How to Plan a Casting Session (Step-by-Step, 2026)

Key Takeaways

• An audition is a test, and a test needs a control. Same sides, same reader, same first direction, or the results are not comparable.
• The redirect is the real measure. How they change is more predictive than how they arrive.
• Record to one standard. Comparable footage is the whole point, and inconsistent framing or audio destroys it.
• Schedule for your own judgment, not their convenience. After about four hours you are no longer evaluating, you are pattern-matching on fatigue.
• How you run the room is a hiring decision too. Actors talk to each other, and a badly run session costs you access to the next pool.

The Controlled Read: Designing a Session That Actually Compares

A casting session produces a decision. The quality of that decision depends entirely on whether the information going into it is comparable. That is a design problem, and it has a simple structure.

Hold three things constant

The sides. Every actor reading for a role gets the same one or two scenes, sent the same amount of time in advance. Different sides mean different difficulty, and difficulty is not a quality of the actor.

The reader. The same person reads the other part for everyone. A reader who is warm with one actor and flat with the next has changed the experiment. Use a consistent, neutral, competent reader, and never let the director read opposite unless every actor gets the same.

The first direction. Whatever you say before the first read, say it to everyone in roughly the same words. "Play it like she already knows" given to three of six actors is not a note, it is an unfair advantage.

Change exactly one thing

The redirect. After the first read, give a substantive adjustment and have them do it again. Not "a bit more energy," which is not actionable. Something that changes the interpretation: play it as if you are trying not to cry. Play it as if you have already decided to leave. Play it to make him laugh.

The redirect is the test. The first read shows you their preparation, their instinct, and their nerve on a strange morning in an unfamiliar room. The redirect shows you three things you cannot get any other way: whether they can hear a note, whether they can execute it without discarding what worked, and how they behave in the moment of being corrected.

The first read tells you how they prepared. The redirect tells you what they will be like on day three.

That is the whole framework. Three constants, one variable, and the variable is the thing you actually score.

Before the Session

Write a real character breakdown. Not a physical description. What the character wants, what they are hiding, what they do under pressure, and one line about the register of the performance. Actors read breakdowns carefully and a vague one produces vague submissions.

Choose sides that test the hard thing. Pick the scene that contains the difficulty of the role, not the showiest one. If the part requires stillness, do not audition with the shouting scene, because everyone can shout.

Send sides in advance, with enough time. Twenty-four to forty-eight hours is a reasonable standard. Sides sent an hour before test cold reading, which is a different skill and probably not the one you need.

Use self-tapes for the first pass. They are now standard, they widen your pool geographically, and they let you filter without occupying a room. Specify framing, background, slate, and file naming in the request, because inconsistent self-tapes are as uncomparable as inconsistent in-person reads.

Decide your scoring before you see anyone. Three or four criteria, written down. Something like: right for the role, takes direction, listens in the scene, and would I want them on set for four days. Written criteria are what stop you from casting the most confident person in the room.

The Self-Tape Request: What to Specify

Self-tapes are the first pass on most productions now, and an unspecified request produces fifty uncomparable files. Specify all of it, in the request itself.

The two lines that do the most work are the second take and the audio standard. Asking for one take as written and one take their own way gives you the remote equivalent of the redirect, which is otherwise the thing self-tapes cannot test. And an explicit audio instruction prevents the systematic bias toward actors who happen to have shot in a quiet room.

The Room: Who Is There and What They Do

The director. Watches, gives the redirect, and mostly stays quiet. Directors who talk a lot in auditions learn less.

The reader. Consistent for every actor, reads competently and neutrally, does not perform.

The camera operator, or a locked-off camera. Someone has to be responsible for the recording being usable.

A session runner. Greets people, keeps time, manages the waiting area, and makes sure nobody sits in a corridor for forty minutes wondering whether they have been forgotten. On a small production this is one person and it matters more than it sounds.

Not everybody else. A crowded room changes the performance and tells you nothing about how the actor works on a quiet set. Keep it to the people with a job.

Running the Session

Block the day. Ten to fifteen minutes per actor for a first session, with a real gap every hour and a hard stop after about four hours. Judgment decays and the actors seen at hour five are being compared against a tireder version of you.

Slate every take consistently. Name, role, agent if applicable, on camera, at the top. This is what makes the footage navigable a week later.

Give everyone the same welcome. A short, identical explanation of what is going to happen. It settles nerves and it keeps conditions constant.

Read once, redirect, read again. That is the core unit. Anything beyond it is usually you enjoying yourself rather than gathering information.

Write your note immediately, before the next actor. One or two lines against your written criteria. Notes written at the end of the day are memory, not observation.

Do not give feedback on their performance. It is not a workshop, and it puts the actor in the position of responding to criticism from someone who has not hired them.

End on time, and tell them what happens next. Even "we will be in touch within two weeks either way" is worth more than silence, and then honour it.

The Recording Standard

Comparable footage is the point, and it is easy to lose.

Same framing for everyone. A medium close-up is the standard for a reason: it holds the face and enough body to read posture.

