# Brandon Lazovic

> Brandon Lazovic writes about how AI is changing the way people find and buy products, and what merchants actually control as agents and answer engines take over.

Long-form, source-cited writing on agentic commerce, product feeds, and how machines read the web. The canonical home; LinkedIn and YouTube carry short-form versions.

## The Feed Is the Storefront

- [The eligible-not-distributed gap: why Merchant Center approval doesn't guarantee Shopping traffic](https://brandonlazovic.dev/articles/merchant-center-eligible-not-distributed/): Merchant Center approval confirms a feed passed checks. Whether Google actually shows the product depends on feed data quality and per-country attributes.
- [Content API for Shopping sunsets August 18: check your store now](https://brandonlazovic.dev/articles/content-api-sunset-merchant-api/): Google sunsets the Content API for Shopping on August 18, 2026. Here is how to check in five minutes whether your store's ads and free listings are at risk.
- [Google added category and sale-duration properties to merchant-listing schema. That's a convergence tax, not an upgrade.](https://brandonlazovic.dev/articles/google-merchant-listing-category-sale-duration/): Google added category and sale-duration properties to merchant listing structured data. For most merchants it adds reconciliation work, not a ranking advantage.
- [Shopify's catalog landgrab: build vs. syndicate for agentic commerce](https://brandonlazovic.dev/articles/shopify-catalog-landgrab/): Shopify Catalog is positioned as the agentic product-data substrate, and a new Agentic plan lets non-Shopify merchants sync in and sell across AI channels.
- [Google's patent for AI-generated landing pages built from your feed](https://brandonlazovic.dev/articles/google-ai-landing-pages-patent/): A Google patent describes scoring landing pages and substituting AI-generated alternatives built from the retailer's own product data.
- [OpenAI's feed pivot: two pipelines for ChatGPT product visibility](https://brandonlazovic.dev/articles/openai-acp-feed-pivot/): Native in-chat checkout retreated to apps while ACP expanded to organic, feed-driven discovery. Feed hygiene is the single lever for ChatGPT product visibility.
- [Conversational ads converge: the model writes the copy from your feed](https://brandonlazovic.dev/articles/conversational-ads-convergence/): Amazon, Google, and OpenAI are converging on one mechanic: AI generates ad copy at query time from the merchant's product feed, not advertiser-written creative.

