9
17 Comments

Why I Got 1 Million Users in 6 Months—Even with So Many B2B Platforms Already Out There

It might sound unlikely.
There are already tons of B2B procurement platforms out there. They all compete on the same things: number of sellers, product SKUs, industry coverage, safety, service...
Many of them have been around for over a decade. Big teams. Big funding. Lots of experience.
We started from zero. And in just 6 months, we reached over 1 million users.
No funding. No ads.
We just did one thing right: we understood what users actually need—better than others did.

The usual "differentiation" no longer works

Visit any major B2B platform and you'll see the same claims:
● 50 million+ products
● 10+ years of industry experience
● Verified suppliers
● Coverage across dozens of industries
All true. All valid. But none of them are true differentiators anymore.
We talked to a lot of B2B users—procurement leads, category managers, e-commerce sellers.
The feedback was surprisingly consistent:
"I'm not worried about finding products. I'm worried about picking the right ones."

"I don't know what to buy" is now the biggest challenge in procurement

In the past, product trends moved slowly. A good product could stay popular for years.
Not anymore.
Social media, global e-commerce, and short video platforms like TikTok have changed everything.
Trends now move like the wind—fast and unpredictable.
● A product can go viral in a week and disappear the next
● Manufacturers are still producing while sellers are already discounting
● If you start sourcing too late, you’re left with dead inventory
For B2B buyers, catching trends early makes the difference between profit and loss.

That’s why we focused on this question: What if the platform tells you what to search?
We launched a feature called Research Insight.
Here’s what we did:
● Collected data from over 4 million products (SKU level)
● Aggregated over 1.5 million active suppliers
● Monitored real-time trends across platforms like TikTok, Instagram, X, Amazon, and AliExpress
Then we built a trend prediction engine using natural language processing, category clustering, and time-series analysis.
When we spot a spike in interest across search, engagement, video reach, and purchases,we can flag a product category as a rising trend—before users even know to search for it.

We don’t just help users find products. We help them make better decisions.

When we embedded Research Insight into our platform Accio, the impact was immediate.
● One customer sourced a smart home gadget we flagged. Three months later, it ranked top 5 in their market.
● A seller found a new beauty device brand we surfaced. It sold out on day one.
● Some suppliers even started using the data in reverse—checking trends before deciding what to produce.
We’re not just building a bigger platform. We’re building one that helps you think.

So why did we grow so fast?

Because users spread the word.
When a procurement tool helps you catch trends, make smart decisions, and act faster,it becomes more than a tool—it becomes your edge.
Many small businesses, cross-border sellers, and distributors now rely on our platform for product insight.
No ads. Just results.

The future of B2B platforms isn’t about offering more. It’s about understanding better.

All the features in the world won’t help if you don’t solve the real problem.
We believe a product becomes valuable when it solves something users truly struggle with—even if they don’t know how to describe it yet.

If you’re building a B2B product—or if you’ve ever worried about helping users make decisions—you’re not alone.
We’re still improving what we’ve built, and there’s a long way to go.But if you’ve got ideas, feedback, or just want to talk shop, I’d love to connect.
I’m Donna, and I didn’t want to build “just another B2B platform.”We looked at the problem from a different angle—and the users came on their own.

on July 30, 2025
  1. 1

    Turning trend prediction into a user advantage is genius...

    1. 1

      Hey, thanks for the kind words. I’m glad that phrase landed well. With Research Insight, we aim to help users spot emerging trends early, and it’s nice to know that message came through. Really appreciate you taking a moment to share your thoughts.

  2. 1

    what is the url of platform and any fee associated with it or not

    1. 1

      Hello! Thanks for your comment and interest! Our platform URL is: https://www.accio.com/
      Currently, the platform is free to use—no fees at all. If you have any other questions or need assistance with anything, feel free to let us know. Happy to help!

    1. 1

      Hey there! Nice to hear from you!

  3. 1

    Works great. But is Accio in essence an AI-powered search layer on top of Alibaba's massive product database?

    1. 1

      You could say Accio sits on top of Alibaba's huge product base, tons of items, millions of suppliers. But it's not just a search tool. It's a chatty, AI-powered sourcing assistant. You just say what you're looking for in plain language (or even upload a pic), and it'll help you figure out what to buy and who to buy from. It also gives you real-time trend insights, smart sorting, and supplier info. Basically it helps you go from 'finding stuff' to 'knowing what's smart to buy and how.

