---
title: A Method for Building a Sales Agent Network in Japan — How to Find, Screen and Approach Agents for Products Sold to Shops and Small Businesses
description: Agent sourcing in Japan stalls because of how the information is structured, not effort. A list stuck at 674 companies gained 88 in two days with a channel-first five-step method. The method, anonymized results, and the DIY failure patterns.
---
[Skip to content](https://nobusato.jp/blog/salespartner-en#main-content)
[](https://nobusato.jp)
- [Contact](mailto:info@nobusato.jp)
- [JobList](https://www.careers-page.com/mycareer-tokyo)
# A Method for Building a Sales Agent Network in Japan — How to Find, Screen and Approach Agents for Products Sold to Shops and Small Businesses
[Nobu Sato October 20, 2026](https://nobusato.jp/blog/author/nobu-sato)
- [代理店](https://nobusato.jp/blog/tag/代理店)
- [Japan Market](https://nobusato.jp/blog/tag/japan-market)
- [Sales Partner](https://nobusato.jp/blog/tag/sales-partner)
- 26 mins
If you sell to shops and small businesses in Japan, your own sales team can't cover the country. You need sales agents — *dairiten*. And once you start looking for them, progress is surprisingly slow. This post lays out why, the five-step method I actually use, anonymized results, and what happens when companies try it in-house.
## Why finding sales agents is such a chore
Agent sourcing stalls not because the person in charge isn't trying, but because of how the information is structured. The companies that can carry your product to shops and small businesses are scattered across industries, and no single directory lists them.
- **There is no directory.** Agents sit in telecom, office equipment, payments, restaurant booking and beauty distribution — separate industries — and many never call themselves "agents."
- **The obvious sources saturate fast.** Comparison and referral sites top out at a few hundred companies. Crossing several sites yields mostly duplicates and almost nothing new.
- **The good agents already sell someone else's product.** A company with a field sales network is almost always already an agent for a competitor. Unless you design the pitch as an "add-on product," you get turned down again and again.
- **Contact details aren't public.** Form only, an address found only on a job posting, a switchboard number and nothing else — this is 10–20% of companies. Mass-mailing guessed addresses doesn't land, and costs credibility.
- **One group recruits under several brands.** Companies that look separate but share a switchboard number are common. Approach the same group three times without noticing and you've already made a bad impression.
- **Some companies say "no sales calls."** Phone them and that company is closed to you for good.
- **A raw list is unusable.** Hand a few hundred unranked companies to a sales team and they burn out before reaching the top candidates.
Start without knowing this structure and you end up where everyone does: the list grows, the contracts don't.
## The method at a glance — five steps
One principle: **start from the channel, not the list.** Decide first which companies visit shops every week, find where those companies gather, then verify and rank before anyone picks up the phone.
| Step | What you do | Output | Where DIY gets stuck |
| --- | --- | --- | --- |
| 1 Define the ideal agent | Work backwards from who visits the target shops weekly | Agent profile (touchpoint, product fit, region) | Searching only "sales outsourcing," missing industry distributors |
| 2 Design the pool by channel | Mine six channel types beyond comparison sites | Candidate list by channel | Collecting a few hundred from comparison sites and stopping |
| 3 Collect, dedupe, rank | Remove duplicates, bundle group companies, S/A/B rating | Ranked master list | The same company enters three times under different names; no priority |
| 4 Verify contacts | Only contacts stated on official sites; zero guessing | Verified contacts + "to confirm" flags | Mass-mailing guessed addresses and losing credibility |
| 5 Design the approach | Per company: relation to existing products, non-compete terms, contact method | Per-company approach and first message | Wasting time on exclusive competitor agents; phoning "no sales calls" companies |
Steps 1 and 2 decide 80% of the result. Steps 3–5 keep quality from slipping; skip them and the sales team's hours evaporate.
### Step 1 — Define the ideal agent
Don't "look for agents." First write down who shows up at your target shops this week. There is no industry called "agents" — companies with shop touchpoints simply go by different names in each industry.
For products sold to restaurants, retail and salons, the candidates fall into roughly five types:
- Multi-product field sales companies selling telecom lines, copiers and electricity to shops door to door
- Official agents who pitch restaurant and booking-site listings to shops
- Beauty distributors delivering products and equipment to salons on weekly routes
- Rounders and field marketing companies that patrol retail shelves on a schedule
- Dealers of POS registers, ticket machines and commercial equipment
The profile needs only three items:
1. Touchpoint — in descending value: weekly route visits \> new-customer field sales \> telesales or referrals only
2. Product fit — can your product ride along as "one more item," or is the company an exclusive competitor agent?
3. Region — include the cities where you have direct sales offices, or only everywhere else?
Start by searching "sales outsourcing company" and you'll find only the first of the five types. This profile decides which channels Step 2 digs into.