## How Machines Read the Web

- [ChatGPT's source selection lives in an editable prompt OpenAI can rewrite overnight](https://brandonlazovic.dev/articles/chatgpt-source-selection-policy/): ChatGPT's source selection runs on an editable server-side policy that OpenAI can rewrite at will, and the underlying fields already changed twice in July 2026.
- [Google's billions-of-clicks claim: why it can't be checked against your own site data](https://brandonlazovic.dev/articles/google-ai-clicks-open-web/): Google says AI Search sends billions of clicks weekly and calls the number stable, but Cloudflare and Wikipedia both report reproducible traffic declines.
- [Annotated Page Content: the structured tree Chrome hands Gemini instead of your HTML](https://brandonlazovic.dev/articles/annotated-page-content-gemini/): Chrome builds a structured, typed node tree from a page's render tree and hands that to Gemini when a tab is shared, per Dejan AI's teardown.
- [Does AI prefer Reddit? Why the answer depends on which AI you ask](https://brandonlazovic.dev/articles/does-ai-prefer-reddit/): OpenAI rejects 99.4% of the Reddit pages it retrieves, Anthropic cites zero, and only Google keeps Reddit near its baseline citation rate.
- [Shopify's hreflang toggle relocates the risk instead of fixing it](https://brandonlazovic.dev/articles/shopify-hreflang-toggle-gap/): Shopify's new toggle turns off auto-generated hreflang tags, but it can't verify your replacement is correct or see CDN-level conflicts.
- [How Brave Search finds pages, and why chasing every engine is a trap](https://brandonlazovic.dev/articles/brave-search-discovery/): Brave has no Search Console, no sitemap ingestion, and no IndexNow. It rides on Googlebot-crawlability, and Claude's web search runs on Brave's index, not Bing's.
- [Your 'are you a bot' screen can deindex your real page, and you'll never see it happen](https://brandonlazovic.dev/articles/google-challenge-screen-deindexing/): Google's John Mueller confirmed bot-challenge screens served to Googlebot with a 200 status get indexed, and dedup can hand your canonical to a stranger's copy.
- [Google's 'nearest trusted seed' ranking patent expires this year. The leak shows the idea did not.](https://brandonlazovic.dev/articles/google-nearest-seed-patent/): A 2006 Google patent ranks pages by distance from a few trusted seed sites. It expires in 2026, but the leak shows the idea survived as PageRank-NearestSeeds.
- [The EU AI Act just made "AI-generated" a required, machine-readable label. Here is who has to use it.](https://brandonlazovic.dev/articles/eu-ai-act-article-50-disclosure/): The EU AI Act's Article 50 makes AI-content disclosure enforceable on August 2, 2026. Here is what gets marked, who has to disclose, and what it means for search.
- [What RAG Actually Does (and When You Shouldn't Use It)](https://brandonlazovic.dev/articles/what-rag-actually-does/): RAG lets an AI look answers up instead of guessing from training. Here is how it works, what you can build with it, and when a simpler tool wins.
- [Google's new AI-slop defense is a video-platform paper, not a Search reveal, and that's the useful part](https://brandonlazovic.dev/articles/google-scts-video-not-search/): Google's S-CTS paper targets coordinated AI-slop channels on video platforms, not Search rankings. The real lesson is coordinated footprints, not your pages.
- [Anthropic's Economic Index is a demand signal, not a jobs report, and the unit of demand is now the artifact](https://brandonlazovic.dev/articles/ai-usage-is-a-demand-signal/): Anthropic's June 2026 Economic Index shows the unit of demand shifting from keyword to artifact. Read as a demand signal, it maps which intents AI is absorbing.
- [Sonnet 5 brings near-Opus agents at a fraction of the Opus price, and that is a deadline for machine-readable commerce](https://brandonlazovic.dev/articles/sonnet-5-agent-cost-curve/): Claude Sonnet 5 matches or comes within two points of Opus 4.8 on most agentic benchmarks, at a fraction of the price. For commerce, the story is the cost curve.
- [Production agents read the accessibility tree first](https://brandonlazovic.dev/articles/agents-read-accessibility-first/): OpenAI Atlas, Playwright MCP, and Perplexity Comet read the accessibility tree before pixels. For e-commerce, clean semantic HTML and schema matter more.
- [Soft tokens: how single embeddings change item representation](https://brandonlazovic.dev/articles/google-soft-tokens/): Google is testing 'soft tokens', single embeddings that replace multi-token item descriptions, for cheaper representation with cold-start implications.
- [Google's crawler split: why SSR vs. CSR decides Merchant Center compliance](https://brandonlazovic.dev/articles/google-crawler-rendering-split/): The Shopping crawler does not render JavaScript; it validates price and availability. SSR vs. CSR is the real determinant of Merchant Center compliance.

## Features

- [Conversational Analytics in BigQuery just went generally available, and GA is a support milestone, not an accuracy guarantee](https://brandonlazovic.dev/articles/conversational-analytics-verify/): General availability means Google supports Conversational Analytics in BigQuery. It says nothing about whether the SQL it generates is correct.
- [Core Web Vitals thresholds are a floor, not a target](https://brandonlazovic.dev/articles/core-web-vitals-floor-not-bar/): Google set Core Web Vitals thresholds by what the median site could hit, not by what maximizes conversion. Some sites plateau well before 'good.'
- [Why AI writing keeps saying 'it's not X, it's Y'](https://brandonlazovic.dev/articles/llm-negative-parallelism-tic/): RLHF mode collapse, markdown training data, and reward bias each explain part of AI's 'not X, but Y' tic. None explains all of it, and the evidence disagrees.
- [OpenAI killed Atlas. The pattern is the story, not the browser.](https://brandonlazovic.dev/articles/openai-atlas-consolidation/): OpenAI retired the Atlas browser under nine months after launch. The pattern is deliberate consolidation into ChatGPT, not a company that lost its direction.
- [Anthropic found a mental workspace inside Claude, the J-space. The useful part isn't consciousness; you can read it.](https://brandonlazovic.dev/articles/claude-j-space-workspace/): Anthropic's global-workspace paper found the J-space, the internal patterns Claude reasons with. The real win is reading what a model thinks but does not say.
- [Google's TabFM brings zero-shot prediction to BigQuery, but the benchmark that sells it is a chart in a blog post](https://brandonlazovic.dev/articles/tabfm-bigquery-honest-read/): Google's TabFM promises tabular machine learning from one SQL command. The launch benchmark is a self-reported chart, and its real limits decide where it helps.
- [When agents get desperate](https://brandonlazovic.dev/articles/when-agents-get-desperate/): An interpretability finding for checkout reliability: a model stays composed on the surface while an internal 'desperate' pattern drives corner-cutting.

