I Work FOR YOU, Not Factories.

I’m Leon Xu, based in Shenzhen with 15 years in consumer electronics. You’ve built a longlist of Chinese manufacturers. Now you need to reduce it to the few worth visiting — without wasting a month or flying to China prematurely.

How to Shortlist Manufacturers in China Before Visiting Factories

By Leon Xu | Easelink Tech | Shenzhen, China

You Should Not Visit Every Factory You Find

Here’s a mistake I see constantly: an overseas team identifies twenty potential manufacturers, then tries to visit all of them. The result is a blur — a week of factory tours, twenty near-identical workshops, and a pile of business cards that all blur together into “they seemed fine.”

The point of a factory visit is not to discover which factory is right. The point is to verify what your research already told you about the two or three that are worth your time. If you’re using factory visits as your discovery method, you’ve inverted the process — and you’ll pay for it in travel cost, decision fatigue, and a final choice made on exhaustion rather than evidence.

The goal is simple: visit the right two or three factories, not all twenty. Everything before the visit exists to get you to that small number.

Build a Longlist First

The process starts, as it always should, with a broad longlist — 15 to 25 plausible manufacturers collected from trade shows, B2B platforms, referrals, and competitive research. At this stage you’re gathering names, not evaluating them. Breadth is the goal, and it should be cheap: an afternoon of research, not a month of correspondence.

The longlist exists to ensure you’re not anchoring on the first factory you found. A team that starts with one referral and immediately pursues it has no basis for comparison. A team that starts with twenty names and works down has a real market view.

If you haven’t built a proper longlist yet, start with the framework for finding manufacturers.

Create Supplier Screening Criteria

Before you screen, define your criteria — and weight them according to your project. The criteria should cover the dimensions that determine fit:

Product experience — have they built this category before?
Engineering capability — electronics, mechanical, firmware, DFM, NPI.
Manufacturing capability — process, equipment, capacity, testing.
Customization — can they do what your product requires?
Communication — language, clarity, responsiveness.
Commercial terms — MOQ, price range, payment terms, lead time.
Quality system — certifications and, more importantly, real QC process.

The key is weighting. A complex, engineering-heavy product weights engineering and DFM heavily; a simple, high-volume product weights manufacturing capacity and cost. The same criteria, different priorities. Write the weights down — otherwise you’ll drift toward whatever is easiest to compare, which is usually price.

Screen Suppliers Remotely

The first screen is done without leaving your desk. Websites, product listings, capability statements, and a short qualification exchange. At this stage you’re answering binary questions: Does this factory make products like mine? Are they in my volume range? Do they claim my required certifications? Do they serve my target market?

This screen should cut the longlist from 20+ down to maybe 8–12 in a few days. It’s fast because it’s shallow — you’re not evaluating deeply, you’re removing the obviously unsuitable. The factories that survive are the ones that plausibly could be right, not the ones you’re sure are right.

Identify 5–10 Qualified Candidates

After the first screen, you have a pool of 8–12. Now you deepen the engagement enough to separate the qualified from the merely plausible — technical discussions, document requests, and video calls. The goal is to reach 5–10 candidates who are genuinely qualified for your project.

This is where the superficial factories fall away. A factory that looked good on its website but can’t answer a technical question, or can’t produce a coherent quality process description, reveals itself quickly under a little scrutiny. You’re not trying to find the winner yet; you’re trying to eliminate the ones that can’t clear a real bar.

Conduct Deeper Evaluation

With 5–10 qualified candidates, the evaluation gets serious. This is where you assess the dimensions that actually determine success — engineering depth, DFM capability, supply chain discipline, quality management, and communication quality. I’ve detailed this process in the supplier evaluation article.

The output of this stage is a clear ranking: which candidates are strong, which are marginal, and why. You should be able to articulate, for each remaining candidate, what makes them a fit or a risk. If you can’t articulate it, you haven’t evaluated deeply enough.

Choose the Right Factories to Visit

Only now — with a ranked set of qualified candidates — do you select the two or three to visit. The selection should be driven by your evaluation, not by geography or convenience. Visit the strongest candidates, even if they’re in different cities; a factory visit to a marginal candidate is wasted travel.

The visit itself should be structured around verification. You’re checking that the factory’s engineering, processes, quality system, and people match what they claimed during evaluation. Bring your specific product requirements, your questions, and your skepticism. A factory visit is a confirmation exercise, not a discovery exercise.

Use Factory Visits to Verify Your Findings

This is the principle that ties the whole process together: a factory visit should verify what you already believe you know, not discover everything from zero. If you arrive at a factory with no prior evaluation and no specific questions, you’ll be shown a polished tour and leave with an impression, not an assessment.

If you arrive with a ranked evaluation and a checklist — engineering team present? quality process documented? equipment matching claims? — the visit either confirms or contradicts your findings. Both outcomes are useful. Confirmation gives you confidence; contradiction gives you a warning you’d have missed otherwise.

For more on what actually happens on a visit and what it can and can’t tell you, see what you see in 5 days versus what you need in 5 months.

Conclusion

You should not visit twenty factories. You should identify the right two or three. That means building a longlist, screening it down, evaluating the qualified candidates deeply, and reserving factory visits for final-stage verification of the strongest few.

Done this way, a factory visit is a high-value confirmation of a well-researched decision. Done the other way, it’s an expensive blur. The process — not the trip — is what finds your manufacturing partner.