Mehdi Oudjida’s Post

Heads-up: the popular Looker Studio-based Chrome User Experience (CrUX) dashboard is headed for sunset. If you’ve been tracking Core Web Vitals in that familiar interface (think: bold LCP charts, crisp design), the countdown has started. - CrUX data demand has exploded. Web performance pros, SEO folks, and site owners all want fresh, real-user metrics. Result? The BigQuery connector is now struggling under the load, with outages and slowdowns becoming the norm. - Google’s not standing still. Official CrUX API endpoints are here: real-time metrics, weekly historical data, and a shift from static dashboards to dynamic, API-driven tools. - RUMvision may have been first to make CrUX data easy in Looker, but the ecosystem’s evolved fast. Google and third parties are rolling out dashboards built on the sturdier CrUX History API, and they’re not looking back. Key date: end of November 2025. Google pulls the plug on the CrUX Connector. Any dashboard depending on it? No new data, no updates, no dice. Anyone relying on automated Core Web Vitals reporting for monitoring or audits: you’re in the crosshairs. Options: • Try Google’s web-based dashboard (CrUX History API under the hood). More stable, future-proof. • Check out third-party or open-source dashboards now tuned for the new API. • Build custom: devs and analysts can use the APIs directly for tailored dashboards, deeper dives, device splits, trends—whatever suits the workflow. Best practices: • Audit what you’ve got. Find everything using the old connector. • Plan your migration (allocate time, resources, sanity) before November 2025. • Stay plugged in by following Google’s CrUX updates and news on web performance tools. Web performance analytics isn’t standing still. Expect more automation, predictive insights, and tighter integration with monitoring platforms. The Looker Studio CrUX dashboard’s sunset isn’t the end—it’s a pivot to more flexible, robust monitoring. Prep now and stay ahead of the curve. #CrUX #WebPerformance #CoreWebVitals #LookerStudio #APIs

#LookerStudio Release - September 11, 2025 ➡️ Deprecation and sunset of Looker Studio 🪦 Explorer The LS team has decided to remove this legacy feature, available from the home page of the tool. You will no more be able to create explorations from September 15 and the feature will be completely removed from the interface on October 25, converting automatically existing explorations to reports. I put a link about the details of this sunset in the comment. I tested this feature when it was first added to Data Studio, but I haven’t touched it since. It allows you to do the same thing as a chart, without the multimodal flexibility of a report page. Its disappearance doesn't surprise me. Really not a big loss. Explorer (beta), rest in peace. ➡️ Hyperlinks & images disabled for data sources with viewer credentials This limitation does not apply if the report creator and the viewer are members of the same team workspace. Yesterday, when the users started to notice this change, the LS staff had explained this choice very clearly on the community forum (see below the explanations), announcing also what it is coming: A viewer opt-in mechanism to use data so making available again this kind of feature. Evolutions for a good data governance (so supervising all the aspects and related risks). " ...We had to make this change to prevent a report creator from gaining access to data when a report viewer clicks a link on a report with viewer’s credentials. Links to URLs that use data from a data source with viewer’s credentials DO work if the report creator and the viewer are in the same team workspace. In this scenario, both the report creator and the report viewer are part of the same organization. We’re going to add the ability for workspace admins to disable these kinds of links, to provide additional security. We’ll also add a way for report viewers to opt in to using data, obtained with their credentials, in links - we’ll do this for both the Free and Pro plans. Stay tuned! " ➡️ Pro feature: Number of scheduled reports increased to 200 ➡️ Official release of modernized treemap chart I already talked about this chart refresh last week, when the feature was technically available. If you missed it, I put a link to the related linkedIn post in the comments. ➡️ Partner connector additions Voila. That's all for this week!