Same background and lighting. A different wall behind actor four is a variable you introduced.

Audio you can actually hear. A room mic across a large space is the most common failure. If you cannot hear the quiet choices, you will systematically favour the loud performers.

Consistent file naming. Role, actor name, date. You will be sorting fifty files.

Do not rely on your memory of the room. The reason to record is that the room is misleading. Presence in person and presence on camera are different qualities, and the camera is what your audience will see.

Callbacks and Chemistry Reads

Callbacks are for a different question. The first session asks "can this person play the role." The callback asks "is this the version of the role I want, and does it fit with the rest of the cast."

Chemistry reads must be paired systematically. If you have three candidates for one role and two for another, read the combinations that are actually plausible rather than whichever pairing is convenient. Chemistry is a property of the pair, not of either person, and it is the most common reason a strong individual audition does not survive contact with the other actor.

Bring more of the script. A callback is where a longer scene, or the difficult scene you held back, earns its place.

Decide who is in the room. Producers and, if the project has one, the casting director. Still not everybody.

Casting Non-Actors and Documentary Contributors

Almost nothing written about casting addresses this, and it is a large share of real production work.

You are not testing performance, you are testing presence and consistency. Can this person be themselves with a camera in the room, and will they be roughly the same person on the third day.

The pre-interview is the audition. A recorded conversation, not a read. What you are looking for is whether they think out loud, whether they answer the question they were asked, and whether they can tell a story with a beginning and an end.

Test the specific thing you need. If the film requires them to walk you through a workshop, do that in the pre-interview. People who are eloquent sitting down sometimes go silent while doing a task, and the reverse is also true.

Consistency beats brilliance. A contributor who is good every time is worth more than one who was extraordinary once and cannot get back there.

Consent is part of casting, not paperwork afterwards. Explain what the film is, where it will be shown, and what you will do if they change their mind. A contributor who does not understand what they agreed to is a problem for them and a risk to the film.

Treating Actors Well Is a Production Decision

This is not only an ethical point, although it is that. Actors talk to each other, and the casting pool for a low-budget project is a community with a memory.

Respect the schedule. Running ninety minutes late tells every actor in the corridor what working on your set will be like.

Have somewhere to wait that is not a corridor. Water, a chair, a door.

Be clear about pay, dates, and what the project is before they audition, not after they are cast.

Reply to everyone. Including no. Especially no.

Do not ask for free work beyond the audition. A read is reasonable. A full day of "workshopping" for an unpaid part is not.

On a production running largely on goodwill, the casting session is the first impression of how the whole shoot will be run, and word travels faster than any of the material you will publish.

Quick Picks: Tools for Casting

Casting has three separate jobs: reaching actors, running the session, and comparing candidates afterwards. Different tools own each.

Best for reaching actors: a casting platform such as Backstage, Casting Networks, Spotlight, or Mandy. These are where submissions actually come from, and which one you use is largely determined by your territory. Paid, and pricing varies by region and role type (verify current pricing).

Best for session logistics on a crewed production: StudioBinder. Breakdowns, schedules, and session organisation connected to the rest of the production paperwork. Free tier, paid from around $29/month (verify current pricing).

Best for the self-tape pass: shared cloud storage. Google Drive or Dropbox with a consistent folder and naming convention. Unglamorous, free at small volumes, and sufficient.

Best for comparing candidates against the role: Storyflow. The canvas where the character breakdown, your written criteria, and each candidate's notes sit side by side on one board, so the decision is made against the brief rather than against your memory of the room. Free plan is $0; Plus is $7.99/month billed annually. The honest limit: it does not host video, take submissions, or schedule sessions.

Pricing checked July 2026. Competitor pricing changes often and varies by territory, plan, and role type, so verify on each vendor's page before quoting it.

Common Mistakes That Ruin a Casting Session

Different sides or a different reader for different actors. The results stop being comparable and nobody notices.

No redirect. You have measured preparation and called it talent.

Casting on the room. Presence in person and presence on camera are different, which is exactly why you record.

Sessions that run too long. The last four actors are being judged by a tired person.

Writing notes at the end of the day. By then it is recall, and recall favours whoever was most recent and most confident.

Inconsistent recording. Different framing, different light, unusable audio. Half the comparison is gone.

No written criteria. Without them, the default criterion is confidence.

Silence afterwards. Cheap to fix, and it costs you the next casting pool.

The Bottom Line

A casting session is not a showcase. It is a test you are running to make a decision, and a test without a control produces a decision without evidence.

Hold the sides, the reader, and the first direction constant. Change one thing, the redirect, and score what happens. Record everything to one standard so that a week later you are comparing performances rather than memories.

The first read tells you how they prepared. The redirect tells you what they will be like on day three. That second question is the one you are actually hiring against, and it is the one an uncontrolled session can never answer.

Author

Justkay is a documentary filmmaker and the founder of Storyflow. He has cast contributors and non-actors for documentary work, where the audition is a pre-interview and consistency matters more than a single brilliant take.