How to set up a custom ad naming convention

A naming convention is a reusable template that turns every uploaded creative into a clean, consistent ad name. Instead of editing names by hand on every launch, you define an ordered list of tokens once, and uplads assembles the name for each creative automatically. This article walks through building that schema in Configuration → Account.

Starting from a preset or building from scratch

You don't have to build a schema token by token. At the top of the Naming Convention card there's a Load preset… dropdown with ready-made schemas you can load with one click:

• Creative Schema - Style / Asset Type / Length / Creator Age / Hook / Dimensions, joined with _.
• Performance marketing - Hook / Audience / Offer / Format.

Loading a preset is non-destructive: it fills in the full schema, but you can still edit, add, remove, or reorder anything before saving. To go back to the built-in starting point, click Reset - that returns you to the default schema (Filename | File type | ID).

Adding, removing and reordering tokens

Each row in the schema is one token. Use the Add token dropdown at the bottom of the card to append a new one:

1. Pick a token type from the dropdown. The basic types are:
   • Filename - the creative's filename (cleaned, see toggles below).
   • File type - resolves to Image, Video, Carousel, or File.
   • Ad set name - the name of the ad set (the ad group on TikTok) the ad is created in. In a launch into several ad sets, each ad carries its own ad set's name.
   • Custom text - literal text you type in (great for a brand or campaign tag).
   • Date - today's date in a format you choose.
   • Auto-id - a sequential id, useful for guaranteeing unique names.
2. For Date, also pick a format (e.g. yyyy-MM-dd, yyyyMMdd, MMM-dd-yyyy). For Custom text, type the text in the box that appears.
3. Click Add.

Below the basic types, the dropdown lists Common fields pulled from the presets (Style, Asset Type, Hook, Audience, and more). Picking one adds a fully configured field token in a single click. The last option, Custom field…, opens the field editor so you can define your own - see the next section.

To reorder tokens, drag any row by its grip handle; a violet insertion line shows where it will land. To remove a token, click the X on its row. Field tokens also have a pencil icon to edit them in place.

Per-field asset-type filters (image-only or video-only fields)

Field tokens are special: they define a slot whose value is filled per creative in the Ad Launcher. When you add or edit one (via Custom field… or the pencil icon), the dialog lets you set:

• Label - what the field is called (e.g. Length, Hook).
• Description - an optional one-line hint shown at launch.
• Applies to - All, Images, or Videos.
• Default value - used when no value is picked at launch.
• Options - predefined chips offered as quick picks (users can still type a custom value).

The Applies to filter is what makes image-only or video-only fields work. If you mark a field as Videos, it resolves to a value only on video creatives. On images it resolves to an empty string and is dropped from the name entirely - no dangling separators left behind. This is how the Creative Schema's Length field only shows up on video ads.

Choosing a separator and the cleanup toggles

Use the Separator dropdown to choose what joins the tokens: underscore (_), hyphen (-), space, dot (.), or pipe (|).

Three checkboxes clean up the filename token:

• Remove dimensions from name - strips size fragments like 9x16, 4x5 or 1080x1350, plus version markers (v1, final, HD), date stamps and a leading order number, so they don't clutter the ad name.
• Remove brand name from filename - removes the start of the filename only when it matches one of the names in the Brand names field or the ad account's own name. Everything else stays: with no brand names listed and no match, the filename is left whole, so construction-ad-9x16.mp4 never shrinks to ad.
• Use the number at the end of the filename as ID - Hook_01.mp4 gets the ID 01 instead of a running number, and the number is dropped from the filename part so it isn't printed twice. Files without a trailing number keep the running number. This only applies when your schema contains the ID token.

The first two only change the Filename token, the third also fills the ID token; other tokens are left as-is.

Previewing names and saving an account-wide default

The Preview box shows a live, fully assembled name as you edit, using sample values so you can see exactly where separators land and how field tokens resolve. Tweak tokens, separator, and toggles until the preview reads the way you want.

When you're happy, click Save. The schema is stored as the default for the selected ad account, so every new launch starts from it automatically. You can still adjust the naming for a single launch - see How to override the naming schema for a single launch. To learn how field values get filled in (including AI vision auto-detection), see How naming field values are auto-detected, and to see names in context before launch, How to review and edit ad names before launch.