Launching in Japan
Building a Sales Agent Network in Japan: Find, Screen, Approach
Start from channels, not from lists
Nobu Sato9 min read
Contents5
If you sell to shops and small businesses in Japan, your own sales team cannot cover the country. You need agents. And once you start looking for them, it goes surprisingly slowly. I have hit this wall in more than one Japan launch. The approach I settled on is simple to say: start from channels, then build the list. Here is how it works.
Why finding agents stalls
Effort is rarely the problem. The problem is how the information is laid out. The companies that reach shops are scattered across industries, and no single directory lists them.
- There is no master list. Agents are spread across telecoms, office equipment, payments, restaurant booking and salon distribution, and many of them never call themselves agents.
- The obvious sources run dry fast. Comparison sites top out at a few hundred companies. Cross-check more of them and you mostly get duplicates.
- The best agents already carry a competitor. If your product can't be sold as one more item in their bag, alongside what they already carry, you will keep getting turned down.
- Contact details are not published. Bulk email to guessed addresses doesn't arrive, and it costs you credibility.
- One group operates under several brands. Approach the same company three times without noticing, and that alone leaves a bad impression.
- A collected list is not a usable list. Hand a sales team several hundred names with no priority and they burn out before they reach the good ones.
If you start without seeing this, you end up where most teams do. The list keeps growing. The contracts don't.
The method in five steps
The core idea is one sentence: start from channels, not from a list. First decide which companies visit your target shops every week. Then find where those companies cluster. Only after you've collected them do you verify, rank and make contact.
| Step | What you do | Where doing it alone gets stuck |
|---|---|---|
| 1 Define the ideal agent | Work back from who touches your target shops weekly | Searching only for "sales outsourcing firms" and missing trade distributors |
| 2 Design the pool by channel | Source candidates from channels other than comparison sites | Stopping after a few hundred from comparison sites |
| 3 De-duplicate and rank | Merge duplicates, bundle groups, grade S/A/B | The same company under three names; no priority |
| 4 Verify contact details | Use only contacts published on the official site; zero guesses | Bulk email to guessed addresses |
| 5 Design the approach | Decide per company: existing products, non-compete terms, channel of contact | Spending time on a competitor's exclusive agent |
Steps 1 and 2 decide about eighty percent of the outcome. Steps 3 to 5 protect quality. Skip them and your sales team's time drains away.
Step 1: Define the ideal agent
Don't go out "looking for agents." First write down who is walking into your target shops this week. "Agent" isn't an industry. It's just the word for companies, in many different industries, that already have contact with shops.
For products sold to restaurants, retailers and salons, the candidates look something like this:
- Multi-product field sales firms selling phone lines, copiers or electricity door to door
- Authorized agents of restaurant-listing and booking platforms
- Salon distributors delivering supplies and equipment on weekly routes
- Retail merchandising firms that visit store shelves on a fixed rotation
- Resellers of POS systems, ticket machines and commercial equipment
The definition needs only three items. The form of contact (weekly route visits beat cold visits, which beat phone only). Product fit: does yours ride along as one more item, or is the agent exclusive to a competitor? And region. This definition tells you which channels to dig in Step 2.
Step 2: Design the pool by channel
Comparison sites are a fine place to start. They are a bad place to stop. New companies come from channels that never show up there.
| Channel | Where they are | Why they are hard to find |
|---|---|---|
| Competitors' primary agents | Agent-recruitment sites where they recruit sub-agents | They are on the recruiting side, so comparison sites never list them |
| Multi-product field sales firms | Telecom, office-equipment and power agent networks | They are labelled "telecom agency" or "OA sales," never "agent" |
| Authorized platform agents | Official agent lists published by restaurant and salon booking platforms | The lists exist only on each platform's own site |
| Trade distributors | Salon distributors, commercial liquor, food wholesale | They are not sales firms, so no agent search finds them |
| Retail merchandisers | Firms contracted for shelf rounds and in-store displays | Classified as in-store promotion, not sales |
| Comparison sites | Sales-outsourcing and telemarketing comparison sites | Easy to find, but every competitor has the same list |
Step 3: De-duplicate and rank
Collected candidates can't go straight to sales. Merge spelling variants and duplicates into one row per company, and record which channel each came from. Spot groups running several brands by a shared main number, address or legal disclosure page, and bundle them as one. Then rank on three axes: form of contact, organizational reach, and how reliable the contact details are.
