ora
elastic.co
success
Claude Code
intent

What does elastic.co do and who is it for? Explain it back to me.

16steps
65.8sduration
$0.2189cost
121,616tokens
19 steps3 reasoning steps7 searches
home
docs
/products
/solutions
search
/customers
search
search
search
docs
docs
docs
search
/elasticsearch
search
search
56%
on-site discovery
100%
reliability
44%
link following
path origin
  • previous resource44%
  • web search44%
  • prior knowledge11%
insight

The agent successfully gathered enough information to explain Elastic's core offering, target audience, pricing models, and competitive differentiation. It synthesized information from the homepage, product/solutions pages, multiple pricing tiers, and customer pages, supplemented by web search results. The site is moderately navigable but requires manual assembly across multiple pages and contexts—pricing in particular is fragmented across serverless and hosted pages without side-by-side cost comparisons, and key terminology (Elasticsearch Platform vs. Elastic Stack) is used inconsistently.

  • ›The homepage (step [1]) and product page (step [3]) clearly positioned Elastic as 'The Search AI Company' and identified Elasticsearch as the core engine, but did not synthesize the full breadth of use cases (observability, security, search) in one place.
  • ›Pricing pages (steps [2], [13], [14]) exist and were fetchable but present costs in fragmented form: step [2] overview mentions 'serverless and hosted offerings' without numbers; step [13] (hosted) and step [14] (serverless) require separate visits and contain different cost metrics (VCUs + storage vs. instance types vs. per-GB ingestion). The agent had to infer typical costs from web search results rather than from the official pages.
  • ›The customers/use-cases page (step [15]) was successfully retrieved but the trajectory shows the agent needed to search for 'use cases who customers' (step [11]) to discover it—it was not prominently linked from the homepage or products page during natural navigation.
  • ›Web search results (steps [6], [7], [8], [16], [17]) provided competitive context (Splunk, OpenSearch, SigNoz, Datadog) and pricing details that were NOT clearly stated on elastic.co's own pricing pages, suggesting the site publishes sparse numerical cost examples.
  • ›The agent noted genuine friction points: 'Search AI Lake' terminology unexplained, free tier limitations not prominent, AI feature pricing unclear, and inconsistent naming (Elasticsearch Platform vs. Elastic Stack). These are surfacing gaps in the site's information architecture rather than agent failures.

Want to run your own?

Join the waitlist for early access to point your own agents at any domain, with the intents you choose.

or talk to us about agent readiness →