AI Citation Baseline Audit (Onboarding)

AI Citation Baseline Audit (Onboarding)

1. 01

   Freeze the client's current AI answer surface before any optimization work begins

   Query ChatGPT, Perplexity, Gemini, and Google AI Overviews with 15 to 25 core category and product prompts, then capture raw answer text, cited URLs, and brand mention position for each. This becomes the immutable comparison point for every later retainer review.

2. 02

   Separate branded prompts from non-branded discovery prompts in the tracking sheet

   Branded queries show how AI engines describe the client; non-branded queries show whether the client appears at all. Agencies that blend the two overstate visibility gains and get caught when a client runs their own spot check.

3. 03

   Log the citation sources shaping each answer, not just whether the brand appeared

   Record which competitor pages, Reddit threads, YouTube videos, or review sites the engine pulled from. Pallix maps these source chains directly, and the same source list is what a manual audit should produce.

4. 04

   Pull the client's existing organic keyword and backlink baseline from a competitive research platform

   Export ranking keywords, estimated traffic, and referring domains for the client and its top three competitors. SpyFu covers organic and paid keyword overlap in one pass, while Linkody handles the backlink side with new, lost, and changed link alerts.

5. 05

   Score each competitor on AI share of voice and traditional rank side by side

   A competitor can rank page one on Google and be absent from AI answers, or the reverse. Flag every gap where the client is cited but not ranking, and every keyword where the client ranks but is never cited.

6. 06

   Set the reporting cadence and the definition of a visibility win before the first client call

   Agree on weekly or biweekly snapshots, the exact prompt list, and what counts as movement. Voxoria tracks prompts daily and benchmarks against competitors, which sets a realistic expectation for how fast answers shift.

7. 07

   Document data-source limitations in the baseline memo

   Third-party AI visibility platforms sample engines rather than query them exhaustively, so numbers carry latency and coverage gaps. State this in writing at onboarding so a later variance between two tools does not read as an agency error.