Boost your skills with Growth Memo’s weekly expert insights. Subscribe for free!
This week, I share my findings from analyzing 1.2 million ChatGPT responses to answer the question of how to improve your chances of getting cited.
For 20 years, SEOs have written”ultimate guides” designed to keep humans on the page. We write long intros. We drag insights all along through the draft and into the conclusion. We build suspense to the final call to action.
The data shows that this style of writing is not ideal for AI visibility.
After analyzing 1.2 million verified ChatGPT citations, I found a pattern so consistent it has a P-Value of 0.0: the “ski ramp.” ChatGPT pays disproportionate attention to the top 30% of your content. Furthermore, I found five clear characteristics of content that gets cited. To win in the AI era, you need to start writing like a journalist.
1. Which Sections Of A Text Are Most Likely To Be Cited By ChatGPT?
There isn’t much known about which parts of a text LLMs cite. We analyzed 18,012 citations and found a “ski ramp” distribution.
- 44.2% of all citations come from the first 30% of text (the intro). The AI reads like a journalist. It grabs the “Who, What, Where” from the top. If your key insight is in the intro, the chances it gets cited are high.
- 31.1% of citations come from the 30-70% of a text (the middle). If you bury your key product features in paragraph 12 of a 20-paragraph post, the AI is 2.5x less likely to cite it.
- 24.7% of citations come from the last third of an article (the conclusion). It proves the AI does wake up at the end (much like humans). It skips the actual footer (see the 90-100% drop-off), but it loves the “Summary” or “Conclusion” section right before the footer.
Possible explanations for the ski ramp pattern are training and efficiency:
- LLMs are trained on journalism and academic papers, which follow the “BLUF” (Bottom Line Up Front) structure. The model learns that the most “weighted” information is always at the top.
- While modern models can read up to 1 million tokens for a single interaction (~700,000-800,000 words), they aim to establish the frame as fast as possible, then interpret everything else through that frame.
18,000 out of 1.2 million citations gives us all the insight we need. The P-Value of this analysis is 0.0, meaning it’s statistically indisputable. I split the data into batches (randomized validation splits) to demonstrate the stability of the results.
- Batch 1 was slightly flatter, but batches 2, 3, and 4 are almost identical.
- Conclusion: Because batches 2, 3, and 4 locked onto the exact same pattern, the data is stable across all 1.2 million citations.
While these batches confirm the macro-level stability of where ChatGPT looks across a document, they raise a new question about its granular behavior: Does this top-heavy bias persist even within a single…
Source link
Disclaimer
We strive to uphold the highest ethical standards in all of our reporting and coverage. We blogs.grocliq.com want to be transparent with our readers about any potential conflicts of interest that may arise in our work. It’s possible that some of the investors we feature may have connections to other businesses, including competitors or companies we write about. However, we want to assure our readers that this will not have any impact on the integrity or impartiality of our reporting. We are committed to delivering accurate, unbiased news and information to our audience, and we will continue to uphold our ethics and principles in all of our work. Thank you for your trust and support.
Website Upgradation is going on for any glitch kindly connect at [email protected]