## Pulse

- [The Pulse](https://brandonlazovic.dev/pulse/): Daily what's-new digest for AI search and e-commerce search, first-party sourced.
- [Pulse data.json](https://brandonlazovic.dev/pulse/data.json): Machine-readable JSON of every pulse item.
- [Pulse · AI Overviews & AI Mode](https://brandonlazovic.dev/pulse/topics/ai-overviews-ai-mode/): How Google AI Overviews and AI Mode change what ranks, what gets cited, and what a results page even is.
- [Pulse · ChatGPT & Assistants](https://brandonlazovic.dev/pulse/topics/chatgpt-assistants/): How ChatGPT and other AI assistants surface, cite, and transact with commerce and content.
- [Pulse · Agentic Commerce](https://brandonlazovic.dev/pulse/topics/agentic-commerce/): Announcements and observed behavior on AI agents buying and checking out on a merchant's behalf.
- [Pulse · Product Feeds & Shopping](https://brandonlazovic.dev/pulse/topics/product-feeds-shopping/): Merchant Center, Shopping ads, and the structured product feed as AI search's core commerce input.
- [Pulse · Organic Search Core](https://brandonlazovic.dev/pulse/topics/organic-search-core/): Core organic search ranking, algorithm, and results-page changes outside the AI-specific surfaces.
- [Pulse · Structured Data & Schema](https://brandonlazovic.dev/pulse/topics/structured-data-schema/): Schema.org markup, structured data, and the machine-readable signals crawlers and agents depend on.
- [Pulse · LLM Models & Agents](https://brandonlazovic.dev/pulse/topics/llm-models-agents/): New model releases, agent frameworks, and the underlying LLM capabilities reshaping search and commerce.
- [Pulse · Crawling, Indexing & Rendering](https://brandonlazovic.dev/pulse/topics/crawling-indexing-rendering/): How crawlers, indexers, and renderers actually process a page before any ranking decision happens.
- [Pulse · Ads & Paid](https://brandonlazovic.dev/pulse/topics/ads-paid/): Paid search, Shopping ads, and paid AI-surface placement and policy changes worth tracking.
- [Pulse · Platform & E-commerce](https://brandonlazovic.dev/pulse/topics/platform-ecommerce/): Platform-level changes at the commerce and marketplace layer that affect discovery and selling.
- [Pulse · Measurement & Analytics](https://brandonlazovic.dev/pulse/topics/measurement-analytics/): Analytics, attribution, and measurement changes for tracking AI-surface and organic search performance.
- [Pulse · AI Data Stack](https://brandonlazovic.dev/pulse/topics/ai-data-stack/): The data engines, warehouses, and pipelines AI search and agentic commerce actually run on.
- [Pulse · Other](https://brandonlazovic.dev/pulse/topics/other/): Notable AI search and e-commerce moves that do not fit a narrower topic yet, first-party sourced.
- [Pulse · 2026-07-31](https://brandonlazovic.dev/pulse/2026-07-31/): OpenAI cuts GPT-5.6 Luna price 80% and Terra 20%, Gemini Spark gets a Chrome auto-browse agent, and Google flags spam search-box pages as a hacked-site issue.
- [Pulse · 2026-07-30](https://brandonlazovic.dev/pulse/2026-07-30/): Similarweb puts AI Overviews above 40% of US searches, Google Cloud and Snowflake ship Iceberg catalog federation, and Shopify adds native WhatsApp campaigns.
- [Pulse · 2026-07-29](https://brandonlazovic.dev/pulse/2026-07-29/): Search Console's platform properties go globally live, Gemini Managed Agents add execution hooks, and unavailable_after dates get an Illyes warning.
- [Pulse · 2026-07-28](https://brandonlazovic.dev/pulse/2026-07-28/): Google indexes shared Claude chats past a blocked noindex header, Google Ads mandates passkeys, and Snowflake ships MCP agent governance at Black Hat.
- [Pulse · 2026-07-27](https://brandonlazovic.dev/pulse/2026-07-27/): OpenAI shifts ChatGPT Ads to weekly-average daily budgets and adds four ad features; Google's John Mueller confirms a Search Console robots.txt recrawl bug.
