The Fairness Algorithm That Should Be in Every Cross-Border Seller’s Playbook
If you manage inventory across three Amazon marketplaces, a Shopify store, and a fledgling TikTok Shop, you know the pain. Every time you decide where to launch a new SKU or shift ad spend, there’s an anonymous voice in your head that says: “We need the cheapest route.” So you optimize for total cost, pick the marketplace with the lowest FBA fees, the cheapest shipping lanes, the smallest tax burden. And then someone — maybe your co-founder in Berlin, maybe your logistics partner — ends up carrying all the hidden weight: longer lead times, higher return rates, customs delays that kill cash flow. The group (your business) got the cheapest ticket, but one person’s seat took three connections and eleven hours.
This is exactly the problem that a tiny travel tool called MeetPoint set out to solve for friend groups planning a trip. Its “Fairest” mode doesn’t just minimize the group’s total flight spend; it scores each destination based on who gets the worst deal, factoring in price, flight duration, and layovers. A city that’s $50 cheaper for one person but forces them onto an 11-hour, three-stop nightmare is penalized. The result: a decision that feels equitable, not just cheap. Most cross-border sellers I know are making allocation decisions with the same flawed “cheapest total cost” mindset. MeetPoint’s core insight — that minimizing disparity often matters more than minimizing average — is a lesson worth stealing, not for your next vacation, but for your next inventory drop.
The Problem That Isn’t Yours — Until It Is
I’ll be blunt: when I first saw MeetPoint’s Product Hunt page, I dismissed it as a nice consumer utility. Another “where should we meet?” app for a world that already has Google Flights and Skyscanner. The maker, Florens von Buchwaldt, explained the backstory simply: his friend group kept going back and forth on WhatsApp, manually checking flights for every possible city, and he needed a tool to answer the question directly. That’s relatable, but not obviously relevant to e-commerce.
Then I read the comment thread. A user asked about the “Fairest” mode, and Florens replied with a formula: flight price + $15 for every hour in the air + $25 per layover. The destination is ranked by whoever gets the worst burden, and the algorithm minimizes that worst case. Suddenly, I saw my own spreadsheet — the one where I compare landing costs across Amazon.co.uk, Amazon.de, and Amazon.com for the same product. I’d been optimizing for average net margin. But my UK operation was swinging with 5-day delivery times and zero returns, while my German warehouse was eating two-week customs holds and a 15% return rate. The average looked fine, but the German warehouse was the friend spending 11 hours on three connections.
Cross-border sellers face this pattern constantly. You have multiple “destinations” (marketplaces, logistics routes, fulfillment centers). Each destination has a “burden” for your business: not just cost, but time, complexity, risk, and opportunity cost. The typical optimization approach — pick the cheapest combined cost — systematically overlooks the player who gets a raw deal. For a business, that “player” might be your cash flow, your customer satisfaction in a specific region, or your sanity as you manage returns. MeetPoint’s lesson is that you need a fairness score for your multi-channel strategy, not just a cost per unit.
How MeetPoint Differs from the Incumbents
I want to name the incumbents in group travel planning: Google Flights, Skyscanner, Kayak. All of them are fantastic for one person. You put in your origin and destination, and they show you the cheapest option. Even when you search for a group, they still optimize per person. No tool asks “What city minimizes the worst travel burden among the four of us?” That’s a fundamentally different optimization problem.
The same gap exists in e-commerce tooling. The dominant category is “product research” and “keyword optimization” — Helium 10 or Jungle Scout will tell you which product to launch on which marketplace based on demand and competition. But they treat each marketplace as a standalone opportunity. They don’t model the global burden of splitting your inventory across markets. If you launch on Amazon.com and Amazon.de simultaneously, Helium 10’s “Profit Calculator” will show you two separate numbers. It won’t warn you that the German account is about to eat a 20% VAT registration cost plus a 10% translation/localization overhead that the US account doesn’t have, and that your total cash outlay will be 30% higher than the combined profit projections suggest.
MeetPoint’s approach is closer to what a multivariate optimization model does: balance trade-offs across multiple dimensions, with a constraint that no single party bears an undue burden. For a cross-border seller, that means weighting not just unit economics, but also time-to-market, compliance risk, logistics complexity, and local customer service capacity. The incumbents don’t do that. MeetPoint’s “Fairest” mode is a prototype for a decision framework that should exist in a tool like Sellics or Pacvue — but doesn’t yet.
