A character loses a finger in episode two and has all ten in episode five. Nobody notices until a cast member asks at the table read, or an editor spots it in the cut. If you work on a returning series, you need a way to track character continuity that does not depend on one person's memory.
This guide is for script editors, story producers and showrunners. It covers what actually drifts, where a tracking system tends to fail, and a routine you can run on every new draft.
What drifts, and why it is hard to see
Continuity is, in the words of Wikipedia's definition, the consistency of the characteristics of people, plot, objects and places seen by the audience over time. In television, that "over time" is long, and the people writing it change.
Most character errors fall into a few groups:
- Identity details: spellings of names, nicknames, who calls whom what.
- Physical facts: age, injuries, scars, glasses, hair, disability.
- History: where someone grew up, who they have slept with, what they know about the crime.
- Knowledge state: what a character knows at this point. This is the one that causes the most damage, because it changes the meaning of a scene.
These errors are rarely careless. They happen because different writers draft different episodes, because scripts are revised under time pressure, and because the thing that contradicts is often a single line of dialogue on page 41, not a scene anyone would flag.
Start with a bible that records what has been established
A series bible is the usual first defence. Wikipedia describes a bible as a reference document for characters, settings and other elements of a project, and notes that bibles are typically updated once information has been established on screen or in scripts.
That second part matters. A bible written before the series is a plan. A bible updated after each episode is a record. Continuity checking needs the record, because the question you are asking is not "what did we intend?" but "what have we already told the audience?"
A useful character entry is short and checkable. For each regular and recurring character, keep:
- The name as written, plus every variant used in dialogue.
- Age or date of birth, and the story-date it is measured from.
- Physical facts that persist: injuries, scars, mobility, appearance.
- Key relationships and the episode where each was established.
- A knowledge log: what they learn, and in which scene.
If an entry has no episode reference, it is an opinion. If it does, anyone on the team can verify it in a minute.
Track character continuity per draft, not per series
Many teams check continuity once, near the end of a block. By then, the error has been carried through two or three drafts and possibly into the schedule. A more reliable pattern is to check every incoming draft against what is already established.
A simple routine for each new draft:
- Read for changes, not for story. The question is what this draft asserts about each character that earlier material did not.
- Compare against the bible and the preceding episodes. Not the whole show: the episodes where that character's facts were last touched.
- Separate error from decision. A contradiction may be an accident, or a deliberate change. Retroactive continuity, or "retcon", is the deliberate version. The fix for an accident is to correct the script. The fix for a decision is to update the bible and check what it breaks downstream.
- Log the outcome. Accepted, fixed, or overruled, and why. This is what stops the same query being raised three times.
- Send one consolidated note. Writers respond better to a single list than to a dozen scattered messages.
The log is the part teams skip, and it is the part that saves time in the second half of a run.
Where spreadsheets and memory break down
A spreadsheet is a sensible start, and it works for a limited cast over one season. It tends to fail in three places.
First, it is only as current as the last person who edited it. Second, it records facts but does not read the script, so someone still has to find every line that touches those facts. Third, it rarely connects episodes to seasons, so a fact established in series one is easy to miss in series three.
On set, the script supervisor holds responsibility for continuity during the shoot, and ScreenSkills describes the role as including breakdowns prepared in advance so that story and continuity queries can be flagged early. The point of that preparation is the same one that applies at script stage: the cheapest place to catch a continuity problem is before anyone has filmed it.
Where CONT'D fits
CONT'D is a screenplay intelligence tool: it reads and analyses scripts, and it does not write or rewrite them. It is currently in a pilot with a small number of UK productions, so we are careful about what we claim.
The pilot is built around exactly this workflow: when a new draft arrives, check continuity across episode, season and project, covering character names, ages and injuries as well as format errors, and gather the results into a single report. A person then accepts or rejects each flag. The aim is a faster, more consistent version of the routine above, with the decision always remaining with the script editor.
The short version
To track character continuity in a TV series, record what has been established rather than what was planned, tie every fact to an episode, and check each new draft against that record before it reaches the table read. Separate accidents from deliberate changes, and keep a log of what was decided. Tools can help with the reading, but the habit of checking every draft is what prevents the missing finger.
If you would like to see how this works in practice, join the pilot waitlist at https://contd.app/waitlist.