Many students do the same thing before contacting a potential advisor: open the public disclosure page, list the number of papers, project levels, academic titles, and admissions quotas in a table, then try to find the one that looks like the best match. But only after actually enrolling do they realize that the disclosure page does not tell them about lab meeting frequency, tolerance for mistakes, equipment queues, topic assignment, or other practical details. Information is asymmetric from the start: hard metrics fill the page, while soft metrics are almost hidden in the gaps between people.
This is not a matter of students lacking search skills. In the current advisor–student mutual selection system, public information naturally favors “output volume,” while training style, mentoring pace, and graduation outcomes can only be learned through scattered rumors or face-to-face inquiries. Some scholars point out that information asymmetry is an inherent flaw of this system; a 2025 Nature survey of 3,785 doctoral students across 107 countries also shows that choosing an advisor and the lab culture are among the matters that most need careful verification before enrollment. So the real task of choosing an advisor is not to determine who is the one right person—such a promise is neither realistic nor responsible—but to clarify your own confusions and check what can be checked first.
01 There are answers on the disclosure page, but not all the answers
▲ Figure 1 | In the existing disclosure system, the output volume column is fully filled, while the training style column is largely blank
The questions the disclosure page can answer are very specific: papers in the past five years, projects led, academic titles, and admissions quotas. Anyone can check this information, and it is also the least likely to be distorted.
But they hardly answer another set of questions: lab meeting frequency, feedback cycles when you get stuck, rules for allocating equipment and funding, and the real rates of delayed graduation or changing advisors. The former determines whether you can even apply; the latter determines your graduate school experience in the coming years. Treating the former as the entire basis is like judging content by paper titles alone—there is a basis, but key information is missed.
02 Break “getting to know an advisor” into three layers
▲ Figure 2 | From publicly verifiable to in-person confirmation, the three layers narrow progressively, and the cost also rises progressively
The process of getting to know an advisor can be broken down into three layers.
The first layer is publicly verifiable. Publication and citation trajectories, projects led, changes in position, and thesis topics they have supervised can all be checked by yourself, yielding a preliminary judgment of whether your direction matches.
The second layer is corroboration from the side. Accounts from current or graduated students, public updates from the lab, and peer evaluations require proactive contact, and their quality depends on whom you ask and how you ask. MuchSci’s role at this layer is to organize accounts from different sources into a structure that allows cross-comparison: whether two people’s answers to the same type of question are consistent, and on which dimension they differ, can be seen at a glance. It does not make judgments; it only helps you align the fragments.
The third layer is in-person confirmation. How lab meetings are run and how often; how problems are handled when you get stuck; where the boundaries of task assignment lie. These questions can only be asked face to face with the advisor, and only the other person’s immediate reaction can reveal the answer.
The three layers are not substitutes for one another: the first two narrow the scope, and the third makes the judgment.
03 Manage “choosing an advisor” as a project with a deadline
▲ Figure 3 | Start laying the groundwork three months in advance; the later it gets, the less you can rely on secondhand accounts
Treat choosing an advisor as a project with a deadline. The schedule can be divided as follows:
Three months ahead: lay the groundwork. List people whose directions match, review public materials in batches, and screen out those who are clearly unsuitable.
Two months ahead: narrow down. Read the representative work from the other person’s last three years; don’t just look at titles—look at what research questions they are addressing and what methods they use.
One month ahead: verify. Contact current students and ask about specific training details: how often lab meetings are held, what first-year students generally do, and how feedback is usually given when mistakes occur.
Before and after contacting: ask questions. Make a list of questions for face-to-face communication, and use questions based on specific situations to observe the other person’s answering style.
▲ Figure 4 | The self-check list only marks “whether you know,” and does not make decisions for you
Before you start contacting anyone, it is advisable to go through the list yourself: whether the direction truly matches, whether there are student accounts that can be cross-verified, and whether the difficulty of graduating can only be asked in person. Mark the status of each piece of information—verifiable, questionable, unverifiable, lead. The purpose of the list is not to make you confident, but to mark what you still do not know.
04 Different ways of asking yield different answers
▲ Figure 5 | Closed-ended questions easily get polite answers, while specific situational questions more easily reveal the real mentoring style
Many students, when contacting current students or advisors, ask, “Is your lab busy?” or “What is the advisor like?” These questions are not impossible to ask, but the answers tend to be polite. A more effective approach is to give a specific situation: “If I can’t produce preliminary results in three months, whom would I usually talk to first?” “How is coordination usually handled when equipment is unavailable?” “What administrative tasks are first-year students also assigned?”
The advantage of specific situational questions is that they force the respondent to describe real processes rather than give an overall evaluation. What you get is not an abstract label like “the advisor is strict,” but judgeable information such as “when you get stuck, first go to a senior student, and the advisor holds an individual discussion every two weeks.”
When contacting the advisor directly, also avoid vagueness. Don’t ask, “What are your expectations for students?” You can ask, “How often does the lab usually meet, and how far does discussion go at meetings?” Don’t ask, “Are graduation requirements high?” You can ask, “What are the approximate graduation destinations of students in recent years?” Questions that point to facts rather than evaluations make it easier to observe how the other person speaks.
MuchSci’s card structure can serve as an aid here: pre-organize on the same page the possible answering styles for each question and the signals to capture. During communication, you don’t need to stare at the cards and read from them; afterward, when you review, you can compare your records to judge which information is clear and which remains speculation.
▲ Figure 6 | After key agreements are confirmed, put them into a written record
After talking with an advisor or student, write down the key agreements while it is still fresh: training pace, task boundaries, graduation requirements, and the conditions you care about. This record does not need to be sent to the other party for confirmation, but you need to have a copy yourself. Oral promises will be reinterpreted by each party after a few months; only what is written down is convenient for later verification.
05 Boundaries
① This article discusses how to clarify confusions before choosing an advisor and does not constitute a guarantee of outcomes. No matter how detailed the questions or how thorough the checks, they cannot guarantee that you will get along smoothly with a particular advisor; they can only reduce the decision-making risk caused by information blind spots.
② The list and question templates do not make judgments for you. They are only responsible for marking “what has not yet been verified” and “which questions need in-person confirmation.” Whether to choose a particular advisor must still be decided by you.
③ Operations involving contacting advisors, admissions policies, and graduation requirements must comply with the current regulations of your institution. Public information inquiries and private communication should both be conducted within the scope of compliance; the questioning methods discussed in this article do not constitute advice to circumvent any system.