Most reference checks end the same way. You call a former manager, they say the candidate was "great to work with," you thank them, and you write "references positive" in the ATS. Twenty minutes spent, nothing learned.
That is not because references are useless. It is because the candidate picked the people who would say nice things, and you asked questions that made it easy to say them. Fix the questions, and the reference call becomes one of the cheapest sources of real signal you have.
Why the standard reference call fails
Think about the incentives on the other end of the line. The reference agreed to the call because they like the candidate. They are busy, they do not want to hurt anyone, and there is no upside for them in being critical. Every open-ended question ("How was it working with them?") invites the safest possible answer.
Add to that the questions themselves. "Would you hire them again?" is a yes or no question where no is socially unacceptable. "What are their weaknesses?" gets you "they sometimes care too much." These questions are not designed to surface information; they are designed to be easy to ask.
The fix is not to interrogate. It is to ask questions that are specific enough that a warm reference can answer them without feeling like they are betraying anyone.
Question 1: "Where did they rank on the team you managed?"
Ask the manager how many people they had on the team and where this person fell. Top third, middle, bottom third? Most references will answer this because it does not require saying anything negative. "Middle of a strong team" is a comfortable sentence to say, and it tells you a lot.
Listen for hesitation. A manager who instantly says "top two, easily" is giving you a different signal than one who pauses and lands on "solidly in the middle." Follow up with "what separated the top performers from them?" You will usually get a specific gap.
Question 2: "What did you have to coach them on most often?"
This works better than asking about weaknesses because coaching is a normal part of managing. Every manager coaches every employee on something. Framing it that way removes the sense that the reference is exposing a flaw.
The answers tend to be specific and useful: "I kept reminding them to loop in the design team earlier," or "we worked on how they handled pushback in meetings." Compare what you hear to the gaps you noticed in the interview. When they line up, you have confirmation. When the reference names something the interviews missed, that is the whole point of the call.
Question 3: "Describe a time they disagreed with you. What happened?"
This tells you about judgment and how the person operates under a manager they may not agree with. A reference who cannot think of a single instance is telling you the candidate either never pushed back or the manager does not remember them well. Both are worth knowing.
A good answer includes how the disagreement was raised, whether the candidate changed their mind when presented with new information, and whether they committed to the decision afterward. You are not looking for someone who always won the argument. You are looking for someone who could have it professionally.
Question 4: "If they came back to you in two years, what role would you want them in?"
This is a more revealing version of "would you rehire them." A manager who says "I would want them running the team" is telling you they see growth. One who says "the same role, they were good at it" is telling you something more measured. One who deflects to "it would depend on what we needed" is telling you the answer is probably no.
The question also surfaces how the reference thinks about the candidate's trajectory, which matters if you are hiring for a role that is a step up from what they did before.
Question 5: "What would the person who liked them least say?"
This is the question most recruiters are afraid to ask, and it is consistently the most productive one. Every reasonably experienced employee has had friction with someone. The reference knows who that was. By asking about a hypothetical third person, you let the reference report the criticism without owning it.
The answer is usually measured and fair: "The finance team found them impatient about approvals," or "some of the junior engineers thought they were too blunt in code review." That is real information about how the candidate shows up when things are not going their way.
Make the call worth the reference's time
A few practical habits improve the quality of every answer above. Send the questions ahead of time so the reference can think, rather than improvising on the spot. Open the call by telling them what the role is and what you are worried about, so they can be relevant instead of generic. And keep the call to fifteen minutes; references get more candid when they are not watching the clock.
Take notes in the reference's words, not your summary. "Middle of a strong team, coached on stakeholder timing, wanted them back as a lead" is far more useful to your hiring committee than "positive reference."
Connect it back to the interview data
Reference calls are most valuable when they are the last input, not a separate one. Go into the call with your interview scorecards open. You already have a list of things you are confident about and things you are unsure about. Point the questions at the uncertain items.
If your interviews flagged concerns about how a candidate handles ambiguity, ask the reference for a specific example of them working without a clear brief. If the panel was split on communication, ask about the disagreement question and listen for how the reference describes their delivery. The reference should confirm or challenge what you already think, not start the evaluation over.
Done this way, a reference call stops being a box to tick after the decision is made and becomes the final piece of evidence before it. That is a much better use of twenty minutes.