Moz Identifies Common Mistakes In AI Prompt Tracking
Moz reveals four mistakes in AI prompt tracking, including focusing on citations and thinking in terms of rankings. These mistakes may lead to misunderstandings about AI visibility.
Original reporting · moz.com
Original: “4 Prompt Tracking Mistakes — Whiteboard Friday”
Read “4 Prompt Tracking Mistakes — Whiteboard Friday” at Moz →
How this was written. Clicktag wrote this article with a language model from Moz’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.
Key points from Moz’s report
- Moz identifies four common mistakes in AI prompt tracking, including focusing on citations and thinking in terms of rankings.
- Tracking the correct quantity of prompts is crucial, and a large brand should track thousands or tens of thousands of prompts.
- Considering a wide range of terms, including long-tail prompts, is essential to get a representative picture.
Moz has been discussing prompt tracking, a big topic in the industry, particularly over the last year. The company has built out prompt tracking in both Moz Pro and STAT, and is including it for all customers.
Moz fans have reservations about prompt tracking, so the focus is on four mistakes people make in any tool, leading to reservations and misunderstandings. The first mistake is focusing on citations rather than mentions.
For example, if Apple is tracking its performance and sees a mention of its brand, but the citation is from an external site like PCMag, it should not be considered a failure. What matters is that Apple is mentioned, not where the information comes from.
AI search starts with a prompt, and identifying prompts that matter to a brand is crucial. Moz Pro's Prompt Suggestions can help with this. The goal is to be in the conversation and have awareness, rather than just getting a click or referral traffic.
The second mistake is thinking only in terms of rankings, like traditional SEO metrics. However, in AI prompt tracking, it's more important to look at how many times a brand is mentioned, what percentage of responses mention it, and what kind of adjectives are used.
The third mistake is not tracking the correct quantity of prompts for a business. A large brand like Apple should track thousands or tens of thousands of prompts, rather than just a few. The outputs vary more from day to day and are less deterministic.
The last mistake is only tracking head terms. When deciding what prompts to track, it's essential to consider a wide range of terms, including long-tail prompts, to get a representative picture.
Questions and answers
What is the first mistake in AI prompt tracking?
The first mistake is focusing on citations rather than mentions.
Why is tracking the correct quantity of prompts important?
Tracking the correct quantity of prompts is crucial because the outputs vary more from day to day and are less deterministic.
What is the last mistake in AI prompt tracking?
The last mistake is only tracking head terms, rather than considering a wide range of terms, including long-tail prompts.



