CyberArk and the Israeli GEO Research Pilot: Measuring Brand Presence and Source Links
Privileged access management is a useful category for studying how an AI answer explains a complex buying decision. Nir Levi’s Hebrew-English research pilot asked which platforms an enterprise in Israel should compare to secure administrator accounts. CyberArk was the preselected tracked brand. The question did not name it, and each language version was repeated in a fresh Perplexity conversation. This created a small, inspectable record of brand presence and supporting exported links.
What the pilot recorded
On 11 October 2026, the pilot collected four answers for this question: two in English and two in Hebrew. CyberArk appeared in 2 of 2 English answers and 2 of 2 Hebrew answers. Links to its official domain appeared in 1 of 2 English exports and 1 of 2 Hebrew exports. The mean overlap between the two languages’ exported source-host sets was 42.1%, calculated across the two rounds. All four observations used Perplexity Search with the GPT-6.1 Sol model label in fresh incognito conversations.
Source-host overlap is the number of shared hostnames divided by the number of distinct hostnames across the two answer exports. Perplexity’s Copy output is the measured record; grouped on-screen sources may contain additional links.
Define the problem before writing the comparison
A practical content brief begins by identifying which privileged identities and administrative workflows the buyer needs to manage. It can then explain the scope of the product, the implementation responsibilities and the information required for evaluation. The goal is to make the buying question intelligible rather than to claim that every organisation needs the same solution. Current product documentation should support the details, and a technical owner should review the terminology.
Separate the company from the page that explains it
The pilot tracks whether the category brand is mentioned and whether a link to its official domain is present in the copied answer. These fields can answer different editorial questions. A brand may be recognized while another source provides the explanation a reader sees. When reviewing a live answer, inspect the particular page and passage, not only the vendor name. An implementation guide, product overview and third-party comparison perform different roles in that source record.
Repeat a stable question after a concrete change
For a follow-up review, retain the wording, language and search settings before altering a page. Describe the actual change, such as clarifying a deployment prerequisite or correcting an outdated product description. Then repeat the same question and inspect the result alongside the original. A longer series can reveal whether an observation recurs. It also gives the team a way to distinguish an editorial task from the larger business question of whether qualified buyers are reaching the site.
Nir Levi’s perspective
First, check which sites are being cited and why. Then identify information missing from the company’s own site and publish useful original content that answers those questions, with real data and examples. The aim is to give AI answers a clearer, better-supported basis for describing the company.
Read the original research
The complete study, paired prompts, evidence downloads and Nir Levi’s commentary are available at https://nirlevi.com/en/research/israeli-geo-hebrew-english/
Put the finding to work
This is the type of evidence an Israeli enterprise software team can bring to a GEO review with Nir Levi: a defined buyer scenario, saved answers, specific source pages and an information task that someone can own and verify.
Explore GEO and AEO consulting in Israel with Nir Levi at nirlevi.com and bring a buyer question to your next review.