Weekly Update

Weekly AI Search Update — W29 2026

Google AI Overviews expands citation panels, Perplexity ships shopping-intent scoring, and what both changes ask of your source layer.

CiteRank AI Research·21 July 2026·5 min readObservation log

Observation log. Observation log. This is a dated record of engine behaviour changes we observed. It carries no quantitative claim and no sampled estimate.

WeeklyGooglePerplexity

Key takeaways

—More citation slots means more room for secondary sources — mid-authority brands gain the most.
—Shopping-intent scoring rewards structured product attributes over persuasive copy.

What changed

AI Overviews widened the citation panel on commercial queries, surfacing more sources per answer. Perplexity began differentiating shopping intent from research intent, changing which attributes it retrieves.

What it means for you

A wider citation panel is an opportunity for brands that were previously just below the cut. If your Citation Score has been flat while your Mention Share rose, this is the week to re-sample.

For commerce brands, audit whether product attributes are exposed as structured data rather than only as prose.

›Re-sample commercial prompts before drawing conclusions from last month's baseline.
›Check that your most-cited page still resolves to a stable, dated URL.
›Confirm product attribute markup is present and consistent across the catalogue.

Watch next week

Whether the widened panel persists after the rollout window, and whether it extends beyond commercial queries into professional-services prompts.

Evidence basis

How this entry was produced

Observation log. This is a dated record of engine behaviour changes we observed. It carries no quantitative claim and no sampled estimate.

Research type: Observation log — dated record of engine behaviour change
Classification: Observation log
Data collected: Observed during the week ending 2026-07-21.
Prompt sample size: Not a sampled study. Changes were noticed during routine monitoring rather than measured against a fixed corpus.
Replays per prompt: Not applicable — no controlled replay design was used.
Engines covered: Perplexity, Google AI Overviews
Engine versions: Not stated. Engine vendors do not expose a stable build identifier for every model, so a version cannot be claimed accurately.
Methodology: Qualitative observation from routine monitoring runs and vendor announcements, written up weekly. No statistical test was performed.

Limitations

An observation is not a measurement. Nothing in this entry establishes the size, duration or generality of the change described.
Changes described may be a staged rollout visible to some accounts and regions and not others.
AI engines are non-deterministic: an identical prompt can return a different answer on replay, so any figure derived from them is an estimate, never a fixed value.
Engine vendors change retrieval and ranking behaviour without notice, so anything stated here describes the stated window only.

Methodology, limitations and disclosure

Every CiteRank study states who produced it, what it measured and where it stops being reliable. The full scoring model is documented on the methodology page.

Author: CiteRank AI Research
Author role: CiteRank AI Research team — measurement, prompt-corpus design and scoring
Review: Internal editorial review by the CiteRank AI Research team. No external or academic peer review was conducted.
Published: 21 Jul 2026
Last updated: 21 Jul 2026
AI engines: Perplexity, Google AI Overviews
Model versions: Specific model build identifiers are not disclosed by every vendor and are therefore not claimed here.
Industry scope: Cross-industry
Geographic scope: Global
Language scope: English
Sample size: Not disclosed for this entry
Conflicts of interest: CiteRank AI publishes this research and sells an AI visibility platform. No third party funded, commissioned or reviewed this entry.
Data availability: Underlying raw data is not published. Method and scoring are documented on the Methodology page.

Limitations

AI engines are non-deterministic: an identical prompt can return a different answer on replay, so every figure is a sampled estimate rather than a fixed value.
Engine vendors change retrieval and ranking behaviour without notice. Findings describe the sampling window stated above, not a permanent state.
Results describe the prompt corpus that was designed for this study. A different corpus for the same brand can produce a materially different picture.
This entry does not disclose a sample size, so its figures should not be treated as statistically representative.

Corrections and revisions

No corrections have been issued for this entry since publication on 21 Jul 2026. If a figure or claim here is wrong, write to research@citerank.in. Substantive corrections are published inline with the date they were made, and the original wording is retained in the note.

Suggested citation

CiteRank AI Research (2026). Weekly AI Search Update — W29 2026. CiteRank AI. https://www.citerank.in/research/weekly-2026-w29

← Back to the AI Research Centre