Salesforce Prospecting Workflow with Google Maps Data (2026)

A modern salesforce prospecting workflow lives or dies on the quality of the data you push into it. Salesforce will happily accept any row you feed it, which is exactly the problem: dirty Google Maps exports turn into duplicate accounts, mismatched leads, and broken reporting within a quarter. This guide shows how to build a clean salesforce prospecting workflow using Google Maps data sourced through MapsLeads, with the right object model, field mapping, dedup rules, assignment logic, and connector setup so your reps actually trust the pipeline.

If you want the cross-CRM version of this same playbook, start with the CRM prospecting workflow complete guide 2026 and the Google Maps leads to CRM workflow overview before diving into Salesforce specifics.

Lead vs Account vs Contact: decide before you import

The biggest mistake teams make with Salesforce and local business data is skipping the object model decision. Google Maps results describe businesses, not people, so you need a deliberate convention.

Use the Lead object when your reps qualify before committing to a relationship. Leads are cheap, disposable, and meant to be converted or trashed. For SMB outbound off Google Maps lists, Lead is usually the right entry point: you import a few hundred plumbers or dental clinics, your SDRs work them, and only the qualified ones get converted into Account plus Contact plus Opportunity.

Use the Account object directly when you are building a target account list (ABM-style) where every business on the list deserves a long-term record regardless of immediate qualification. Mid-market and enterprise teams often skip Lead entirely.

Contact is for the human. Google Maps does not give you named contacts, so Contact records appear later, after enrichment or first reply. Do not try to force a business name into the Contact object.

Pick one convention per campaign and document it. Mixing Leads and Accounts inside the same Google Maps import is what creates the duplicate chaos most Salesforce orgs eventually hire a consultant to clean up.

Build the list in MapsLeads

The list itself is built in MapsLeads before Salesforce ever sees a row. Run a Search by category and city, scope it tight enough that the result set is workable (one industry, one metro), and review the preview. Run Contact Pro on the rows you want to keep so you get the website, phone, and email signals. Add Reputation when the campaign cares about review count, average rating, and the keywords customers actually use in reviews — this is the data that lets you write a relevant first-touch message instead of another generic outbound blast. Add Photos only when visual context matters (interior design, hospitality, real estate).

Save the result into a group named after the campaign. The group name should match the Salesforce campaign name exactly, because that string becomes the Lead Source detail or a custom field downstream. Export to CSV when you are ready.

CSV vs Data Loader vs Salesforce Inspector

Salesforce gives you three realistic ways to import a MapsLeads CSV.

The Data Import Wizard inside Setup handles up to 50,000 records and is the right tool for occasional, smaller list loads. It supports matching rules natively and will respect duplicate rules, which makes it safer than Data Loader for one-off campaigns.

Data Loader is the classic desktop tool. It scales past 50,000 records, supports upsert against an external ID, and is scriptable for recurring loads. If your salesforce prospecting workflow runs every week or you sync more than 50k rows at a time, Data Loader with a saved field mapping is the right answer.

Salesforce Inspector (the browser extension) is the power user option. It lets you paste a CSV directly into a query result and update or insert in seconds. Use it for surgical work — patching review counts on existing accounts, fixing a city field on 200 records, backfilling a custom property. Do not use it for first-time bulk imports; the audit trail is weaker.

The pattern most healthy teams settle on: Data Import Wizard for the first import of a new campaign (so duplicate rules fire), Data Loader for recurring loads against an external ID, Inspector for one-off corrections.

Field mapping that survives a year

Map MapsLeads CSV columns to Salesforce fields explicitly. Do not let Salesforce auto-map; the column names will not match cleanly and you will lose data silently.

Account name maps to Company on the Lead or Name on the Account. Website maps to Website on both objects and should be normalized (lowercase, no trailing slash, no www) before import — this becomes your dedup anchor. Phone maps to Phone, but normalize to E.164 first if you have international data. Address fields map one-to-one (Street, City, State, Postal Code, Country).

The interesting fields are the Google Maps-specific ones, which need custom fields on Lead and Account. Create a custom number field for Rating (one decimal place), a custom number field for Review Count (integer), a custom long text field for Review Keywords (the top words pulled from the Reputation feature), a custom URL field for Google Maps URL, and a custom text field for Place ID. Place ID is the cleanest external identifier you will ever get for a local business — it never changes when the business updates its name or phone, and it is your best upsert key in Data Loader.

Add a custom picklist for Maps Category (the primary Google category) and a multi-picklist for Sub Categories. These two fields make territory and segment reporting trivial later.

Dedup: matching rules and duplicate rules

Salesforce dedup is a two-layer system. Matching rules define what counts as a duplicate (the comparison logic). Duplicate rules define what happens when a match is found (block, allow with alert, or report).

For a salesforce prospecting workflow built on Google Maps data, configure three matching rules. The first matches Lead to Lead on normalized website domain plus city — exact match required, fuzzy disabled. The second matches Lead to Account on the same keys, so a new lead on an existing customer account gets caught. The third matches Account to Account on Place ID where present, falling back to website domain plus postal code.

Pair each matching rule with a duplicate rule set to "Block" on insert for high-confidence matches (Place ID), and "Allow with alert" for medium-confidence (website plus city). Allowing with alert preserves rep judgment while still flagging the conflict.

Run the standard duplicate jobs weekly. The Salesforce Duplicate Jobs feature surfaces matches that slipped past insert-time rules — usually because the website was missing on one record and present on the other.