### Step 2 — Design the pool by channel
Comparison sites are the right entrance but not the exit. Sweep comparison sites and equipment makers' dealer lists and you get a few hundred companies; after that, more sites return only duplicates. New companies come from six channel types that comparison sites don't carry.
| Channel | Where they are | Why they're hard to find | Suited products |
| --- | --- | --- | --- |
| Competitors' primary agents | Agent-recruitment sites where they recruit sub-agents | They are the recruiters, so they don't appear on comparison sites | Same-category products pitched as add-ons |
| Multi-product field sales companies | Telecom, office-equipment and electricity agent networks; several primary agents run agent networks of their own | Their industry label is "telecom agent" or "OA sales" — they never say "agent" | Anything sold to shops and small businesses |
| Platforms' official agents | Official agent lists published by restaurant and salon booking platforms | Exist only inside each platform's site | Products for restaurants and salons |
| Industry distributors | Beauty distributors, commercial liquor wholesalers and food-service wholesalers on weekly delivery routes | Not sales companies at all, so no agent search ever surfaces them | Light products that ride along on existing visits |
| Rounders and field marketing | Companies contracted for shelf checks and POP installation in retail | Classified as "in-store promotion," not "sales" | Retail products that need installation |
| Comparison and referral sites | The usual sales-outsourcing and telesales comparison sites | Easy to find — and every competitor holds the same list | Only for the first few hundred at the entrance |
Run it and the five non-comparison channels fill up with companies the existing list didn't hold a single one of. Industry distributors in particular: zero on the list when the search began.
Seen this way, the pool keeps growing. Merchant-acquisition agents for food-delivery platforms, dealers of ticket machines and self-checkout, commercial liquor wholesalers, mid-sized regional beauty distributors — all of them are in shops every week.
### Step 3 — Collect, dedupe, rank
Collected candidates can't go straight to sales. They pass through three steps.
1. **Dedupe** — merge spelling variants (position of "Co., Ltd.", kana vs. roman letters) and the same company from different sites. One company, one row, with the channel it came from recorded as the source. The source shapes the approach: a company found on an agent-recruitment site is a company looking for products.
2. **Bundle group companies** — fold a group recruiting under several brands into one. Tells: same switchboard number, address, careers page or legal notice. Note in the remarks which contact point is the group's front door.
3. **Rank** — on three axes: touchpoint, organizational reach, contact reliability.
| Rank | Criteria | How sales treats it |
| --- | --- | --- |
| S | Own field sales network calling on shops, and a primary agent running its own agent network; or a distributor with weekly route visits. Contact verified on the official site | The lead approaches directly |
| A | Can do field sales but single-product or single-region. Contact verified on the official site | The sales team's first priority |
| B | Telesales- or referral-based, or too little information to confirm the touchpoint | Form-based batch proposal; promote on response |
Ranking exists to set the order of approach. Hit the S tier first and you learn how the product lands before grinding through the B tier.
### Step 4 — Verify contacts
Only contacts explicitly stated on the official site. Not one guess. Guessed addresses (info@, surname@) don't just bounce — they get flagged as spam and block your legitimate follow-ups too.
Keep contact status in three levels, in a column separate from rank.
| Status | Meaning | Handling |
| --- | --- | --- |
| Verified | Phone, form or email confirmed on the official site's company or contact page | Use as is |
| To confirm | Information found only in unofficial sources: job postings, map services, trade-association lists | One phone call before approaching, to confirm existence and department |
| Unconfirmed | Company name identified, official site not found | Managed separately as manual follow-up leads |
What a second check still can't find is not deleted; it's parked as "to confirm" in remarks. Delete it and you research the same company again next time.
In practice this step eats the most time. Opening each official site and matching the company profile against the contact page, a few hundred times over — and the temptation to "just fill it in by guessing" always arrives midway.
### Step 5 — Design the approach
The approach is not a mass mailing. It is decided per company — what, how, who — before anything is sent. Four items to decide.
| Item | Decision | Why |
| --- | --- | --- |
| Relation to existing products | If they already sell a competitor, pitch an add-on, not a switch | Nearly every company with a field network sells some competitor. A switch pitch ends at the first sentence |
| Non-compete and exclusivity | Ask in the first conversation whether existing contracts carry non-compete or exclusivity terms | Finding out "the contract doesn't allow it" after a warm meeting is the biggest time loss |
| Group front door | Approach a bundled group through one contact point only | Hit separate brands separately and the same person hears the same pitch three times |
| Contact method | Companies stating "no sales calls" get the form only, and the message opens by stating it is a partnership proposal | A phone call ends the relationship before it starts |
The first message leads with their benefit (higher ticket size from add-on proposals to existing customers, something to carry on existing visits) and puts your product description after it. They are looking for products, not for your company profile.