Where the Math Breaks
Let’s get specific about MeetPoint’s math because it’s where the biggest lesson hides. The maker updated “Fairest” after feedback: originally, it only looked at price. Users (rightly) pointed out that the person paying least might be the one doing three connections and eleven hours. So Florens added a burden formula: price + $15/hour in the air + $25 per layover. Then the algorithm evaluates each destination by the maximum burden among all travelers, and it tries to minimize that maximum.
This is a classic “minimax” optimization. It’s powerful because it prevents one extreme. In travel, it stops the group from choosing a city that’s super cheap for three people but brutal for the fourth. In e-commerce, the analogous scenario is choosing which marketplace to prioritize for a new product launch. Suppose you sell in the US and the UK. Launching on Amazon.com might cost you $1,000 in initial fees and shipping, while Amazon.co.uk might cost $1,200 because of higher VAT compliance and translation. A total-cost optimizer picks Amazon.com. But what if your UK business is already underperforming, and the additional burden of launching a new product there without support would stretch your UK operations to the breaking point? The “minimax” approach would say: don’t launch anywhere until you can support both markets at a manageable burden level. Or it might suggest a different product for the UK that fits the existing logistics setup.
The math breaks, however, when you try to set the weights. $15 per hour in the air? $25 per layover? Those are arbitrary. For your e-commerce burden model, you’d need to define your own weights: maybe $50 per day of customs delay, $200 per compliance error, 10% of margin for returns risk. That’s not trivial. But the insight remains: the optimization should flatten the worst-case burden, not just average cost.
What Cross-Border Sellers Can Borrow
You don’t need to build a new tool. You need to change how you evaluate channel decisions. Here’s a practical framework borrowed directly from MeetPoint’s “Fairest” mode:
Identify your “travelers.” These are not people; they are your business units, marketplaces, or even individual SKUs. If you sell on Amazon US, Amazon EU, eBay, and your own Shopify store, each is a “traveler” with its own cost structure and pain points.
Score each destination (decision option) by burden for each traveler. Don’t just look at cost. Include time (lead time from factory to warehouse), complexity (number of customs steps, translation needs), and risk (returns rate, fraud rate, account suspension risk). Assign a dollar value or a weight to each dimension, just as MeetPoint assigned $15/hour and $25/layover.
Compute the “worst burden” for each destination. For example, if you’re deciding whether to expand into Japan, your US marketplace might have a burden score of $5,000 (low risk, quick shipping) while your Japan entry has a burden of $12,000 (high translation costs, longer customs, higher returns). The worst burden for Japan is $12,000. For a new SKU launch in the US alone, the worst burden might be only $4,000. The minimax optimization says: choose the option where the maximum burden is minimized. In this case, the US-only launch wins — not because it’s cheaper total, but because it doesn’t overload any single unit.
Apply this to your tooling stack. Use existing data from Helium 10 or Jungle Scout for cost estimates, but overlay your own burden model in a spreadsheet. If you’re using a logistics aggregator like ShipBob or Flexport, pull their performance data on delivery times and returns to inform your burden scores.
This framework is not perfect. It’s subjective. But it’s infinitely better than the default habit of “cheapest total cost,” which ignores that one channel is bleeding while another is fine. I’ve seen sellers drain their German Amazon account by overloading it with SKUs that have high return rates, while the UK account sat idle. MeetPoint’s “Fairest” mode would have flagged that imbalance.
Why Amazon Sellers Should Care More Than Shopify Ones
I’ll make a blunt distinction: Shopify store owners have more flexibility to adjust shipping methods, suppliers, and destinations. A Shopify seller can turn off a product listing in a high-cost region with one click, or run a local campaign to a specific country using Klaviyo segments. The burden per transaction is more divisible. Amazon sellers, on the other hand, are locked into marketplace-level fixed costs: FBA fees per marketplace, VAT registration (often required if you store inventory there), translation, and advertising budgets that need to be allocated per country. The “traveler” on Amazon is the entire marketplace account, and the burden of launching a new SKU is not marginal — it’s incremental with non-linear costs. A single bad marketplace decision can tie up inventory for months, generate storage fees, and tank your account performance metrics. For an Amazon seller, the minimax approach is not a nice-to-have; it’s a survival tactic.