Assignment rules

Assignment rules turn a generic import into a routed pipeline. Build the rule set around the fields you imported: territory by State or Postal Code prefix, vertical by Maps Category, account size by Review Count buckets (under 50 = SMB, 50 to 250 = mid, 250+ = key account). The Review Count proxy is rough but it correlates well with foot traffic and revenue for most consumer-facing categories, which is good enough for round-robin routing.

Assign a fallback queue for any record that does not match. Never let a record sit unassigned — orphaned leads are how good lists die.

Connectors: Outreach, Salesloft, Apollo

Salesforce alone does not sequence outbound. Pick one engagement platform and wire it tightly.

Outreach syncs bidirectionally with Salesforce out of the box. Map MapsLeads-derived custom fields (Rating, Review Count, Review Keywords) into Outreach as snippet variables so reps can pull them into email templates. A first-touch email referencing a specific review keyword outperforms a generic intro by a factor that justifies the Reputation credit on its own.

Salesloft works the same way and has a slightly cleaner cadence editor. Pick whichever your team already prefers; the data flow is identical.

Apollo serves a different role — it is your enrichment and contact discovery layer on top of MapsLeads' business data. Use Apollo to find named contacts at the businesses you imported from Google Maps, then push those contacts back into Salesforce as Contact records linked to the Account.

The integration sequence matters: MapsLeads to Salesforce first (the business record), then Apollo for contact enrichment (the human record), then Outreach or Salesloft for sequencing.

Reporting templates

Build three reports day one and never delete them. The first is Lead Source Performance by MapsLeads group: rows are campaign names, columns are stage conversion rates. This tells you which categories and cities convert. The second is Review Count vs Win Rate: a matrix report showing close rate by Review Count bucket. The third is Time to First Touch by assignment queue, which surfaces routing problems early.

If you want the parallel HubSpot setup for comparison, see HubSpot prospecting workflow with Google Maps.

MapsLeads to Salesforce in five minutes

This is the minimum viable flow. Repeat it as often as you run new campaigns.

Step one: open MapsLeads and run Search for your target category and city. Filter to the result count you want.

Step two: select the rows you want, run Contact Pro on the selection, then run Reputation on the same selection. Wait for both passes to complete. Skip Photos unless you need them.

Step three: export the group to CSV. Open the file in Excel or Google Sheets and confirm Place ID, Website, Phone, Rating, Review Count, and Review Keywords columns are populated. Normalize Website to lowercase without protocol or trailing slash.

Step four: open Salesforce Data Loader. Choose Upsert. Select the Lead object (or Account, per your convention). Set the external ID to your custom Place ID field. Map columns explicitly — do not auto-map. Run the upsert against a sandbox first if this is a new mapping.

Step five: watch the success file. In Salesforce, open the matching duplicate rule report to confirm no high-confidence duplicates were created. Confirm assignment rules fired by spot-checking three records. Trigger the Outreach or Salesloft sync and add the new records to a cadence.

That is the entire loop. Once the field mapping is saved in Data Loader, every subsequent run takes five minutes start to finish.

A typical campaign runs 1 credit per row for the Search and Base data, plus 1 credit per row for Contact Pro, plus 1 credit per row for Reputation, plus 2 credits per row for Photos when you need them. Budget against that math before you build the list, and check your wallet on the Pricing page if you need to top up.

Common mistakes

Importing without a Place ID column. You lose the cleanest upsert key on the planet and will eventually get duplicates.

Letting Salesforce auto-map columns. The auto-mapper guesses, the guesses are silently wrong, and you find out three months later when a report breaks.

Skipping the website normalization step. Your dedup rules cannot match "https://www.example.com/" to "example.com" unless you make them identical at import time.

Mixing Lead and Account imports in the same campaign. Pick one and stick to it.

Sequencing before enrichment. Pushing raw business records into Outreach without named contacts produces generic outreach that hurts your sender reputation.

Checklist

Object model decided and documented. Custom fields created on Lead and Account (Place ID, Rating, Review Count, Review Keywords, Maps URL, Maps Category). Matching rules and duplicate rules configured and tested in sandbox. Assignment rules cover every territory plus a fallback queue. Data Loader field mapping saved. Outreach or Salesloft sync verified on three test records. Three reports built and pinned.

FAQ

How do I import Google Maps leads to Salesforce? Build the list in MapsLeads with Search plus Contact Pro plus Reputation, export to CSV, normalize the website column, then upsert through Data Loader against a custom Place ID external identifier with explicit field mapping.

Should I use Lead or Account for SMB Google Maps prospects? Lead. SMB outbound has a high disqualification rate and Leads are designed to be disposable. Convert to Account plus Contact plus Opportunity only when a rep confirms the business is worth a long-term record.

What is the best dedup tool for Salesforce when importing local business data? Native matching rules plus duplicate rules are sufficient if you import a Place ID and normalize the website column. Third-party tools become useful only above 100,000 records or when you need fuzzy address matching.

What sequencing options work with this Salesforce workflow? Outreach and Salesloft both sync bidirectionally and accept custom field variables, so review keywords and ratings flow into email templates. Pick whichever your team already uses.

Can I run this workflow weekly? Yes. Save the Data Loader field mapping, save the MapsLeads search filters, and the recurring run takes five minutes.

How do I avoid duplicate accounts when reps import overlapping cities? Configure the Account-to-Account matching rule on Place ID first, website domain plus postal code second, set the duplicate rule to Block on insert for Place ID matches.

Get started

A clean salesforce prospecting workflow is a one-time setup that pays back every campaign. Build the list in MapsLeads, push it through Data Loader with proper mapping, and let dedup plus assignment rules do the routing. Get started and run your first five-minute import today.