Workflows
How to Filter Video Search Results by Date, Duration, or Resolution
A search query might return hundreds of results. Narrowing by when it was shot, how long it is, or what resolution it was captured at turns a broad search into a precise one.
You search your library for "interview" and get 340 results. Somewhere in those 340 clips is the specific interview you need, the one from last month's client shoot, the short version, the one shot in 4K. But you are going to scroll through a lot of clips to find it.
This is the point where keyword search stops being helpful and starts being a wall of noise.
Why keyword search alone falls short
Text search is powerful. Being able to find every clip where someone says "product launch" or every scene with a whiteboard is genuinely useful. But power without precision creates its own problems. The bigger your library gets, the more results any broad keyword returns.
The issue is not that the search is returning wrong results. It is returning too many right results. You searched for interviews and got interviews. All of them, when you only wanted one.
What you need is a way to layer constraints on top of your keyword search. Keep the content match, but narrow by metadata: when was it shot, how long is it, what format is it in.
The filters available in FrameQuery
FrameQuery's search combines content-based matching (what is in the video) with metadata filtering (facts about the file itself). The metadata filters include:
Date range. Filter by when the video was shot or when it was added to your library. Use a specific range (March 1 to March 31) or a relative range (last 30 days, this quarter). Date filtering is the single most effective way to narrow a broad search because you almost always have at least a rough idea of when something was shot.
Duration range. Set a minimum and maximum length. This is particularly useful when you know the approximate length of what you are looking for. A 30-second social clip and a 45-minute full interview are very different things, even if they both match "interview."
Source folder. Filter by where the file lives on disk. If your footage is organized by project, client, or shoot on your drives, source filtering lets you constrain your search to a specific location without having to browse there manually.
Project. Filter by which project the video belongs to in FrameQuery. Similar to source folder filtering, but based on your FrameQuery project structure rather than your file system.
Resolution. Filter by the resolution of the footage. Find only 4K clips, or only HD, or only footage above a certain resolution threshold. Useful when you need delivery-quality originals rather than phone footage.
Codec. Filter by the codec or container format. Find all R3D files, all ProRes clips, all H.265 recordings. This matters when you need to know what format something is in before pulling it into an edit.
These metadata filters combine with the content-based search modalities: transcript (what was said), scenes (what was shown), objects (what appeared), filenames, color, and people.
Combining filters for precision
The real value is in combining multiple filters in a single search. Each additional filter narrows your results significantly.
Consider the difference:
- "product demo" returns every product demo in your library. Potentially hundreds of clips.
- "product demo" + last 3 months returns only recent ones. Maybe 20 clips.
- "product demo" + last 3 months + under 10 minutes returns short, recent product demos. Perhaps 5 clips.
- "product demo" + last 3 months + under 10 minutes + 4K returns exactly what you need for that client deliverable.
Each filter is simple on its own. The combination is what makes them powerful. You go from hundreds of results to a handful in seconds, without needing to know the exact filename or remember which folder it was in.
Practical filtering scenarios
Here are common situations where filters solve real problems:
Finding footage from a specific shoot. You know the shoot happened last Tuesday. Filter by date (last week) and source folder (the drive you ingested to). Even without a keyword, this narrows your library to just that shoot's footage.
Pulling clips for social media. You need clips under 60 seconds that work as standalone pieces. Filter by duration (under 60 seconds) and search for the relevant topic. Smart collections can automate this further by saving the duration filter as a persistent view.
Locating 4K originals for delivery. The client wants the highest quality versions. Filter by resolution (4K or above) and codec (ProRes or R3D) to find delivery-grade files, skipping over proxies and screen recordings.
Reviewing recent additions. A team member processed new footage yesterday. Filter by date added (last 24 hours) to see exactly what came in, without it getting lost in the larger library.
Finding everything from a client. Filter by project or source folder to see all footage associated with a specific client. Add a date range to narrow to a particular engagement period.
Saving filter combinations
Once you find a filter combination that solves a recurring need, you do not have to rebuild it every time. FrameQuery lets you save complex filter combinations as saved searches for one-click reuse.
A saved search stores the complete query: keywords, content filters, and metadata filters together. Click it and the search runs with all filters applied. This is particularly useful for queries you run weekly or daily, like "what new footage came in this week" or "short clips from the current project."
For filters you want to apply permanently as a persistent view, smart collections take saved searches a step further by updating automatically in the background.
Filters as a habit
The editors who get the most from search tools are the ones who develop a habit of filtering, not just searching. The instinct to type a keyword and scroll is strong. But adding even one filter (usually date) before hitting search cuts your results dramatically and saves the scrolling time entirely.
Start with date. If you have any idea of when something was shot, apply the date filter first. It is almost always the fastest way to cut a large result set down to a manageable one. Layer duration and source on top when you need further precision.
Turn broad searches into precise ones. Join the waitlist to try filtered video search with your own library.