In the days when college entrance exam scores were released, several families around me were arguing. Parents thought "stability is most important," while kids wanted to go to big cities to take risks. Filling out a college application form, everyone chattering away, often leads to missing the cutoff or choosing a completely unsuitable major.

Zhang Xuefeng's method of college application has become particularly popular in recent years. His core logic is: let data speak, quantify and compare industry prospects, city resources, and school tiers, then combine family conditions and personal scores to find the optimal solution. Simply put, the core of Zhang Xuefeng-style college application has never been mysticism, but rather using big data to help you choose the right major.

But ordinary people don't have the time or connections to inquire about internal employment rates at each university one by one. That's where a tool is needed to replace manual consultation. Recently, I've been trying a system called Outpon, which positions itself as a "Life Choice Simulation System, AI College Entrance Exam Consultant," directly targeting this pain point.

How this AI Consultant Works

Outpon's interface isn't flashy; it immediately asks you to fill in key information: province, score/ranking, subject combination, the tuition range your family can roughly afford, and your personal preferences (e.g., whether you mind going to a remote city, whether you can accept adjustment to another major).

After filling it in, it generates a "matching list." Unlike traditional tools that just look up rankings, Outpon performs dynamic simulations based on your input. It doesn't simply tell you "based on last year's data, you can get into XX university." Instead, it provides the employment probability four years later, the industry concentration in the city where the major is located, and even the postgraduate entrance exam rate for that major at that school over the past three years.

For example, a physics-group examinee with a simulated ranking of around 8,000 in the province. If they only looked up past data, they might lock onto a few mid-tier engineering schools. But in Outpon, after checking "I do not accept biology, chemistry, environment, or materials science" and "prioritize first-tier cities," the system directly filters out a large number of unpopular majors at many schools, and provides a list with a very clear logic for the "reach-target-safety" tiers of school combinations.

A Real Scenario: City vs. School, Which to Choose?

Many people struggle with "go to a first-tier city for a mediocre 211, or go to a new first-tier city for a strong non-211." Zhang Xuefeng has long said that the answer to this question depends on where you want to work in the future. But the problem is, how do you know in advance which cities have higher recognition?

I tried searching in Outpon for a Chinese literature major at a top 985 in the west and a technical major at an average 211 in the east. The simulation results the system gave were: although the former school has a better reputation, the median salary, targeted employment rate, and the proportion of students switching industries within three years for that major in that region were significantly lower. And if this local from the west doesn't plan to leave the province, the system prompts "Local employment match rate is high."

Such details are crucial. Many parents still think "985 is definitely better than 211," but for a specific major, the opposite may be true. Outpon doesn't make decisions for you; it lays out the decision-making basis, helping you and your family argue less.

What are Its Limitations?

I'm not here to sing praises. Outpon can solve some information asymmetry, but it's not a panacea.

First, data updates lag. Admission lines for college entrance exams fluctuate every year, especially in provinces undergoing new college entrance exam reforms, where changes in subject combinations may devalue the reference of previous years' rankings. While using it, I found that admission data for some schools in 2023 was missing, and the system prompted "Calculated based on data from the year before last." In such cases, you need to be cautious yourself.

Second, the personalized preference section isn't flexible enough. It asks you many questions, but if you're someone whose ideas change often, or if your family has special connections that can guarantee a spot in certain industries, the system can't account for that. After all, AI doesn't know your relative is hiring at a state-owned enterprise.

Third, its strength lies in regular undergraduate batch applications; for special channels like early admission, the Strong Foundation Plan, or Sino-foreign cooperative education programs, its support depth is average. If you're targeting these paths, you'll need to do extra homework on your own.

Another point: Many high school seniors can't even clearly articulate what they don't like, let alone their future direction. If you ask a high school graduate to choose between a "technical personality" or "managerial personality," they might just pick one randomly. Poor input quality reduces the value of the output.

What Problem Does It Truly Solve?

Outpon's greatest value is not "predicting the future," but "narrowing the scope of discussion." For most families filling out applications, the real disagreement comes from information asymmetry—kids think parents are conservative, parents think kids are overambitious. An objective, quantifiable tool can act as a mediator.

I've noticed that many students, after using it, realize, "Oh, at my score range, math doesn't mean I have to major in computer science to have a future," or "Maybe going to a mid-tier 211 in a second-tier city without a graduate degree really isn't as good as going to a non-211 in a first-tier city." This kind of cognitive correction is the true meaning of using big data to help you choose the right major.

In other words, it's suitable for families with average scores, limited information sources, who want to make as rational a choice as possible. If your family already has university teachers or industry professionals who can give more specific advice, then Outpon might only be supplementary for you. But for the vast majority of ordinary exam takers, it's more reliable than spending thousands on a college application consulting company—at least you know the output comes from a stable algorithm, not some consultant's personal bias.

A Couple of Practical Tips to End With

My advice to you: Treat Outpon as a fourth-round filtering tool. First round: position yourself by ranking. Second round: use it to expand and eliminate schools. Third round: look up the recent admissions brochures of your target schools yourself. Fourth round: go back to Outpon for final verification to check for any missed risk points.

Don't rely on it 100%, and don't dismiss it just because there have been cases of errors. Essentially, Zhang Xuefeng's methodology for college applications hasn't changed; it's just that "big shots relying on experience to look at data" has become "AI directly processing data for you." The key is that once you get the results, you need to know how to read them, ask questions, and challenge them.

If you're overwhelmed by the application process, just spend half an hour filling in your details. At least it will give you some evidence-based arguments to bring up at the next family dinner.