Guides
Videography hiring: a reliability scorecard for interviews
A videography hiring scorecard for reliability: which of five interview questions brings out which behavior, and the warning sign in each weak answer.
What to take away
- Reliability is a separate quality from talent, and it is the one that decides whether you can sleep the night before a shoot.
- A reel cannot show it. A question about the worst day, and a reference asked the right way, can.
- Ask the same five questions of everyone and write the answers down. An unstructured interview measures rapport and nothing else.
- Ask references what they observed, not which dates they remember.
- Map each question to one behavior and score the behavior, so a smooth answer cannot hide a missing step.
Four behaviors to listen for
Listen for four behaviors, none of which appears in a portfolio. Each has a question that brings it out, so decide what you are listening for before the conversation starts and judge the answers against that list rather than against your first impression.
Four Reliability Behaviors
- Verifycopy cards to two destinations
- Finish the last ten percent
- Escalate early, not at wrap
- Say the uncomfortable thing politely
- They verify. Cards are copied to two destinations, checked and confirmed out loud before anyone leaves. This is the one non-negotiable.
- They finish the last ten percent. Files are named to convention and the project is handed over in a state someone else could open.
- They escalate early. A problem raised at ten in the morning is a scheduling adjustment; the same problem raised at wrap is a reshoot.
- They say the uncomfortable thing to a client, politely. Someone who agrees to everything on site will agree to a sixth deliverable and tell you afterward.
Candidates describe their best work, so ask about the worst day. Every useful signal in this trade lives in how someone handled a shoot that went wrong.
Write the role down so the answers can be compared
A vague advertisement produces vague applicants. State the actual work: project types, typical day length, travel, whether the role is shoot, post or both, and what the person will own without supervision.
What the Ad Must State
- Project types and typical day length
- Travel requirements
- Shoot, post, or both
- Software and gear used
- Unsupervised responsibilities
- Honest working pattern and weekends
State the working pattern honestly too. A candidate who learns about weekend event work in month two will leave in month three, and the interview is the cheapest place to find out that the pattern does not suit them.
Keep the advertisement and the process the same for every applicant. The EEOC publishes guidance for small employers on nondiscrimination in recruitment and employment, and a consistent, written process is the practical way a small company complies without a human resources department.
Five questions, asked of everyone
Ask the same five of every candidate and write the answers down. Identical questions make the answers comparable, and a written record stops the memory of the friendliest candidate from deciding for you.
- Tell me about a shoot that went wrong. What happened, what did you do, and what would you do differently? Listen for whether they own any part of it.
- Walk me through your backup routine on a two-camera day. Listen for verification, not for the number of copies.
- A client asks for a change after sign-off that will take a day. What do you say to them, and what do you say to me? Listen for whether they escalate.
- Describe a project you handed to someone else. What did you leave them? Listen for naming, notes and organization.
- What part of this work do you dislike? Listen for candor.
Five Structured Questions
- Tell me about a shoot that went wrong
- Walk me through a two-camera backup routine
- Client asks for a change after sign-off
- Describe a project you handed over
- What part of this work do you dislike
Compare the five answers side by side once every interview is finished. The candidate who says nothing to question five is either inexperienced or managing you, and either is worth writing down.
Keep the notes short: one line per question, quoting what the candidate actually did rather than how the answer felt. Notes taken that way can be set beside the reference answers and the trial result, so three separate sources point at the same four behaviors.
Which question brings out which behavior
Each of the four behaviors has one question that brings it out. Score the behavior, not how the answer sounded, and write the evidence beside each mark. Read the warning sign first, because a polished answer is easy to give and a missing step is hard to hide.
| Behavior | Question | A usable answer contains | Warning sign |
|---|---|---|---|
| They verify | 2. The backup routine on a two-camera day | A check that each copy is good, said out loud | Counts the copies and stops there |
| They finish the last ten percent | 4. A project handed to someone else | Naming, notes and a state someone else could open | Describes the edit and skips the handover |
| They escalate early | 3. A change after sign-off that takes a day | Tells the client the cost and tells you before work starts | Agrees on the spot, or absorbs it quietly |
| They say the uncomfortable thing | 5. The part of the work they dislike | A specific, candid answer | Nothing, or a polished non-answer |
Question one, the shoot that went wrong, runs under all four. Listen for whether the candidate owns any part of it.
Mark each behavior present, partly or absent from three sources: the interview, the reference call and the trial. Verification is the one non-negotiable, so an absent mark there ends the discussion, whatever the other three say.
References that report behavior, not dates
A reference who confirms dates tells you little about day-to-day reliability. Ask for people who watched the candidate work, such as a producer, a lead shooter or a client contact, and ask what they directly observed:
- What work did the candidate own, and what did you observe them hand off?
- When did they spot or escalate a problem, and how did they handle it?
- Did they meet deadlines and leave files or notes organized for the next person?
- What supervision or working conditions helped them do their best work?
Reference Questions That Work
- Alone with a difficult client, and why
- Anything you had to check behind them
- What did they need most support with
- What would you have changed to keep them
Finish every reference call with one more question: would you hire this person again for a similar role, and why or why not? The answer can surface both the candidate's real friction and the previous employer's, and it shows how to manage them from the first week.
Test one claim with a paid trial
Use a paid trial to test what the answers claimed. For an editor, send real rushes, a real brief and a real deliverable schedule, set a realistic deadline, then score what comes back against the schedule rather than against taste.
Scoring the Paid Trial
- Hit runtime, versions and ratios
- Files named to your convention
- Asked a good clarifying question
- Audio survived a phone speaker
- Delivered on date, flagged slippage early
- Pay for the trial
For a shooter, spend half a day beside them on a real project and watch the setup and the wrap rather than the shooting. Do they monitor audio on headphones? Do they record room tone unprompted? Do they count gear back into cases?
Pay for the trial and agree its task, time limit, deliverables, deadline and pay in writing before it starts. An unpaid test selects for people who can afford to work for free and repels the ones already busy.
What the conversation cannot settle
Reliability does not decide how you engage someone. Whether a person is an employee or a contractor depends on facts about the working relationship, not the invoice, and the consequences of getting it wrong fall on the business.
The Department of Labor publishes compliance assistance for new and small businesses covering wage and hour obligations. Have an accountant or employment attorney review classification and paperwork before the arrangement starts, and keep time records from the first day.
What to do with a gap the screen leaves
A candidate can pass every question and still fail in an undefined role. Whatever the interview, the references and the trial left open becomes the first observed task, and the training sequence structures the sign-offs that follow.
Record what the person demonstrated at each stage, because attendance is not sign-off. The order in which to add people at all, and the role map behind it, is in hiring and training a crew.
Common questions
These come up once the scorecard is in use, and each answer leans on the evidence rather than on first impressions.
What if the trial is good and the references are not?
Trust the references. A trial is one supervised task done under observation; a reference describes a hundred unsupervised days.
What if a candidate has no usable references?
Lean on the trial and on the answers to questions one to four, and ask for a project they handed to someone else. A missing reference is a gap to fill by observation, not a verdict.
Should I hire someone better than me at their craft?
Yes, and expect it to be uncomfortable. Hire for reliability and for a standard you can define; you do not need to be the strongest editor in your own company.
Does hiring change if I want to work in another city?
Considerably. A remote operator you cannot stand beside needs a much stronger written system and a much higher bar on reliability, which is one of the harder parts of expanding into another market.