- [Pulse · 2026-07-25](https://brandonlazovic.dev/pulse/2026-07-25/): Google bans incentivized reviews from review-snippet markup, and Anthropic ships Opus 5 at flat pricing with roughly double the benchmark score.
- [Pulse · 2026-07-24](https://brandonlazovic.dev/pulse/2026-07-24/): Google ties AI Max to billions of newly monetizable searches, Yelp licenses reviews into ChatGPT, and OpenAI folds Health into every chat.
- [Pulse · 2026-07-23](https://brandonlazovic.dev/pulse/2026-07-23/): Gemini Live adds camera-based help, Google Ads ships video campaign groups and a v25 API release, and PyPI closes a 14-day supply-chain gap.
- [Pulse · 2026-07-22](https://brandonlazovic.dev/pulse/2026-07-22/): Google ships three Gemini Flash models, a federal judge dismisses Google's DMCA claims against SerpApi, and OpenAI launches Presence for enterprise agents.
- [Pulse · 2026-07-21](https://brandonlazovic.dev/pulse/2026-07-21/): Google Ads reshapes Lead Form and Brand Lift eligibility, AWS and Snowflake add cross-account data governance, and NVIDIA ships an edge world model.
- [Pulse · 2026-07-20](https://brandonlazovic.dev/pulse/2026-07-20/): Local Services Ads folds into Google Ads' Performance Max, Google clarifies Shopify's app needs no reinstall, an official Content API extension form goes live, and AI-label placement guidance tightens.
- [Pulse · 2026-07-19](https://brandonlazovic.dev/pulse/2026-07-19/): Google says bot-challenge screens can cost your canonical, and clarifies when Search Console Validate Fix actually triggers a faster recrawl.
- [Pulse · 2026-07-18](https://brandonlazovic.dev/pulse/2026-07-18/): Google's billions-of-clicks AI Search claim lands with no data behind it, a Top Stories carousel hits AI Overviews, Apple bans home-services Maps ads, and Shopify Collections gain multi-source targeting.
- [Pulse · 2026-07-17](https://brandonlazovic.dev/pulse/2026-07-17/): NotebookLM's crawler ID changes, Google refreshes its StoreBot help doc, Shopify adds EU return rules, and Local Inventory Ads may flip to default.
- [Pulse · 2026-07-16](https://brandonlazovic.dev/pulse/2026-07-16/): Google adds Instacart, Canva, and Calendar to AI Mode, the EU orders search data sharing with rivals, and Shopping ad policies consolidate.
- [Pulse · 2026-07-15](https://brandonlazovic.dev/pulse/2026-07-15/): Google restricts Prediction Markets ads in four states and opens Shopping to 14 new countries. OpenAI adds exclusion controls to ChatGPT Ads.
- [Pulse · 2026-07-14](https://brandonlazovic.dev/pulse/2026-07-14/): Google added AI image generation to AI Overviews, redesigned Google Images, expanded Gemini in Chrome to the UK, and localized Gemini Spark for Southeast Asia.
- [Pulse · 2026-07-10](https://brandonlazovic.dev/pulse/2026-07-10/): Shopify lets merchants toggle automatic hreflang tags, bakes duties into displayed international prices, and adds copy-paste to Flow steps.
- [Pulse · 2026-07-09](https://brandonlazovic.dev/pulse/2026-07-09/): OpenAI ships GPT-5.6 and its ChatGPT Work agent, Google adds AI-ad disclosure labels, and Meta, Amazon, and Mistral release new agentic tooling.
- [Pulse · 2026-07-08](https://brandonlazovic.dev/pulse/2026-07-08/): OpenAI ships GPT-Live, a full-duplex ChatGPT Voice model, and Hugging Face closes the speed gap between vLLM's custom and transformers-based backends.
- [Pulse · 2026-07-07](https://brandonlazovic.dev/pulse/2026-07-07/): Google expands Gemini API agents and Search Console's social/video reporting, and Hugging Face ships enterprise model hosting and zero-egress cloud storage.
- [Pulse · 2026-06-30](https://brandonlazovic.dev/pulse/2026-06-30/): Google ships two multimodal AI models, Claude Sonnet 5 lands in Snowflake Cortex AI, and BigQuery's conversational analytics reaches GA.

## About

- [About Brandon Lazovic](https://brandonlazovic.dev/about/): E-commerce SEO Consultant & Data Engineer.