Find local points of sale

Finding local shops to stock a collection usually means someone opening a map, searching one obvious phrase, and stopping at the first page of results, then emailing every one of them the same note whether or not the shop is a fit. Half the good candidates never turn up in that first search, and half the emails go to places that were never going to say yes.

Finding retailers for your product is two jobs: seeing every shop in a city that could carry it, and deciding which of them are worth a message. An AI agent with search and email can do both, city by city, and keep the results in one place so a later pass never starts from zero.

How the agent finds retailers in each city

The run starts once a list of target cities exists. Each city is searched three ways, since no single query returns everyone.

• Local listings. A places search through Serper for your category and positioning ("independent [category] shop [city]"), then two or three variants by neighborhood or style keyword.
• Web guides. Searches for "best [category] shops [city]" roundups, which often name five to fifteen shops on a single page that a places search alone would miss.
• The shops those guides name. Each roundup is scraped for its full list of names, addresses and descriptions instead of being read as prose. Most of the volume comes from here.

A places search alone misses the shops that only appear in a roundup; a web search alone misses the ones with no press and a strong map presence. Running both, then scraping what the roundups list, is what closes that gap.

Every candidate lands in one running list: name, city, neighborhood, address, website and a free-text note. It is one table filtered by city, never a new list per city, so nothing has to be found twice.

Qualifying retailers before any email

Positioning fit is checked from each shop's own site or social presence: does what they carry sit next to what you would want to sit next to? The signals weighted on top of that are an active, recently updated social account, a strong aggregate review score, and a price point compatible with your wholesale terms.

Shops are ranked into a few tiers instead of a single score, which keeps the reasoning visible and tells the follow-up where to spend its persistence. Anything that fails on positioning is dropped before it reaches a person's review of the list.

For the highest tier only, the agent reads the shop's public reviews and pulls one real detail its customers mention: style, selection, service. That step doesn't scale to the whole list, and doesn't need to. A generic compliment reads as generic; a specific one reads as homework done.

Retailer outreach on a fixed follow-up cadence

Each first message, sent from Gmail, carries that detail where there is one, two lines on what you make and where it sits, and a concrete next step: a lookbook or catalog with terms, or a short call.

The follow-up is a schedule decided once and applied the same way to every shop: a short nudge at day 7, a different channel at day 14, and the shop is archived at day 21 if nothing has come back. Every contact and its outcome is logged against the shop's record, so a later pass never sends the same opener twice, and a promising shop doesn't quietly fall off the list because nobody remembered to check back in.

Shops that start stocking you stay on the same list, past the yes. Each season they get the new collection and a real check-in on a fixed cadence, and the list tracks which accounts are active and which have gone quiet, the same way it tracked prospects.

Each run reports how many shops were found per city, how many qualified, how many were contacted, the response rate, and how many became active accounts.

Beyond retail

The shape doesn't depend on the category. Swap the search terms and the positioning criteria, and the same three-angle search and qualify-before-contact order applies to whatever you are sourcing.