Before We Begin
This is Lesson 1 of Finding Factory Customers: 10 Lessons for Industrial Sales.
The full series builds a complete playbook for finding factory customers from the ground up: Lesson 1 answers who to call, Lesson 2 answers who to call first, Lesson 3 teaches how to build the list, and subsequent lessons walk through the opening, the decision chain, timing, lead qualification, long-cycle deal management, and finally, in Lesson 10, how to turn every won deal into the next batch of leads. Ten lessons in sequence, each doing exactly one thing.
Lesson 1 starts with the most fundamental question: Before you pick up the phone, do you actually know which type of factory you should be selling to?
I. 600 Calls in a Quarter, 3 Deals Closed
A company selling industrial dust-removal equipment. Sales team of 4. Quarterly target: 12 deals.
Their list came from filtering a business registration database by "business scope contains: machinery/manufacturing," then manually going through the results by province. Over the quarter, they called roughly 600 companies and closed 3 deals.
The team did a post-mortem:
- About 120 of the 600 were traders or stall traders — they generate no dust at all
- About 90 had workshops too small to need new equipment; their existing units were sufficient and they had no motivation to upgrade
- About 150 were textile or food-processing factories, whose dust-removal needs are entirely different from metalworking shops — the equipment models simply don't match
- About 80 did have a genuine need, but their annual output was below 5 million yuan, meaning a single dust-removal unit would consume a third of their entire annual equipment budget, making the decision cycle extremely long
- Factories that truly matched on all three counts — right size, right process, motivated to upgrade — numbered only about 60
60 target customers were scattered across a list of 600. 90% of the calls were wasted.
Translated into cost: 4 salespeople, a quarter of 60 working days, effective contact rate of 10% — meaning roughly 54 days were spent on fruitless outreach. At an all-in cost of 15,000 yuan/month per salesperson, 4 people × 3 months = 180,000 yuan in labor, of which at least 160,000 yuan was spent chasing the wrong list.
This is not a pitch problem. If the list is built wrong, no pitch will save you.
II. Why "Business Scope" Can't Filter What You Need
The "business scope" field in a business registration is, in essence, self-reported by the company. It tells you what the company is registered to do — but not:
- What it is actually producing today
- What type of process its workshop runs
- Whether that process generates dust, fumes, or exhaust
- Whether its scale clears the threshold for equipment investment
- Whether it has any motivation to replace its current equipment
The four characters "mechanical manufacturing" can cover everything from a small forge with 2 million yuan in annual output to a precision-parts plant with 2 billion. The two have completely different needs, budgets, and procurement processes. Putting them on the same call list means running two entirely different businesses at the same time.
The root problem is this: when most salespeople define "target customer," they start from their own product — but never complete the chain from product to factory pain point to factory characteristics.
III. The Product-Backward Method: Three Steps to a Profile
The logic of the "product-backward method" is: what you sell eliminates a specific loss a factory suffers; that loss is largest at a certain type of factory; that type of factory is your ideal customer. Starting from the loss and working back to factory characteristics is more precise than starting from an industry category.
Step 1: Write Down Exactly Which Specific Loss Your Product Eliminates for Factories
Not "improves efficiency." Not "reduces costs." It must be specific to a process and a magnitude.
Force yourself to answer this one question: If a factory doesn't use your product, what happens six months from now?
- Industrial dust-removal equipment: Workshop dust concentration exceeds the legal limit, triggering liability risk for occupational disease compensation, or a mandatory production halt during an environmental inspection — losing a week of output
- Injection-mold temperature controllers: Surface temperature variance exceeds 5°C across the mold, causing defect rates to rise 3–5 percentage points per batch; at 5,000 units per day, that translates to tens of thousands of yuan in wasted materials per month
- CNC tooling management systems: Unpredictable tool life means either late tool changes that scrap workpieces, or premature changes that waste remaining tool life — both cost money
After writing, check your answer: if you wrote "reduce headcount" or "digital transformation," you need to go one level deeper — which specific operation, which type of labor, how many hours saved, and how much does that work out to in yuan?
Step 2: Which Type of Factory Suffers This Loss Most?
Trace the loss upward and you will find four converging dimensions:
Scale band: The loss must be large enough that a factory is willing to pay for your solution. If your equipment costs 200,000 yuan per unit, the target factory's annual equipment investment budget should ideally be at least 800,000 yuan — otherwise your single unit exhausts their budget flexibility. Factories with annual output of 50 million to 300 million yuan are typically the sweet spot for industrial sales: large enough to have budget, small enough that the decision chain isn't too long.
