6 Months Building a B2B Platform: What I Got Wrong About Buyers

Every founder has a story about building something they were certain buyers wanted, and discovering they had built something slightly, or entirely, different from what buyers needed.

Here is mine.

What I Assumed About Buyers

When I started building TradesLynk, my mental model of the “buyer” was an import manager at a mid-sized food distribution company in Europe or the UK. They’re experienced, they know what they want, they have a sourcing process, and they’re looking for a more efficient way to find verified exporters (https://tradeslynk.com/exporters).

Based on that model, I prioritised two things early: exporter discovery (robust filtering, search by category (https://tradeslynk.com/categories) and country) and trust signals (verification badges, company profiles). I assumed the buyer’s primary need was finding credible exporters (https://tradeslynk.com/exporters) they couldn’t find before.

That assumption was directionally correct but incomplete in ways that took six months to fully understand.

Mistake 1: Underestimating the First-Time Sourcer

The experienced import manager is real. But a significant portion of people who search for agro-commodity suppliers online are not experienced sourcers. They are:

• Business owners starting to import for the first time

• Restaurant and food chain procurement teams sourcing specialty ingredients

• Consumer packaged goods companies exploring new commodity inputs

• Retail buyers looking to add private-label products (https://tradeslynk.com/products) to their range

These people don’t have a sourcing process. They don’t know what documents to ask for. They don’t know the difference between a CIF and an FOB price. They’re not looking for a filtered directory — they’re looking for a guided sourcing experience that tells them what to do next. We’ve tried to address this with our Knowledge Hub (https://tradeslynk.com/learning), which covers everything from Incoterms to certification requirements.

I had built for the expert and underbuilt for the beginner. The feature set was correct for experienced buyers but intimidating or insufficient for everyone else.

This changed how I thought about content and onboarding. A buyer who doesn’t know what an IEC certificate is needs to know why it matters before the verification badge means anything to them.

Mistake 2: Assuming Buyers Know What They Want

The second mistake was assuming that buyers arrive at the platform with a clear, formed requirement. In practice, requirements are often approximate.

A buyer might know they want “turmeric (https://tradeslynk.com/category/spices/turmeric)" but not know whether they need Alleppey finger turmeric (high curcumin, for pharmaceuticals) or Erode turmeric (brighter colour, for food manufacturing). They might know they want “20 metric tons” but be unclear on whether they mean 20 MT per shipment or 20 MT per quarter. They might have a destination market in mind but not fully understand the import regulations in that market.

I designed the RFQ system assuming buyers would arrive with complete specifications. What I found in early testing was that many buyers needed a discovery process before they could even write a proper RFQ. They needed to see what was available in spices (https://tradeslynk.com/category/spices), grains (https://tradeslynk.com/category/grains), or pulses (https://tradeslynk.com/category/pulses), what specifications exporters typically offered, what price ranges looked like — before they could crystalise their requirements.

This taught me that the platform needed to support exploratory sourcing, not just confirmatory sourcing. Browsing, comparing, and learning should be first-class experiences, not just a pre-filter before the inquiry form.

Mistake 3: Thinking Buyers Would Always Initiate

My initial model assumed buyers drive the transaction. They search, they find, they contact. Exporters (https://tradeslynk.com/exporters) respond.

What I underestimated was the degree to which well-matched RFQs and notifications change exporter behaviour — and how that exporter responsiveness signals quality back to buyers.

When we built the RFQ matching system that notifies verified exporters (https://tradeslynk.com/exporters) of relevant RFQs, response rates went up. And when buyers saw faster responses, their perception of platform quality increased — even though the buyer hadn’t changed anything about how they searched. The exporter side had changed, and that improvement was visible to the buyer.

This sounds obvious in retrospect. But it changed my view of who I was building for. I was not building for buyers and exporters (https://tradeslynk.com/exporters) separately. I was building for the quality of the interaction between them. The platform’s value is not the sum of what it does for each side independently — it’s the quality of what it facilitates between them.

Mistake 4: The Response Time Assumption

I assumed buyers would be patient enough to wait a week for an exporter response, particularly given the cross-timezone nature of most trade interactions.

They are not. Even in B2B trade — even where deals take months to close — buyers make snap judgements about platform quality based on initial response times. If a buyer sends an inquiry on Tuesday and hears nothing by Thursday, they have already mentally discounted that exporter.

This changed how I thought about the notification and engagement system. Email notifications for inquiries are insufficient. We needed to make the case to exporters (https://tradeslynk.com/for-exporters) directly: the speed of your first response affects your trust score, which affects your visibility in search results, which affects your inquiry volume. Response time is not a nicety. It is a ranking signal.

What I Got Right

To be balanced: some assumptions held.

The hypothesis that buyers filter heavily by verification status held. When we added the verified badge and allowed filtering by verification tier, conversion rates from browse to inquiry went up. Buyers do use trust signals if those signals are clearly communicated.

The hypothesis that a zero-commission model would be a genuine differentiator also held. Multiple buyers told us — unprompted — that they appreciated knowing TradesLynk wasn’t incentivised to push them toward any particular exporter (https://tradeslynk.com/exporters). That trust in the platform’s neutrality transfers to trust in the search results.

The Honest Summary

Six months of building a B2B platform taught me that buyers are not a monolith. They are experienced sourcers and first-time importers. They are clear on requirements and still forming them. They are patient and they are also making instant judgements about whether a platform is worth their time.

The right product serves all of them — not by being all things, but by being clear about what guidance it provides and making every interaction feel credible.

We’re still building. But the mistakes made the product better.

TradesLynk (https://tradeslynk.com) is live and evolving. Browse verified exporters (https://tradeslynk.com/exporters), explore products by category (https://tradeslynk.com/products), or visit our Knowledge Hub (https://tradeslynk.com/learning) to understand the platform. If you are a buyer sourcing agro-commodities, I’d genuinely welcome your feedback on what’s missing.