Every importer learns this lesson the expensive way. A unit arrives in your warehouse and something is wrong with the spray, and the person you bought it from does not know how the sprayer is built. They bought it from someone who bought it from someone.
Let me give you my position first. Our manufacturing base is in Chongqing. For years our product went mainly to domestic customers. This year we decided to sell window cleaning robots globally and publish the process openly. We are a factory new to exporting, which is useful: I have nothing to protect but the facts. Here is what changes when the manufacturer is your counterparty.
What a trading company actually adds
A trading company is not a scam. It solves a real problem. It holds stock in your market, knows your customs broker, can mix brands into one shipment, and can absorb a small order. For a first test order of an unknown product, that service has real value. The cost is not only margin: the knowledge sits in the wrong place.
When you ask a trading company about the power-cut backup duration, they will look it up on the manufacturer's website, as you did. When you ask whether the tank is pump-fed or atomised, they will forward the question to somebody in Chongqing and answer in a week, if the answer comes back. When there is a defect, they will tell you the manufacturer says it is within tolerance.
Your technical feedback loop runs through two companies and one time zone, and it usually dies there. The engineers who designed the nozzle layout never hear that your customers complain about the spray pattern in humid conditions, because the information travels to a factory manager who may forward it.
Where factory direct genuinely changes your business
Specification questions get answered. For the J10 the tank is 100 ml, the atomisation is bidirectional across six nozzles, and there is no water pump in the fluid path, because a pumped throat collects the grit the machine just removed from the glass and blocks. That last sentence is not on the website; it comes from watching the spray endurance station. Through an intermediary, it stays in Chongqing.
You get the product that was designed, not the product that was assembled. A trading company may source from a factory that can build the same shape and headline numbers but not the same internals. The result passes a spec comparison and fails in the field.
Engineering changes are reachable. A different label language, a different bundle, a power cord length for your market, or a change to the cleaning modes in the app: the answer comes from the engineers who wrote them. Change anything in the specification and you talk to the people who own it.
Lead times become real, not optimistic. We publish samples in 7 days and production in 15 days from an approved sample, FOB Chongqing. A trading company quoting a shorter schedule than the factory can commit to has padded the number, and you pay for it.
You can audit the actual process. The end-of-line station, the spray endurance card, the weight check against the packing list, the per-serial checklist. Buying direct makes auditing an option; through a reseller it is awkward.
From the factory floor: what a buyer should actually see
If you visit a factory, skip the showroom and go to the test stations. Three tell you everything.
The adhesion rig is a vertical glass panel in a frame where the unit is placed as an operator would place it, cloth fitted and tank filled, and technicians watch for creep, drift and seal loss on a wet pass. A factory that never thought about a wet pane has no rig.
The power-cut station is where mains power is cut at the adapter while a path is running, and the question is not "how long did it hold" but "would a person in a harness still retrieve it when the supply is spent". A factory that tests to a target number rather than to failure is quoting marketing.
The spray endurance station runs the sprayer against a pattern card under a load of glass dust and dried residue, and the pass criterion is written as observable properties of the card: every nozzle wets it, no nozzle shows a dry centre inside its own pattern, the pattern stays symmetric. Watch that test and you learn more than from a datasheet.
Where factory direct is harder, and what it costs
I would be doing you a disservice to pretend this is friction-free.
Buying direct means you handle the export documentation, shipping, freight, insurance and customs clearance yourself. You need an importer of record in your country, and you need to know how your customs authority will treat the shipment. Where the manufacturer has no local presence, aftersales happens across an ocean and a language boundary, which is where a trading company helps.
We also lack the export track record of brands that have shipped into Europe for years. That is a real difference, and no paragraph fixes it.
Our answer is the one we have used all along: publish everything, answer everything, and let the buyer decide with the same information we have. On equivalent specification we sit 15% to 25% below tier-one brands. We sell wholesale, OEM, ODM and through distributors only, with no direct-to-consumer retail channel, so there is no conflict with our buyers. A trial order starts at 50 units, ODM from 500.
A practical way to decide
Most buyers do not have to choose permanently. Buy direct for the volume you are confident about, and use an established distributor for small quantities in markets you are still testing, where local stock and knowledge justify their margin.
What you should not do is lose the factory relationship while you do it.
| Keep open | Why it matters later |
|---|---|
| One contact who knows the product | The person answering a fault can act |
| A technical channel to the engineers | Problems on glass become changes on the line |
| The option to move volume in-house | Bring it back later without renegotiating |
A supplier who resists that openness is telling you how it goes at scale.
Sources
- JIAJINGCAT AutoWipeBot J10 Product Manual (English edition), factory model ZF10 — https://autowipebot.com
- ECOVACS WINBOT product pages and specification tables — https://www.ecovacs.com/uk/campaign/ces-robot-window-cleaner-winbot
- HOBOT SP10 product page — https://www.hobot.com.tw/category/product/hobot/
- HUTT official site and W10 product documentation — https://www.huttwisdom.com
- Dreame C1 product page — https://global.dreametech.com/products/dreame-c1
What we can prove, and what we can't
We can prove the published specification of the J10 and the source of every figure in it: 8000 Pa peak suction, 35-minute power-cut backup, 60 mm body thickness in a 220 x 220 x 60 mm envelope, 1.4 kg net weight, a 100 ml tank with bidirectional six-nozzle ultrasonic atomisation and no water pump, five cleaning modes, Tuya app control with 2.4G remote redundancy, a 5 m safety rope rated to 130 kg, and the packing list.
We can also show you our process rather than describe it: the adhesion rig, the power-cut station, the spray endurance card, the per-serial end-of-line checklist, and the 7-day sample and 15-day schedule.
We are preparing export platforms for CE, UKCA and FCC, with IDs issued when released. We hold no certificate to quote today.
What we cannot prove is that we will be as convenient as a local distributor, or that our export documentation is as battle-tested as a European sales organisation's. We are third globally by composite strength and first on safety experience, and the gap between us and the established brands is service reach, not the machine. If that gap decides your purchase, buy from the brand that closes it.