## Results — a case from in-store payment terminals (no names)
A list that had topped out at 674 companies from comparison sites and equipment makers' dealer directories gained 88 companies in two days through channel design. All 88 came from five channels the existing list held none of.
| Metric | Result |
| --- | --- |
| Entrance pool (comparison sites + equipment makers' dealer directories) | 674 companies |
| Added through channel design (two days) | 88 companies, from five channels beyond comparison sites |
| Rank breakdown | S 10 / A 33 / B 45 |
| Region | Outside the three major cities 43, Tokyo 26, Osaka 14, Fukuoka 5 |
| Contact status | Verified 75 (85%), to confirm 11, unconfirmed 2 |
| Email addresses | Only the 15 stated on official sites; 0 guessed |
| Manual follow-up leads (name identified, no official site found) | 16 companies |
Three things surfaced along the way.
- Comparison sites stop yielding from the second one on. Overlap was extremely high; staying on comparison sites alone would never have found the 88.
- Almost every candidate already handled a competitor's terminal. Every approach was switched to an add-on pitch, and a non-compete/exclusivity check was added to the first contact.
- Several candidates that looked like separate companies shared a switchboard number — the same group. They were bundled and approached through one front door.
The pool isn't exhausted. Regional members of platform agent lists, merchant-acquisition agents for food delivery, ticket-machine and self-checkout dealers, commercial liquor wholesalers and mid-sized regional beauty distributors remain untouched.
## What happens when you do it in-house
Start agent sourcing internally and you hit the same walls in nearly the same order. None of it is a question of ability; all of it is the cost of not knowing the method.
| Common approach | What happens | What you lose |
| --- | --- | --- |
| Collect 200 from comparison sites and hand them to sales | You call the same list your competitors hold, in the same order | New touchpoints as a first mover |
| Sweep a second and third comparison site | Duplicates, with a handful of new names | Days of someone's time |
| Mass-mail guessed addresses | Nothing lands; where it does, it's flagged as spam and blocks legitimate follow-ups | The company's trust |
| Approach one group's brands separately | The same person receives the same proposal three times | Their goodwill |
| Phone a "no sales calls" company | That company is closed to you for good | One S-rank candidate |
| Work top to bottom with no priority | The team burns out before reaching the top candidates and concludes "agents don't work" | The agent channel itself |
The last one costs the most. Companies that would have appeared with a different pool design go unfound, and the agent strategy is shut down as "we looked, there weren't any."
## How to work with me
This is for companies with products sold to shops and small businesses — payments, POS, SaaS, telecom, equipment, shop services — that are building an agent network from scratch or whose network has stalled. It suits in particular companies that have just entered the Japanese market and don't know where the domestic agents are.
The engagement follows the five steps above.
1. Discovery call (30 minutes) — product, target shops, existing channels, whether you have direct sales offices
2. Agent profile and channel design — which pool types to dig, in what order
3. List delivery — S/A/B rank, verified contacts, source, remarks (group, competitor products handled, phone OK or not)
4. Approach design — per-company approach and first message
5. Optionally, executing the first round of outreach
Working languages: Japanese, English and Chinese. English or Chinese progress reports for an overseas head office come from the same material.
### Nobu Sato
[mailto:info@nobusato.jp](mailto:info@nobusato.jp)
## Latest posts
- [ALL POSTS](https://nobusato.jp/blog/all)
A recruitment service backed by knowledge, insight, and expertise, delivering well-informed support tailored to your needs.
```json
{
"@context" : "https://schema.org",
"@type" : "BlogPosting",
"author" : {
"@type" : "Person",
"name" : "Nobu Sato",
"url" : "https://nobusato.jp/blog/author/nobu-sato"
},
"dateModified" : "2026-10-20T08:49:32.058Z",
"datePublished" : "2026-10-20T07:59:31.000Z",
"headline" : "A Method for Building a Sales Agent Network in Japan — How to Find, Screen and Approach Agents for Products Sold to Shops and Small Businesses",
"mainEntityOfPage" : {
"@id" : "https://nobusato.jp/blog/salespartner-en",
"@type" : "WebPage"
},
"publisher" : {
"@type" : "Organization",
"logo" : {
"@type" : "ImageObject",
"url" : "https://nobusato.jp/hubfs/nobusatojp-1.jpg"
},
"name" : "nobusato.jp"
}
}
```