      1. 1

        Thanks for your reply.

    2. 1

      This comment was deleted a year ago

  4. 1

    A Question. What do you mean when you say No Ads? I mean is it a buzz word because on the internet no matter how much great product you build, how big the problem you solve, what your offer is, if you not present it infront of your audience unless you show up infront of them through ads or do cold out reach you will not get any interaction...

    1. 1

      Great question. Yeah, I totally get what you're saying. It's true: even the best product won't grow if nobody sees it.

      When I said "No Ads", I meant we didn't spend on any paid campaigns—no Google ads, no Meta, nothing like that. But we didn't just sit and hope, either.

      What we did do:

      Built something people actually wanted to share. When a tool helps someone spot a trend early, they'll usually tell their team or network.

      Spent time in real communities—industry forums, private groups, even Reddit sometimes. Not selling, just being helpful.

      Got a lot of referrals through word of mouth, especially between buyers and sourcing teams. Procurement's more connected than people think.

      And yeah—we also set up some mutual partnerships with other platforms in related spaces (like logistics tools, niche marketplaces, research tools). Nothing paid, just "hey, let’s help each other’s users."

      So yeah, we didn't do ads. But we did work hard to show up where it mattered. Visibility still matters—we just took a different route.

      1. 1

        Thanks It will help. I'm at the MVP stage, testing my service but the problem is how can I get inside private groups, niche communities, and share what I have made. Because It's hard to grab people attention, they not come and give feedbacks. Could you help me on this in a bit detail?

        1. 1

          Yeah, I get you — breaking into private or niche groups at MVP stage is tricky.
          One thing to watch out for first: if you come across as “the guy selling something,” you’ll get muted or kicked fast. So go in as a contributor, not a promoter.
          What worked for me:

          1. Know your target — list exactly who you want feedback from and where they hang out.
          2. Bring value first — make a short, useful resource for that community and share it with the admin, no hard pitch.
          3. Leverage intros — if you know someone inside, ask for an intro and offer something in return.
          4. Start small — 50–200 member invite-only groups are easier to stand out in than giant public ones.
          5. Feedback-for-value — once inside, DM a few members personally, offer early access in exchange for their thoughts.
            And golden rule: don’t drop your link on day one. Spend a week or two engaging, answering questions, and showing you get their problems. Then your MVP will feel like a natural next step, not an ad.
  5. 1

    This is one of the sharpest takes I've seen on the current state of B2B platforms. You're not competing on features, you're competing on understanding.

    I had a couple of questions I'd love your thoughts on:

    How do you filter noise in trend data? Especially from platforms like TikTok, a lot of "hype" can be manufactured. Curious how Accio separates real signals from fake spikes.

    Do different industries need different trend logic? Fast-moving categories like beauty or home gadgets make sense, but what about slower ones like industrial goods or machinery? Do you handle those differently?

    Really loved the insight that "more" isn't the future — "smarter" is. Would love to hear your take, and open to thoughts from others too.

    1. 1

      Really good questions, happy to share how we're thinking about it, and would love to hear others' thoughts too.

      On the noise filtering: yeah, totally agree. TikTok especially is full of hype that burns out in a day. What we do is:
      First, we check across platforms. If something's only trending on TikTok but not showing up in search (like Google, Amazon), or marketplace data (AliExpress, etc.), we treat it as a maybe, not a signal.

      Second, we watch for short vs. sustained spikes. Real trends usually ramp up—people start watching, then searching, then actually buying. One-day spikes without follow-up = filtered out.

      Third, we look at actual behavior. Not just people talking about a product, but whether sellers are adjusting listings, bundling products, or creators are mentioning use cases, not just aesthetics.

      As for different industries, yeah, big difference.

      In fast-moving stuff like beauty or gadgets, influencer signals and short-term search spikes matter a lot.
      For slower stuff like machinery or B2B hardware, it's more about regulatory news, procurement directories, even patent filings. The signals are there, just different.
      Would honestly love to hear how others approach trend prediction, especially in those "slower" verticals. Let's swap notes in the thread.

      1. 1

        Thanks for sharing! Your core approach makes total sense. I think the real challenge is getting consistent, high-quality data from multiple sources - that's a huge technical and resource hurdle. Excited to see how you improve data depth and real-time updates!

        1. 1

          Thanks for the support! We’ll keep working on improving our product to give users a better experience. Really appreciate your continued interest, and feel free to reach out anytime to share thoughts or feedback.