Consulting Should Be Explainable

A Market Where Nobody Asks Why

Last fall, after a consultation, a parent pulled me aside and asked, “The other places just told us not to apply to this school and never said why. Can you tell me why?” I couldn’t give a clean answer on the spot. It embarrasses me to admit it, but back then we were still running half on instinct ourselves.

There are structural reasons admissions consulting stays opaque. Results get verified only once a year. Samples are always small. When a student gets in, the consulting takes the credit; when a student is rejected, there is room left to blame the student. So you can answer the question “why?” with “that’s what my experience says” and still do good business, because parents have no data in hand to compare. It is a market where the information asymmetry itself is the product, and the structure is lopsided in the seller’s favor.

Are we a company that is somehow free of that temptation? I still can’t promise that.

Every Piece of Advice Carries Its Evidence

So as we built Epic Prep, we narrowed our principles down to one: every piece of advice comes with its evidence. When we propose a school list, the conditions behind that school’s actual admits come attached. Score bands, courses completed, the grain of the student’s activities. If we want to say “this school is a reach,” we have to show, in numbers, what happened to students with a similar profile. In a recent consultation, we laid out four cases side by side whose conditions resembled the student’s. In the same score band, two got in and two did not, and the difference came down to course selection. That day I watched the parent’s question shift from “which schools do we apply to?” to “what more should we be preparing?” It was the moment evidence changed the direction of the conversation.

To get there, we spent the entire second half of last year organizing the score data and admissions cases scattered across ELA, sorting them by condition. Cleaning up a single case until it could stand as evidence for advice took forty minutes on average, and the cases we have built that way still fall short of three digits. I know how modest that number is. We judged it better than “that’s what my experience says” all the same.

Evidence Is Not Free

The cost showed up immediately. Preparing one consultation began taking two and a half hours on average. Before the principle, thirty minutes was plenty. I still remember what Jamie said in the first week of pilot consultations: “The consultations are backing up while I dig for evidence. On instinct I would have seen three families today. I saw one.”

We have lost things, too. Last month we withheld our judgment on a particular school because we had only two cases to draw on, and that parent left with the words “the other place gave me an answer on the spot” and never came back. Put a shop that sells fast certainty next to a shop that sells slow evidence, and the fast one wins the short game. That is simply a fact we have to accept. That week I sat down with the team and ran the numbers again. If two parents out of ten walk away over this principle, will we go that much further with the remaining eight? For now, the only answer we have is closer to faith than calculation.

Attach evidence, and consulting slows down. It gets harder. Sometimes we have to say “we don’t know yet.” All three are bad for revenue.

I Still Think the Math Works Out

Run the numbers to the end and it looks like this. When advice made on instinct turns out wrong, you never learn what was wrong. When advice with evidence attached turns out wrong, it becomes an asset. A record remains of the conditions under which the prediction missed, and the next cycle’s advice gets that much more accurate. Instinct accumulates in individuals and walks out the door at the end of the day; evidence accumulates in the company and stays.

We did the plain arithmetic, too. If preparation takes five times as long, the same team can handle one fifth as many consultations. There are only two ways to close that gap: charge more per consultation, or spend less time finding the evidence. We bet on the latter. The more cases we organize, the shorter the next consultation’s prep becomes, and the two and a half hours of that first month has now come down under two. Whether it is coming down fast enough is something we will have to keep watching this year.

Still, whether this math is right will only be proven after a few admissions cycles. If this year’s results are bad, this post will make for a fairly embarrassing record. I am leaving it here knowing that possibility. Because I believe that even one more parent asking “how do you know that?” is better, for us and for this market.