BusinessScoreStatus
Hydraulics supplierManchester84New
Industrial coatingsLeeds78New
CNC machining shopSheffield71Queued
Packaging converterBristol64Queued
Cold storage operatorHull52Held
Illustrative preview - lead pool after a discovery run
Where a list really comes from

Prospecting is not difficult. It is just never finished.

The list gets built. It simply eats the morning that was meant for selling, and the half-run search is still sitting in a browser tab on Thursday.

  1. A search that has to be run again next quarter

    The niche has not changed and the city has not moved. The result was pasted into a spreadsheet that started ageing the day it was made, and nobody volunteers to do it a second time.

  2. Addresses that were guessed rather than checked

    First name, dot, surname, at the domain is a hypothesis. Sending to a hypothesis is how a bounce rate climbs before the campaign has said anything at all.

  3. An export that arrives without a shape

    Every tool names its columns differently. Mapping them by hand is the quiet tax paid on every list somebody hands over.

  4. One company, two lists, two messages

    The same business turns up from a search and again from a spreadsheet. The second message lands on a person who already read the first one.

How a lead reaches the pool

Four steps between a niche and a lead you can actually write to.

Scout does the searching, the verifying and the scoring. It does not do the deciding.

  1. 01

    Describe the niche, not the query

    A project carries the niche and the location. Scout turns that into a Google Places Text Search (New) call, so nobody has to think in search operators to get a list of B2B companies back.

  2. 02

    Search once, buy once

    Places results are cached in a places_cache table with a seven-day TTL. The same niche and location inside that window is served from the cache rather than bought a second time.

  3. 03

    Find the address, then prove it

    Scout looks for email addresses and confirms each one with an SMTP handshake, EHLO followed by MAIL FROM, so the mailbox is known to exist before anything is written to it.

  4. 04

    Score, then stop

    Every lead is scored from 0 to 100 before it enters the pipeline. Discovery mode is human-gated: Scout saves the pool and halts, and research does not start on its own.

The lead pool

A pool you can sort, filter and leave alone.

Discovery writes leads into a pool rather than straight into a campaign. Sorting and pagination happen on the server, so a long list behaves like a short one, and the AI filter narrows the pool in plain language.

  • Every lead carries a score from 0 to 100
  • Sorting and pagination are handled server-side
  • AI-assisted filtering runs behind /api/leads/ai-filter
  • Nothing leaves the pool until a person moves it
Scout - discovery results5 of 5 shown
BusinessScoreStatus
Hydraulics supplierManchester84New
Industrial coatingsLeeds78New
CNC machining shopSheffield71Queued
Packaging converterBristol64Queued
Cold storage operatorHull52Held
Illustrative preview - lead pool after a discovery run
Scope

One project per client, and no pool that bleeds into the next.

Leads belong to a project, and a project represents one client. The pool built for one account is not reachable from another, because every query carries the organisation as well as the project.

  • Leads are scoped per project and per organisation
  • A project maps to a single client
  • Leads and Leads / Import are per-project screens
  • The Pipeline view reads the same scoped data
Projects - workspace switcher4 projects
Client project A4 campaignsOwner
Client project B2 campaignsAdmin
Client project C6 campaignsMember
Internal pipeline1 campaignsViewer

Leads, campaigns, mailboxes and analytics are scoped per project. Every query carries the organisation and the project.

Illustrative preview - one organisation, separate client projects
What changes

Automated lead generation software should end the search, not just speed it up.

A search that runs without you

Saved searches run on a daily, weekly or monthly schedule. The list keeps arriving long after the person who set it up moved on to something else.

Addresses proved, not patterned

An SMTP handshake asks the receiving server whether the mailbox exists. The answer is a fact, not a guess drawn from a naming convention.

The same search paid for once

Places results are cached for seven days, so repeating a niche and location inside that window reads from places_cache instead of buying the result again.

A spreadsheet that maps itself

CSV import takes up to 20,000 rows, matches the columns automatically and deduplicates on the way in.

A score before a decision

Each lead arrives with a 0-100 score, so the question becomes which end of the pool to work rather than where on earth to start reading.

Discovery that waits

Scout in discovery mode saves leads and stops. Research, drafting and sending all sit behind a person who has to say yes.

The situations you are about to ask about

What happens in the awkward cases

Sourcing is mostly edge cases. These are the ones that come up first and how the platform behaves in each.

  • A repeated search

    You run a niche you already ran

    Inside the seven-day TTL the answer comes from places_cache, so a second run of the same niche and location does not buy the same Places result twice.

  • Overlapping runs

    Two saved-search runs land together

    The scheduler claims work with an atomic UPDATE that only matches rows whose lastRunAt is older than the threshold, so concurrent runs cannot queue one search twice.

  • Somebody else's export

    A CSV shaped by another tool

    Columns are mapped automatically on import and duplicates are dropped as rows are brought in, up to a ceiling of 20,000 rows in a single file.

  • Nothing to write to

    A company with no verified address

    It stays in the pool as a lead. It is simply not one anything can be sent to, and the SMTP check is what makes that visible before a campaign rather than after it.

  • Several clients

    One agency, separate lead pools

    Leads are scoped per project and per organisation. A pool built for one client cannot be read from another client's project.

  • A very long list

    A pool bigger than the appetite for it

    Sorting and pagination run on the server, and the AI filter narrows the pool in plain language, so size is a detail rather than an obstacle.

The specifics

What lead discovery actually is

The short version, without adjectives, for the person who wants to know exactly where the data comes from.

Discovery source
Google Places Text Search (New)
Cache table
places_cache
Cache TTL
7 days
Address check
SMTP handshake (EHLO, MAIL FROM)
Lead score
0-100
Discovery mode
Human-gated, saves to pool
CSV import
Up to 20,000 rows
Import handling
Auto column mapping, dedup
Saved searches
Daily, weekly or monthly
Scheduler claim
Atomic UPDATE on lastRunAt
List handling
Server-side sort and pagination
Scope
Per project, per organisation

What building a list stops costing you every month.

How each task is handled today compared with MyLeadTeam
TodayIn MyLeadTeam
A seat in a contact database such as Apollo, paid for in months when nobody builds a list at all.Discovery runs from a niche and a location, and the Places result is cached for seven days so a repeat search does not buy the same data twice.
An address assembled from a naming pattern and sent to hopefully.Each address is confirmed with an SMTP handshake before anything is written to it.
A spreadsheet passed between people, mapped by hand and deduplicated by eye.CSV import of up to 20,000 rows, with automatic column mapping and deduplication as the rows land.
A recurring calendar reminder to re-run last quarter's search.A saved search on a daily, weekly or monthly schedule, claimed atomically so two runs cannot duplicate the queue.
One shared list that every client's campaign is drawn from.Leads scoped per project and per organisation, with one project standing for one client.
FAQ

Questions people ask about this

MyLeadTeam's Scout agent finds B2B companies through Google Places Text Search (New), using the niche and the location attached to a project. Results are cached in a places_cache table with a seven-day TTL, so running the same niche and location again inside that window is served from the cache rather than bought a second time. Leads can also arrive by CSV import, which accepts up to 20,000 rows per file with automatic column mapping and deduplication. Both routes write into the same per-project lead pool.

Get started

Start with a niche and a city, not with an empty spreadsheet.

Scout searches, verifies and scores. The pool sits there until you decide what happens to it next.

No payment method or recurring charge during early access.