Process type: Dust-removal equipment applies to metalworking, casting, grinding, and welding — not food processing or textile dyeing. Mold temperature controllers apply to injection molding — not stamping or forging. List out the process dimension and directly eliminate inapplicable industries.
Export presence: Export factories hold product consistency and equipment reliability to a standard one level higher than domestic-only factories. If your equipment's primary value is improving yield rates, factories with export orders make procurement decisions faster than purely domestic ones — they cannot afford the cost of high defect rates.
Industrial cluster affiliation: Industrial clusters where a given process is concentrated often develop industry-wide purchasing norms. Once a piece of equipment builds a reputation in a cluster in Guangdong, the sales cycle for dozens of similar factories nearby shrinks noticeably. The more concentrated the cluster, the more effective word-of-mouth spreads, and the more valuable it is to plan your sales route around it.
Cross the four dimensions and you will find your profile condensing from the vague "manufacturing sector" into a recognizable shape.
Step 3: Write the Profile as a Single Filterable Sentence
The format of an Ideal Customer Profile sentence:
"Factories with annual output in the range of [revenue band], engaged in [process type], located in [region/industrial cluster], and with [signal condition]."
The "signal condition" is the key — it is your filter for judging whether a factory has purchasing motivation right now. Common signal conditions include:
- Recent equipment investment record (indicates capital-expenditure willingness)
- Currently hiring for production, engineering, or equipment-maintenance roles (indicates capacity expansion)
- Has export orders (higher product consistency requirements)
- Recent bidding history (equipment procurement follows a structured process, indicating this is not a small, ad hoc operation)
A profile without a signal condition is just an industry-plus-scale box. Calls made from it will still produce large volumes of "wrong timing" contacts.
IV. Two Mini Case Studies
Case 1: Industrial Dust-Removal Equipment
A company selling industrial dust-removal systems. Unit price: 150,000–500,000 yuan. Core applicable scenarios: dust control in metalworking, casting, and grinding workshops.
Step 1: Loss eliminated — liability risk of occupational disease compensation and forced production shutdown due to excess workshop dust. One week of halted production for a factory with 80 million yuan in annual output means a direct loss of more than 1.5 million yuan.
Step 2: Four-dimension convergence:
- Scale: Annual output 30 million–300 million yuan (has equipment investment capacity, but not so large as to require a formal bidding process)
- Process: Metalworking, casting, grinding (generates dust; excludes food/textile)
- Region: Prioritize the Pearl River Delta (concentrated electronics/precision-parts metalworking) and the Yangtze River Delta (concentrated machinery manufacturing)
- Signal: Recent record of workshop expansion, or currently hiring for production safety management roles
Step 3: ICP in one sentence — "Factories with annual output of 30 million–300 million yuan, engaged in metalworking or casting, located in the Pearl River Delta or Yangtze River Delta, and with recent workshop expansion or current openings for safety management roles."
Tianxia Gongchang covers 4.8 million China-based physical manufacturing enterprises. Filtering this profile by crossing industry tags and scale bands, this company estimated that roughly 20,000 factories nationwide meet the criteria. 20,000 factories for a 4-person sales team is a scale at which a systematic approach — divided by region and by batch — is genuinely executable: large enough to sustain growth targets, small enough not to be paralyzing.
Only after refining the profile to this precision did they begin building the list.
Case 2: Injection Mold Temperature Controllers
Another company, making injection mold temperature controllers. The product stabilizes mold temperature to reduce defect rates in injection-molded parts. Primary customers: injection molding factories for automotive components, home-appliance housings, and precision consumer electronics.
Step 1: Loss eliminated — material waste caused by unstable mold temperature. For an injection molding factory with 50 million yuan in annual output, dropping the defect rate from 4% to 2% saves roughly 300,000 yuan per year in material costs (assuming raw material costs at 30% of output value), with the equipment paying back within two years.
Step 2: Four-dimension convergence:
- Scale: Annual output 20 million–150 million yuan (below 20 million, the factory typically has only a single injection molding machine and the mold temperature controller is low-priority; above 150 million, large factories usually have in-house equipment maintenance systems and longer procurement cycles)
- Process: Must be injection molding (an extremely constraining dimension that directly eliminates 90% of manufacturing)
- Export presence: Injection molding factories with export clients hold higher requirements for product appearance consistency and lower tolerance for defect rates — purchasing decisions are faster
- Signal: Currently hiring injection molding operators or mold technicians (indicates capacity expansion, with correspondingly higher yield requirements)
Step 3: ICP in one sentence — "Factories with annual output of 20 million–150 million yuan, engaged in injection molding, with export orders, and currently hiring injection molding or mold-related production roles."
