NO BS Marketplace

05/06/2026

Alphabet’s Q1 2026 earnings got plenty of coverage for the 63% Cloud revenue growth and $35.7 billion in quarterly CapEx, but the number that apparently tells the bigger story received almost no attention. Google Cloud backlog, the contractually committed future revenue that enterprise customers have already signed up to pay, nearly doubled quarter over quarter to over $460 billion.

Revenue captures what happened last quarter, while backlog captures what has been committed to happening over the next three to five years. When enterprise customers sign multi-year contracts, the full commitment appears in backlog immediately. Unlike revenue, backlog cannot decline unless contracts get broken, and enterprise procurement teams structure those contracts specifically to make breaking them expensive.

Roughly $230 billion in new long-term commitments were signed in a single 90-day window. Companies committing that kind of money to Google Cloud are choosing which AI infrastructure will power their operations through 2029 and beyond. The enterprise market has effectively voted on AI search permanence with binding contracts, not opinions.

https://nobsmarketplace.com/blog/why-google-cloud-backlog-matters

05/05/2026

Alphabet’s Q1 2026 earnings included three numbers that make it difficult to keep calling AI content consumption an “emerging” channel. Google’s Gemini is processing more than 16 billion tokens per minute through its API alone, up 60% from last quarter. Paid subscriptions across Google’s services reached 350 million, with the strongest quarter ever for consumer AI plans driven by the Gemini App. And Gemini Enterprise paid users grew 40% quarter over quarter.

For context, Netflix has approximately 300 million paid subscribers globally. Spotify has roughly 260 million premium subscribers. Google’s paid subscription base, which increasingly includes AI-powered features, now exceeds any single streaming platform. Nobody describes Netflix at 300 million subscribers as an emerging entertainment channel. It seems fair to apply the same standard here.

All three layers of AI consumption (consumer, enterprise, developer) draw from the same trust signals when deciding which content to surface. A brand visible across credible third-party publications is visible across all three. A brand without that citation presence is invisible across all three. The audience is already there. The question is whether a brand’s content strategy has caught up.

https://nobsmarketplace.com/blog/16-billion-tokens-per-minute-and-what-it-means-for-content

04/30/2026

For two years, the dominant SEO narrative has been that AI Overviews would cannibalize Google Search. Fewer clicks, smaller ad inventory, shrinking organic ecosystem. Apparently, somebody forgot to tell Google.

Alphabet released Q1 2026 earnings on April 29, and Google Search and Other revenue grew 19% year over year to $60.4 billion. That is up from 12% growth in Q1 last year. Total Alphabet revenue hit $109.9 billion, the 11th consecutive quarter of double-digit growth.

Sundar Pichai, CEO of Alphabet and Google, tied the Search performance directly to AI features, saying Search had a strong quarter “with AI experiences driving usage, queries at an all time high, and 19% revenue growth.” The CEO of Google is not describing AI Overviews as a defensive move to retain users. He is describing them as the cause of accelerating Search growth.
The panic framing depended on three assumptions: AI Overviews would shrink clicks, drive users to ChatGPT, and collapse publisher traffic. The Q1 data seems to contradict all three. None of this means publisher CTRs have stayed the same, but the broader story about Search being in structural decline has run into a 19% growth number it cannot explain.

https://nobsmarketplace.com/blog/google-search-19-percent-growth-ai-overviews

04/27/2026

A study of 1.4 million ChatGPT prompts found that the median page cited from the search channel is about 500 days old. Non-cited pages from the same channel were significantly younger. In the search channel, where 88% of all ChatGPT citations come from, older pages outperform newer ones.

The reason connects to how authority accumulates over time. A page live for over a year has had time to earn backlinks, get referenced by other publications, and build the kind of external validation that a language model reads as trustworthiness. A newer page with better content but no external references has not yet passed that threshold.

The news channel works in reverse. Cited news pages have a median age of about 200 days versus 300 for non-cited. When relevance scores between competing news pages are nearly identical, ChatGPT uses recency as the tiebreaker.

Two channels, two opposite age dynamics. Evergreen content benefits from longevity and sustained authority building. Time-sensitive content benefits from speed and precise titling. Knowing which channel a page is likely to enter through determines which strategy applies.

https://nobsmarketplace.com/blog/chatgpt-cites-older-pages-500-days-old

04/20/2026

Similarweb, a digital intelligence platform, surveyed US consumers on how they use AI tools versus search engines at each stage of the purchase journey.
35% of consumers use AI at the product discovery stage. Just 13.6% use search. AI holds a 2:1 or greater advantage through every stage from discovery through evaluation. The gap only closes at the final purchase step.
The attribution problem is real. A consumer discovers a brand inside ChatGPT, researches it there, builds a shortlist there, and then visits the website through a branded Google search. The conversion shows up as search-driven. AI gets zero credit for the work that made the conversion possible.
AI referral traffic to external sites remains flat despite platform usage growing 28.6% over the past year. The platforms were designed to answer, not to route. Measuring AI performance by referral clicks is measuring something the architecture intentionally does not produce.
The metric that maps to buying behavior is brand mention share: how often a brand appears in AI responses relative to competitors. Across every sector Similarweb measured, the brands leading in AI visibility are the ones cited most often by trusted third-party sources, not necessarily the biggest.

Full breakdown on our blog as usual : https://nobsmarketplace.com/blog/brand-discovery-now-happens-before-a-single-search-query-gets-typed

04/15/2026

Google Search Advocate John Mueller replied to a question on Reddit last week about having multiple URLs pointing to the same content. The site owner had changed their URL structure and left the old paths accessible, creating duplicates. They were worried about a penalty.

