Video Script Planner & Teleprompter

Start with a decision the viewer can make

A useful script begins with a small, observable outcome. Instead of planning a video about your whole service, decide what someone should understand or do after watching. A viewer might need to identify the correct collection shelf, compare two versions of a sample, or complete one preparation step. Write that outcome in a sentence before filling the narration. It becomes a practical test for every scene: does this part help the viewer reach that outcome?

Next, gather the facts you can actually support. Put the real object on the table, inspect the relevant screen, or confirm the instructions with the person responsible. Separate an observation from a claim about results. Showing that a pouch has three pockets is different from saying that it prevents every lost accessory. The first can be demonstrated in this video; the second needs evidence beyond a tidy shot. Your script should not quietly turn one into the other.

Draft in language you could comfortably say to one person. Read a sentence aloud before polishing it for appearance. Written introductions often contain background that the camera will establish immediately, while important instructions hide inside long clauses. Move the useful noun and action forward. Keep necessary conditions close to the instruction they qualify, particularly when the viewer could otherwise follow the wrong procedure.

Choose a project title that identifies the particular version you are making. A title such as Workshop kit return check is more useful during a handoff than New video. Select the destination context, then write your opening, demonstration, and ending in your own words. Nothing in this workspace generates the argument, verifies the facts, or replaces your judgment. When an idea remains uncertain, resolve it before recording rather than making the narration sound more confident.

Build scenes around changes in the explanation

Use a scene for a meaningful unit of communication, not automatically for every sentence or camera angle. One scene can contain a spoken explanation and several related visual instructions. Start a new scene when the purpose changes, when a silent action needs its own allowance, or when rearranging the material would otherwise become confusing. A compact plan is easier to revise than twenty fragments whose titles say almost the same thing.

Give each scene a descriptive title and choose its role: opening, point, demo, transition, or close. The title helps you find the material; the role describes its job in the explanation. Neither is spoken by the teleprompter. A demo role does not create footage or add instructions automatically. It simply helps you recognize whether the plan contains the evidence promised by the opening, rather than several points followed by an unsupported conclusion.

Add, duplicate, remove, and move scenes up or down as the draft develops. Duplication is useful for comparing two phrasings, but remember that both copies remain part of the plan until you remove the unused one. Moving a scene changes the intended viewing and narration order. Read the joins after each move: a sentence beginning with this result may no longer have a clear result immediately before it. The planner does not rewrite those connections for you.

The project supports up to twenty scenes and always retains at least one. A scene becomes active when its narration or visual notes contain something beyond whitespace. An unused scene has no estimated duration, even if its pause control still shows a value. By contrast, a scene with visual notes and no narration can represent a silent action. Keep abandoned alternatives outside the working project or save a separate JSON version; otherwise their words and allowances can still affect the total.

Separate what is said from what must be filmed

Narration is the text the audience should hear. Visual notes tell the person making the video what must appear. Keep stage directions out of the narration unless you genuinely intend to read them aloud. A note such as Hold the adapter beside its label belongs with the visuals. The spoken sentence might instead explain why that match matters. This separation makes the production worksheet useful without feeding camera instructions into the teleprompter.

Make visual notes specific enough to execute. Name the object, relevant detail, starting state, and action. Replace Show the kit with Open pouch on desk; all three empty pockets visible; place each accessory beside its label. Those are ordinary notes you write, not additional structured fields. You can include a location, prop reminder, or pickup instruction in the same space. The storyboard and shot-list guide develops this approach for a small shoot.

Decide which actions occur under speech and which need extra silence. If a hand places an accessory while the narration continues, its entire action time should not automatically be added again. If the audience must inspect the finished arrangement after the sentence ends, add an allowance for that inspection. The planned pause is an extra duration in the estimate, not an analysis of what the camera or speaker will actually do. Its position within your performance remains your production decision.

Do not make essential meaning depend only on a gesture or an unexplained color. Naming the pocket and accessory in speech can help someone who cannot see the detail clearly. Also plan the later captioning and accessibility work your finished video needs. This tool does not produce captions, audio description, or an accessibility certification. Its contribution is earlier: it gives you somewhere to notice an information gap while changing a sentence or shot is still straightforward.

Work through a complete three-scene example

Consider a hypothetical video showing how to check a borrowed presentation kit before returning it. Create three scenes called Inventory, Check the pockets, and Arrange the return. Use opening, demo, and close roles. Set the reading pace to 140 words per minute. The following narration is deliberately short enough to inspect directly; it is a worked planning example, not a claim about how long a real person will take.

Inventory, 24 words: Before you return a shared presentation kit, check the small accessories as well as the main item. This pouch should contain three separate pieces. Write visual notes showing the open pouch and accessories together. Set a two-second planned pause for an additional view of the complete set.

Check the pockets, 36 words: Lay the adapter, cable, and remote beside the checklist. Match each item to its printed name, then place it in the marked pocket. Leave the checklist visible until every pocket has been checked by one person. Allow six extra seconds for a silent inspection after the spoken demonstration, rather than assuming all handling happens during speech.

Arrange the return, 20 words: Close the pouch only after the check. If something is missing, tell the kit owner before you arrange the return. Add three seconds to hold on the closed pouch and return note. This ending gives the viewer a concrete next action without inventing a universal returns policy.

The scene estimates are 24 x 60 / 140 + 2 = 12.3 seconds, 36 x 60 / 140 + 6 = 21.4 seconds, and 20 x 60 / 140 + 3 = 11.6 seconds when displayed to one decimal. The underlying total is 80 x 60 / 140 + 11 = 45.2857 seconds: displayed as 45.3 seconds and approximately 00:45. Rehearse the actual pocket check next. If inspection needs longer, change the allowance or simplify the demonstration; do not treat the calculation as evidence that the action fits.