This profile is narrower than the first. Based on industry distribution and industrial cluster distribution in Tianxia Gongchang, this company's national TAM is approximately 5,000–8,000 factories, concentrated in Zhejiang (Taizhou injection molding, Ningbo precision manufacturing) and Guangdong (Dongguan, Shenzhen consumer electronics).
The TAM is not large, but it is sufficient — a 4-person sales team covering 6,000 factories over two years for one deep-touch cycle is achievable. If the number were too small, you would need to loosen one layer on the scale floor or the process type; if too large, add one more signal condition to narrow it. This step is about checking the profile against real numbers, not guessing.
V. What Tianxia Gongchang Does at This Step
Once the profile is written, the next action is validation: How many factories in the country actually match this profile?
That number determines your market-entry decision.
Too few (say, fewer than 500): the profile is too narrow, or the market itself is too small to sustain a sales team's growth targets — loosen one or two dimensions.
Too many (say, more than 100,000): the profile is not yet tight enough; the signal condition is not filtering out non-buyers, and outreach will still generate large volumes of noise — add one more dimension.
The right range depends on team size. A sales team of 4–6 people typically works well with a TAM of 5,000–30,000 factories: enough to sustain systematic outreach, with sufficient funnel depth.
Tianxia Gongchang integrates industry tags, industrial cluster affiliation, scale tiers, and signal tags into a single structured database, covering 4.8 million China-based physical manufacturing enterprises. Salespeople can filter using the dimensions from their profile and immediately see "how many factories satisfy these conditions" — without building a list first and filtering by hand.
The value of this step is not just the number. It lets a salesperson know — before making the first call — how large a market they are working, whether there is enough space, and whether the profile needs adjustment.
A profile that has never been validated against TAM will either overestimate the market (wasting resources on unreachable customers) or underestimate it (hitting a ceiling too soon). Work out the numbers first. Then begin.
VI. Three Common Profile Mistakes
Mistake 1: Treating an industry as a profile
"Our target customer is automotive components factories" — that is an industry, not a profile. Automotive components factories range by scale from a small shop with 3 million yuan in annual output to a Tier 1 supplier with 5 billion. Their needs, budgets, and decision processes are entirely different. A profile must include scale and signal conditions.
Mistake 2: A profile so broad it is only there to reassure yourself the market is large enough
"Our target customers are all manufacturing factories with annual output above 50 million yuan" — this covers millions of factories, a number no sales team has the reach to touch. A broad profile is just a way of avoiding the truly hard question: which type of factory is most likely to close?
Mistake 3: A profile that only exists in someone's head, never written into a filterable sentence
A profile must be translatable into database query conditions. If you cannot write out "scale range + process type + signal condition" as three explicit items, the profile is not finished. A profile that lives only in someone's head will be understood differently by every member of the sales team — and list quality will be inconsistent as a result.
VII. A Profile Template to Take Away
After completing a round of thinking, use this template to check whether your profile is complete:
Ideal Customer Profile Checklist
Which specific loss does my product eliminate for factories? (Write it down to the process and the dollar amount — "improves efficiency" is not acceptable)
What scale range do the factories that suffer this loss most fall into? (Write out the lower and upper bounds of annual output)
Which process type causes a factory to encounter this loss? (Specific to process, not industry category)
Is there a regional or industrial cluster preference? (Spread across the country vs. concentrated in a few clusters)
What signal indicates that a factory "has purchasing motivation right now"? (Write at least one observable signal)
Assemble the five items above into a single sentence. Can that sentence be translated into database filter conditions?
Approximately how many factories in the country match this profile? (Does the TAM support the team's scale?)
One-sentence ICP template:
"Factories with annual output of [A 10,000s–B 100 millions yuan], engaged in [process type], located in [region/industrial cluster], and with [signal condition], in the [industry] sector."
After each quarterly won-deal review, revise this sentence once using actual closed-deal data. The first version of the profile will rarely be fully accurate — but it is your baseline for building the list and making outreach. Without this sentence, every step that follows is guesswork.
With the profile defined, the next step is prioritization — not every factory on the list deserves equal effort. Lesson 2, Not Every Factory Is Worth Calling — Tiering Factory Customers into S/A/B/C, covers how to score factories on four to five dimensions and allocate your limited sales time to the batch most likely to close.