The "Impossible Triangle" of Vendor Selection

*This article is reprinted in full from the WeChat official account "Supply Chain Management Practitioners".*

Economics has an "impossible triangle": free capital flow, a fixed exchange rate, and independent monetary policy cannot all be achieved at once. The theory was proposed by Robert Mundell, the father of the euro. The Nobel-winning theory behind it may be hard for ordinary people to digest, but the "impossible triangle" itself is something none of us finds the least bit unfamiliar.

Take a girl looking for a partner, for instance: the dream is a man who is handsome, wealthy, and devoted. Men with all three do exist, but they are exceedingly rare. In the vast majority of cases, the best you can hope for is two out of three. The handsome, wealthy young man usually faces too many temptations and tends to stray; the devoted, wealthy one is often not much to look at; and the devoted, handsome one is, as a rule, flat broke.

Or take the "iron triangle" of project management: once schedule, cost, and scope have been optimized to a certain point, no further improvement is possible—improving one of them must come at the expense of the rest. To go faster, you must cut scope or spend more; to save money, you must go slower or cut scope; to add scope, you must pay more. In job hunting we want high pay, light work, and a short commute; in sourcing suppliers we want high quality, fast delivery, and low cost. In the end we all run up against the same reality—you can't have your cake and eat it too. Clashing objectives multiply, and we are left with an "impossible triangle." Of course, depending on how many objectives collide, there may equally be an "impossible quadrangle" or an "impossible pentagon"; for the sake of discussion, this article refers to them all as the "impossible triangle."

Facing vendor selection under the "impossible triangle," here is a three-step decision process:

First, set hard thresholds to screen out the worst suppliers;

Second, compare the candidates against one another and eliminate those at a comprehensive disadvantage;

Finally, make strategic trade-offs and select the supplier that best fits your top priorities.

Let's work through a case to see how it plays out (the case comes from DeepSeek). A technology company needs to purchase 50 servers, with a total budget of ≤ RMB 2.1 million, a quality pass rate of ≥ 98%, a delivery cycle of ≤ 30 days, and 24/7 technical support. Six candidate suppliers are in the running, with the following metrics:

Step 1: Hard thresholds—one strike and you're out. On price, quality, delivery, and other dimensions, the buyer sets the following hard requirements: unit price ≤ RMB 42,000 (total budget ≤ RMB 2.1 million); quality pass rate ≥ 98%; delivery cycle ≤ 30 days; technical support must be provided; and no major risks (litigation, environmental penalties, or missing ISO certification). Clearly, supplier A misses the quality bar (97%), provides no technical support, and holds no ISO certification—out. B blows the delivery deadline (35 days)—out. F misses the quality bar (95%), provides no technical support, and has an environmental penalty on record—out as well. With these three suppliers with "fatal flaws" eliminated, three remain: C, D, and E.

Step 2: Eliminate the comprehensively disadvantaged. C: the second-highest unit price, plus a litigation record. E: the highest unit price (RMB 42,000)—the best quality and delivery, but a first-time partnership carries risk. D: the lowest unit price (RMB 36,000), the fastest delivery (20 days), and no risk. Among the three, C is behind D on both price and delivery and carries a litigation record besides—it is comprehensively disadvantaged and out. That leaves D and E, where improving either one's objectives means sacrificing the others. In negotiations, for example, getting D to raise its quality means a higher price; getting E to cut its price means accepting lower quality standards or thinner after-sales service.

Step 3: Strategic trade-offs and the final decision. Whether through hard thresholds or the elimination of the comprehensively disadvantaged, what we have been doing is squeezing out the "slack," and the remaining D and E now stand as the "impossible triangle." Between D (RMB 36,000, no risk) and E (RMB 42,000, high quality but higher risk), the call depends on the buyer's priorities. If cost and risk control weigh more, choose D; if quality and delivery speed weigh more, choose E (the potential risks of a first-time partnership can be hedged with tools such as advance-payment guarantees and installment payments). Incidentally, some may say E's quality (99%) is only one percentage point above D's (98%)—hardly much better. For high-stakes products like servers, though, quality cannot be read that way; read it the other way around: supplier D's servers are down 2% of the time (a quality problem)—twice E's rate. That is a world of difference.

Moved by a fit of poetic inspiration, DeepSeek improvised a quatrain to sum up this three-step elimination: "One-vote vetoes cut out the fatal flaws; the all-round weak are shown the door. The final pick comes down to two or three—saving time, dodging pitfalls, losing your way no more." Ha! The first three lines are spot-on; the fourth, well, has a certain doggerel flavor to it.

The verse may be a bit doggerel, but the theory behind this methodology is anything but humble—it is called "Pareto optimality." Its core idea: once multi-objective optimization has been pushed to its limit, improving any one objective requires sacrificing the rest. That is the "impossible triangle"—there is no "wanting it all and getting it all," only strategic trade-offs and deliberate priorities. Facing the "impossible triangle," our task is not to hunt for the "perfect supplier" (there isn't one), but to rank our priorities among the conflicting objectives and pick the most suitable candidate. In this example, both D and E are optimal solutions; which one to choose depends on whether the buyer's first goal is "saving money" or "safeguarding quality."

On the "impossible triangle," a few points I would add: 1. Hard thresholds are the bottom line—they keep out the "bad money" and prevent the worst choices. 2. The optimal solution is not perfect, nor is it unique; it is the least-bad option after every effort at optimization, and there can be more than one. 3. Vendor selection is not a one-time, either-or call; it should flex with dynamic needs. In a shortage, supply security comes first—choose the supplier with timely delivery even at a higher cost; in a glut, cost comes first—choose the cheapest.

Finally, failing to find a perfect supplier under multiple conflicting objectives does not mean we resign ourselves to fate and give up on "having it all"; it means we must close the supplier's gaps through follow-up management. That is what "selection plus management" of suppliers means. It is just like the girl looking for a partner: land a wealthy, handsome man, and you had better watch him closely to guard against straying; land a handsome, devoted one, and you had better work hard alongside him to make money. Girls all know that selection without management leaves those shortcomings free to make trouble for you. And in procurement, "selection without management" of suppliers is the biggest problem I have seen in my two or three decades in the field—bar none. It guarantees downstream performance problems with suppliers, and more than half of my "red book" is devoted to tackling exactly this issue: *Procurement and Supply Chain Management: A Practitioner's Perspective*.