| Rank | Criteria | Handling |
|---|---|---|
| S | Runs its own field sales force and its own agent network, or a distributor with weekly routes | Approached directly by a senior lead |
| A | Can do field sales, but mainly one product or one region | Prioritized by the sales team |
| B | Mainly phone or referral, or the form of contact cannot be confirmed | Proposal by web form; promoted if they respond |
The point of ranking is the order of approach. Talk to the S tier first and you learn how the market receives your product before you get to the B tier.
Step 4: Verify contact details
Use only contact details published on the company's official site. Not a single guess. Track status in three levels (confirmed, needs checking, unconfirmed), separately from rank. Don't delete what you couldn't find. Keep it as "needs checking," or you will research the same company again next time.
In practice this step takes the longest, and somewhere in the middle you will be tempted to fill the gaps with guesses. A list that cuts corners here will cost your sales team its credibility later.
Step 5: Design the approach
No bulk sends. For each company, decide four things before you reach out.
- Relationship to existing products: if they carry a competitor, propose an addition. Don't ask them to switch.
- Non-compete and exclusivity terms: confirm them in the first conversation. Finding out "we can't, contractually" after the conversation has warmed up is the biggest waste of time there is.
- Group contact point: approach a bundled group through one central contact only.
- Channel of contact: if a company says "no sales calls," use the web form only. Call them anyway and the relationship ends there.
Open the first message with what the agent gets out of it: a higher ticket from upselling existing customers, something that rides along on visits they already make. Put your own company description last. They are looking for products to sell. Nobody is reading your brochure.
What happened on one engagement
One example. For a hardware product sold to shops, a list built mainly from comparison sites and reseller directories had stalled at a few hundred companies. After switching to channel design, we added new companies equal to just over ten percent of the original list in two days. Every one of them came from a channel other than comparison sites. Trade distributors in particular had been completely absent from the list when the search began.
About half the new candidates ranked S or A. More than eighty percent had officially confirmed contact details, and there were no guessed email addresses. A few things became clear along the way.
- After the first comparison site, more sites produce almost nothing new. If we had stayed with comparison sites, we would have found none of the additions.
- Nearly every candidate already carried a competitor's product. So we positioned everything as an add-on and built a non-compete check into the first contact.
- Several candidates that looked separate turned out to be one group sharing a main phone number.
What happens when you do it alone
Teams that start agent sourcing in-house tend to hit the same walls in the same order. Ability has little to do with it. It's what not knowing the method costs.
| Common approach | What happens | What it costs |
|---|---|---|
| Collect from comparison sites and hand to sales | You work the same list as your competitors, in the same order | First-mover access |
| Work a second and third comparison site | Mostly duplicates; a handful of new names | Several days of someone's time |
| Bulk email to guessed addresses | Nothing arrives; spam filtering blocks legitimate contact too | The other side's trust |
| Approach one group through separate brands | The same person receives the same pitch three times | Their goodwill |
| Work the list top to bottom with no priority | The team tires before reaching the best candidates and concludes "agents don't work" | The agent channel itself |
The last one is the most expensive. The companies a different pool design would have surfaced are never found, and the whole agent strategy gets closed down because "we looked and there was no one."
In the end, it is a question of sequence
When agent sourcing fails, the instinct is to search more and call more. Volume rarely fixes it. Sequence does: define who visits the shops, work out where they are, verify, then start with the strongest. In a Japan launch the order of your first hires decides almost everything, and a sales network is the same. Most of what happens to it is settled by what you do first.
If you are thinking through a sales network for the Japanese market, feel free to get in touch.
On how to screen a competitor list, see Picking the wrong competitors is worse than having none.
Building the agent pool is close to building a talent map, and the map has to answer a specific question: A talent map is not a list. Here's the question it answers. For the problems foreign companies hit before they even get to agents, see What foreign companies struggle with most in Japan.
On how the customer layer decides the channel, see SMEs are 99% of Japanese companies.
What happens after signing, and why distributors don't sell: Why your distributor isn't selling your product.
Related: A joint venture with a Japanese partner: shortcut or detour?