I’d argue that if you run a multi-marketplace Amazon operation, you should be using a weighted decision matrix every time you allocate inventory to a new country. Pull data from Amazon Seller Central on cost of goods sold, storage fees, returns rates, and inbound shipping times. Then apply a burden factor for compliance risk (e.g., EU CE marking, UKCA marking). The “fairest” allocation is the one that doesn’t make any single marketplace carry 60% of your total risk. MeetPoint’s algorithm taught me that it’s better to have all three marketplaces at 30% risk each than to have one at 5% and another at 85%.
Where My Judgment Says It Falls Short
MeetPoint is a great idea, but it’s early. The maker admits the tool currently supports only 4 people and uses a single API — the Tequila API from Kiwi. For a travel tool, that’s limiting. For a cross-border seller trying to adapt the concept, the limitations are even more glaring.
First, the “fairness” score is static. $15/hour in the air and $25 per layover are generic placeholders. For your business, the real weights vary by market: a customs delay in Germany costs more than in the Netherlands because of different VAT rates. A return in Japan costs 30% of the item value just for restocking, while in the US it’s 15%. You need a system that lets you define your own weights per “traveler” — and that requires data integration. MeetPoint is not built for that.
Second, the minimax approach can lead to under-optimization. If you have a market that is inherently high-burden but high-margin (say, selling premium electronics in Japan), minimizing the worst burden might cause you to avoid that market entirely, leaving money on the table. The trick is to balance minimax with a total profit floor. MeetPoint doesn’t have a toggle for “cheapest but within a fairness threshold.” That’s a feature I’d want in an e-commerce version.
Third, scaling. MeetPoint’s algorithm works for 4 people. For 300 SKUs across 10 marketplaces, the number of permutations explodes. You can’t run this in a spreadsheet; you need a proper optimization engine. Tools like Oracle’s Network Optimization or even a custom Python script using linear programming would be necessary. The concept is borrowable, but the implementation is not turnkey.
Finally, there’s a behavioral challenge. In a friend group, everyone has a voice. In a business, the “traveler” with the worst burden might be a warehouse manager who doesn’t speak up. Equitable allocation requires visibility across all silos, which many cross-border sellers lack. MeetPoint’s idea assumes you can quantify each person’s burden. In e-commerce, you often don’t have the data.
What I’d Watch / Test Next
I’m not going to tell you to abandon your tool stack and build a custom algorithm this week. But here are three concrete next steps you can take — no coding required.
1. Build a one-sheet burden matrix for your top three marketplaces. For each marketplace (Amazon US, UK, DE, etc.), list five dimensions: total landed cost per unit, average delivery time to customer, return rate, customs/tax complexity (rank 1-5), and cash flow impact (how quickly you get paid). Assign a dollar weight to each dimension based on your historical data. Then, for a new product launch, calculate the “worst burden” for allocating to each marketplace. If Amazon DE’s worst burden is $15,000 (due to high return rate + slow customs) and Amazon US’s is $8,000, you know not to split inventory equally. Focus on the US first, or reduce the DE allocation until you can improve its burden.
2. Audit your current inventory allocation. Pull a report from Amazon Seller Central or your ERP showing units sold per marketplace, returns per marketplace, and average days to delivery. Use the burden framework to see which marketplace is “carrying the worst deal.” If one marketplace has 20% of your sales but 50% of your returns and 40% of your delayed deliveries, that’s the friend doing three connections. Either fix that marketplace (better supplier, different carrier) or reduce your allocation to it. Don’t let the total cost average deceive you.
3. Subscribe to meetpoint-2 on Product Hunt and follow Florens’ updates. He’s iterating fast—he already added a region constraint feature based on user feedback. If he decides to open up the API or add multi-parameter weighting, that tool could become a lightweight template for burden optimization. More importantly, watch how he handles user requests for “custom fairness factors.” That’s the exact feature that would make his framework directly applicable to cross-border logistics decisions.
The takeaway is not that MeetPoint is your new SaaS tool. It’s that the most dangerous metric in cross-border operations is the cheapest average cost. It hides the bleeding. The next time you’re in a WhatsApp thread — or a Slack channel — debating where to launch next, ask: “Who’s going to carry the worst burden?” Not just “What’s the cheapest total?” That question alone will change how you allocate.
