Naming Convention Proposal

Variables

Replace each placeholder before you run the prompt.

• {{current_names}}A representative sample of current campaign / ad-group names in the account.

  Campaigns: "Brand", "NB Search - Generic", "PMax Industrial 2024", "Search - Wide Match Test", … (paste 30+ rows)

• {{context}}Account complexity — how many regions, brands, channels, languages, levels of segmentation.

  1 brand, 3 regions (US, CA, UK), 4 product lines, search + shopping + pmax channels, English + French.

Propose a naming convention for the account below. Audience: an experienced agency team that will roll this out across 30+ accounts.

CURRENT NAMES (representative sample):
{{current_names}}

CONTEXT:
{{context}}

OUTPUT:

1. The proposed convention: a single template, e.g. "BRAND_REGION_CHANNEL_OBJECTIVE_AUDIENCE".
2. Token definitions: for each token, list the allowed values (with abbreviations).
3. The regex that validates the convention.
4. Migration plan: how to rename existing campaigns without breaking history. Specifically, what to copy vs rename in place, what to leave alone.
5. Edge cases: campaigns / ad groups that don't fit the template — call them out and propose extensions.

Constraints:
- Convention must be ≤80 characters total (Google Ads field limit).
- Tokens must be unambiguous (a token can be parsed back without lookups).
- Convention must work for both English and French campaigns.

Do NOT use spaces — use underscores. Do NOT use special characters — only [A-Z0-9_].

Expected output shape

Template + token definitions + regex + migration plan + edge cases.

Why we wrote it

Naming conventions get debated for hours in agency meetings and rarely documented well. This prompt forces a complete, rule-driven proposal.

How to use

1. Open Claude or ChatGPT. The recommended model for this prompt is claude-sonnet-4-6 — opus when the prompt requires deep reasoning, sonnet for the rest.
2. Replace every {{variable}} with content specific to your account. The examples above are starting points, not templates to ship as-is.
3. Paste the prompt and run.
4. Read the output against the expected shape above. If the model produced a structurally different response, re-prompt rather than accept the drift.