Understand the timing model before making cuts

For each active scene, estimated seconds equal counted narration words multiplied by 60, divided by your reading pace, plus the planned pause. The total adds the unrounded scene values before displaying the result. Reading pace accepts whole numbers from 60 to 240 words per minute, with 140 as the default. Those are input boundaries and a starting setting, not a recommended speed for every speaker, audience, or subject.

The word count uses whitespace-separated narration tokens that contain at least one letter or digit. A hyphenated expression without spaces counts as one token. Standalone punctuation or an emoji does not count, while a numeral does. Scene titles and visual notes do not contribute words. Consequently, 250 can count as one token even though its spoken form takes several words. Write numbers and abbreviations as you intend to say them when that makes rehearsal and estimation clearer.

This is a rough model for space-separated text, not linguistic analysis. It is especially unreliable for languages that do not normally separate words with spaces. Even in English, a product name, unfamiliar term, or long number can take longer than another token. The estimate also cannot know whether someone breathes, hesitates, repeats a sentence, waits for a screen, or pauses naturally between clauses. It is useful for comparing drafts under stated assumptions, not predicting a precise recording.

Use the total to ask a better question. If the kit example feels rushed at 140, trying 120 changes its modelled total to 51.0 seconds; trying 160 gives 41.0 seconds. Neither setting proves the narration will sound better. If you have a real delivery deadline, measure a rehearsal and revise the content that causes difficulty. The video script timing guide explains how to calibrate pace and avoid counting the same silence twice.

Rehearse narration without confusing scroll speed and pace

The teleprompter displays narration from the scenes in order, with separation between their texts. It does not display scene titles or visual notes. Before opening it, check that every visible word belongs in the spoken delivery. If the presenter needs a direction, agree that cue separately or keep the production worksheet nearby. A clean narration view reduces the chance of accidentally speaking a reminder intended only for the person operating the camera.

Start with the text paused and choose a comfortable font size. Play advances the script; pause lets you stop without abandoning the rehearsal; restart returns to the beginning. The available font range is 24 to 72 pixels, and scroll speed runs from 10 to 100 pixels per second. A mirror toggle is available when your physical setup needs reversed text. It does not replace checking that your particular reflection arrangement is readable.

Reading pace and scrolling speed solve different problems. Words per minute drives the duration estimate, while pixels per second controls how quickly the text travels on screen. Changing font size or line wrapping changes how much text occupies the view. Do not expect a particular pixel speed to equal a particular speaking pace. Planned pauses also do not stop the scrolling automatically, and the tool does not listen to your voice or follow your progress.

Rehearse once for language and again for delivery with the planned actions. Use a separate timer if you want a measured duration, and record elsewhere only when you are ready. Pause whenever the wording causes a stumble; revise the sentence rather than training yourself to fight it. Closing the dialog, using Escape, or hiding the page pauses the prompter. No camera or microphone permission is required because this is a scrolling text aid, not a recording session.

Choose an export that preserves the work you need

Copy narration when another recording workflow needs only the spoken text. That copy does not include the production notes, so it is not the right handoff for someone arranging props or capturing supporting shots. For a fuller handoff, export the TXT production worksheet or PDF. These include the narration, scene information, visual notes, timing, and assumptions. Review what you are sending: notes can contain internal instructions that should never appear in a public post.

Use CSV when you need the scene information in a spreadsheet. It contains all scenes, including inactive ones, along with the exported settings and mapping information. It is a data handoff, not a promise of a live production board or an importer for another service. If you reorganize that spreadsheet into a shooting schedule, keep the original scene references so the editor can still relate footage to the intended viewing order.

Save JSON to retain a complete version 1 project that the planner can reopen. Give saved files meaningful names in your own file manager, such as kit-return-draft-02, so an experimental rewrite cannot be mistaken for the approved narration. Reopen a saved version when you need to continue; there is no autosave or browser local-storage recovery. Refreshing or closing the page is not a substitute for saving your work. The JSON import limit is 512 KiB.

The PDF is a rendered production worksheet, not an editable text document. Retain TXT or JSON when later editing or reusable text matters. Very expanded documents can reach the forty-page PDF guard; use TXT or CSV for those handoffs. Script text and project files are processed in your browser rather than transmitted by this tool. The wider website still has ordinary site analytics, so that privacy boundary should not be confused with a claim that the whole site makes no network requests.

Adapt the plan to the video job, not an invented platform rule

The destination selector labels the project; all nine contexts use the same planning and timing engine. It does not simulate native interfaces, check upload eligibility, or impose a platform-specific duration. Choose the guide that matches the work. An Instagram process reveal can focus on one visible change, while a TikTok product-use demonstration needs a clear starting condition and an honest reveal of what the product does.

A YouTube tutorial plan may require prerequisites, screen captures, and pickup shots across several stages. A Facebook community explainer instead needs the event sequence and practical information to agree. For a LinkedIn evidence-led explanation, keep claims close to the example and caveat that support them. These differences belong to the assignment and audience, not a hidden scoring system in the planner.

Use a Threads video reply to answer one conversation while supplying enough context for a new viewer. An X video update can explain a visible comparison while leaving supporting dates and units clear. A Snapchat walkthrough depends on connected locations and silent beats. A Reddit process demonstration benefits from a reproducible baseline, test conditions, and observable result rather than a promotional opening.

Before leaving the plan, ask someone to identify the promised outcome, necessary evidence, and next action without your verbal explanation. Resolve missing information, save the editable project, and take the appropriate worksheet into production. Browse the free tools directory for other preparation tasks. When you need a broader social media workflow, SMTasker is a separate product with a five-day trial. It is not the engine of this free planner, and starting a trial is not required to write, rehearse, or export here.