Mueller confirmed there’s no penalty for duplicate URLs. Google can handle them. But he added a caveat that matters more than the reassurance: “you’re making it harder on yourself” because Google will pick one version to keep, and it might not be the one you’d prefer.

He described technical SEO as “basically search engine whispering, being consistent with hints, and monitoring to see that they get picked up.” Canonical tags, 301 redirects, sitemap inclusion, internal link consistency. None of them are directives. All of them are hints. Google usually follows them when every signal points in the same direction, and guesses when they conflict.

The part most people overlook: when backlinks point to different URL variations of the same page, the authority gets scattered instead of consolidating into the URL you’re trying to rank. The links are real. The equity is real. But it’s diluted across versions Google may or may not treat as canonical.

Full guide covering methods, common mistakes, and how to audit your setup on the blog: https://nobsmarketplace.com/blog/canonicalization-seo-best-practices

04/14/2026

Google CEO Sundar Pichai confirmed on a recent podcast that Google is running Search and Gemini as separate products that will “profoundly diverge” over time. Add AI Mode as a third surface, where Google tests advanced features before migrating them into main search, and the picture becomes clear: Google is no longer one search product.

Each surface handles content differently. Traditional Search still shows ranked results, now topped by AI Overviews that synthesize answers before users scroll to any links. AI Mode runs multi-turn conversations where users explore complex questions without necessarily visiting a website. Gemini operates as a standalone assistant where users may never see a URL, but the model can still reference brands it associates with relevant topics.

For anyone building organic visibility, the question is whether your content and brand are discoverable across all three. A backlink from a credible publication now builds traditional ranking signals, reinforces the authority that AI Overviews draw from, and strengthens entity association in the models behind AI Mode and Gemini.

Most SEO strategies are still optimized for one results page. That results page is the surface Google is investing in least.

Full analysis on the blog: https://nobsmarketplace.com/blog/why-optimizing-for-google-results-page-isnt-enough-anymore

04/13/2026

Google’s capital expenditure budget for 2026 sits between $175 and $185 billion. CEO Sundar Pichai said on a recent podcast that the only reason the number isn’t higher is because the physical materials to build more don’t exist. Wafer capacity, memory chips, power infrastructure, even electricians to wire the data centers. Every bottleneck is physical.

All of that infrastructure is purpose-built for running AI workloads. AI Overviews in search, AI Mode, Gemini. The products already live on this infrastructure, and the buildout is designed to scale them further.

For anyone doing SEO or content marketing, the signal is hard to miss. Companies don’t spend $185 billion on infrastructure they plan to walk away from. AI search features are not a phase to wait out. The traditional results page is being built over, layer by layer, with hardware that has a decades-long operational lifespan.
The foundational work of earning authority on credible third party sites and building brand recognition still matters, possibly more than before. Those are the trust signals AI systems draw from when generating answers. Adapting strategy now beats waiting for concrete and silicon to expire.

Full analysis on the blog: https://nobsmarketplace.com/blog/why-google-is-spending-185-billion-on-ai-infrastructure

04/07/2026

Google disclosed on April 3 that a logging error has been inflating impression counts in Search Console since May 13, 2025. Nearly eleven months of overstated data across every property. Clicks weren't affected. Impressions were.

The fix is rolling out over the coming weeks. When impression numbers drop, nothing changed about a site's actual visibility. The measurement is getting accurate.

Any report or strategy decision that used impressions or CTR during the last eleven months was working with inflated inputs. CTR looked artificially low because the denominator was too large. The SEO community had been noticing the pattern for months, calling it "alligator trending" where the impression and click lines diverge on the chart. Turns out at least some of that was a bug.

The bug also overlaps with Google merging AI Mode data into Search Console totals in June 2025 with no way to separate it. Two significant data discontinuities in the same window makes year-over-year impression comparisons across that period unreliable.

Clicks remain the cleanest signal. Google confirmed they were unaffected by the logging error. For the affected period, clicks are the metric to anchor analysis around.

Average position is worth understanding too. It blends a page's ranking across every query it appeared for. A page ranking position 2 for its main keyword and position 45 for twenty long-tail queries shows an average in the 40s. Misleading on a dashboard unless you filter to specific queries.

If you manage Search Console data: annotate May 13, 2025 as a discontinuity point, use clicks as primary metric for the affected period, export historical data before the fix propagates, and explain the forthcoming impression drop as a correction.

Full breakdown on our blog :

How to Read Google Search Console Metrics in 2026 (And Why Your Impression Data Has Been Wrong for 11 Months)

04/06/2026

Someone on Reddit asked whether splitting a sitemap into multiple files would hurt SEO. Google's John Mueller jumped in with a list of reasons why you'd actually want to split them, and some of them go beyond what most sitemap guides cover.

The standard reason: Google caps a single sitemap at 50,000 URLs or 50MB. Hit the limit, split the file.

The strategic reasons are more useful. Splitting by content type (products, categories, blog posts) gives you per-group indexing data in Search Console. If 200 product pages drop out of the index, you see it immediately in the product sitemap instead of it being buried in one giant report.

Mueller also mentioned splitting by freshness so crawlers can theoretically focus on the sitemap that actually changes, though he was candid about it: "I don't know if this actually happens tho." He suggested proactive splitting before hitting limits, and noted that hreflang annotations can make sitemap files grow fast for multilingual sites.

His last reason: "my computer did it, I don't know why."

The broader takeaway is that sitemaps work better as an active monitoring tool than as a set-it-and-forget-it file. How they're organized determines how granular your Search Console data is, which directly affects how fast you spot indexing problems.

Google ignores the priority and changefreq tags entirely. Only the URL and lastmod date matter, and faking lastmod dates can cause Google to distrust those signals for the whole site.

https://nobsmarketplace.com/blog/xml-sitemaps-in-2026