Lore

Native format plus customer-language creative strategy

# Native format plus customer-language creative strategy

A three-part process for ads that convert, the intermediate level of AI UGC three-level ladder (beginner, intermediate, expert).

The three parts

  1. Find a native content format that already performs well on the platform (see Scroll behaviour: why ads should look like native organic content for why native matters). tiktok-format-to-prompt-pipeline is one way to source these.
  2. Do deep customer research on a specific avatar. The source names Reddit, TikTok and similar places.
  3. Rewrite the format's words in the customers' own language, so the proven format carries avatar-specific wording.

Example

For busy single working moms, replace a generic caption like 'Watch me apply this makeup' with 'When you have 5 minutes to apply your foundation before your work husband arrives'.

Notes

The format supplies the structure that holds attention, and the customer research supplies the language that makes it relevant to one avatar. Both are inputs to the generation step, not prompt tricks, which fits the thesis that inputs and workflow matter more than fighting the model.

Related

Custom GPT for ad-variation prompts, Custom GPT for viral-optimized scripts

Angles as the reusable customer-language layer

At the expert level the two jobs, format earning attention and customer-specific words converting, are stored as two separate libraries. Formats go in a format library. Angles in the avatar's own language, found by researching customers on Reddit, TikTok and similar places, go in a winning-angle library. Because the two are kept apart, a proven angle can be applied to many different formats. See format-library-and-winning-angle-library.