If you're using #GA4 with #BigQuery, and you've noticed inconsistent user counts in Looker Studio reports, you're not alone. We've outlined the reason these numbers don't always match, why user counts in GA4 can look inflated, and we'll show you how to deduplicate the data in this article 👇 𝗚𝗲𝘁𝘁𝗶𝗻𝗴 𝗔𝗰𝗰𝘂𝗿𝗮𝘁𝗲 𝗨𝘀𝗲𝗿 𝗖𝗼𝘂𝗻𝘁𝘀 𝗶𝗻 𝗟𝗼𝗼𝗸𝗲𝗿 𝗦𝘁𝘂𝗱𝗶𝗼 🔗 https://lnkd.in/gZ7vBX_X #LookerStudio #DataAnalytics #AnalyticsTips #LookerStudioTips #DataDeduplication #CleanData

⚠️ Sampling Destroys Trust in Looker Studio x GA4 Reports. I was recently digging into a Looker Studio dashboard connected to GA4, analyzing revenue by item_promotion_name for the purchase event. We are running serious segment-driven personalization on the web, and we need to evaluate it accurately. But here’s what we found instead of accuracy: For a single day: €3,000 (accurate based on data in DWH) For a shorter date range (month): €3600 (overestimated???) For a wider date range (this year): €600 (estimated) WTF 🤯 👉 Same dataset. Same orders. Completely different results. Why? Because GA4 applies sampling when dealing with large datasets. Instead of processing all your events, it takes a fraction and estimates the rest. It looks like it even overestimates... And that’s the real problem — you start questioning every number, every chart, every conclusion in the dashboard. Trust in your data is gone. 🎯 Takeaway: If you care about accurate reporting, don’t rely on sampled data. 👉 Use BigQuery — it gives you unsampled, raw GA4 data so you can confidently make decisions. But WAIT ‼️ ‼️ you can’t just plug raw GA4 data into Looker Studio — that’s too expensive - you can burn thousands of euros for just refreshing the dashboard multiple times per month. What you need is modeled, clean / dashboard-ready data with tools like DBT or GA4Dataform.com - to make your reporting ultra fast, scalable, and trustworthy. #BigQuery #GA4 #DigitalAnalytics #MarketingAnalytics #LookerStudio

The 'GA4 Annotations' connector has finally been approved and is now an official Looker Studio connector 🎉 It automatically imports annotations from Google Analytics into Looker Studio just like any other data source. To use it, search 'GA4 Annotations' in LS data sources, add the connector, select a GA property, assign categories to colors and off you go. Then you can: - Create a table of annotations that changes when you switch dates or filters.  - Blend annotations with GA4 data (or any other source) to overlay annotations on your charts. - Assign categories to the annotation colors. Let me know how it works for you (or if you run into any issues 🙏 ). Happy annotating 🙂 =========================== And in case you missed it, here is some other GA4 annotation fun: - Push annotations from Google Tag Manager to GA4: https://lnkd.in/gsmpfz3G - Sync your marketing calendar with GA4 annotations: https://lnkd.in/gYj8-bNB - Make bulk GA4 annotation updates from Google Sheets: https://lnkd.in/gNFUUPh3 And for the uber GTM geeks, GTMFixer makes GTM even better: https://lnkd.in/gWKzti3T

CrUX Dashboard will be retired 2025 If your Core Web Vitals reports live in Looker Studio, this matters. Google says the dashboard wasn’t built for wide use. It hit outages (often when new data dropped). And the BigQuery “connector” that powers it goes away in late November 2025. I get the stress. Dashboards break. Teams lose history. Stakeholders still want answers. Here’s what I’d do this week: →Move monitoring to CrUX Vis. It’s faster, built for scale, and already seeing far more usage than the old dashboard. →Switch trend tracking to the CrUX History API. You’ll get weekly data instead of monthly, so issues surface sooner. →If you must keep legacy reports, connect to BigQuery with your own credentials. The public dataset stays. The old connector doesn’t. Key notes: →Expect the CrUX Connector to shut down in late Nov 2025. →When it does, Looker Studio dashboards that rely on it stop updating. →BigQuery data itself continues, so your historical source remains intact. This isn’t just a tool swap. It’s a chance to simplify your perf stack, cut outages, and make your CWV reviews more frequent and useful. #seo #corewebvitals #webperformance #digitalmarketing #analytics #marketingwithsumitdas

Remember the nightmare of segmenting GSC data in Looker Studio? It’s over. No matter how good your Search Console Looker Studio template is, for each new customer website with 1k+ URLs, you have to update the CASE WHEN formula to define page segments and update all your charts. A new report at Sitechecker kills this problem: - creating page segments is much easier - you can create them by a list of dozens of different URLs, even without a subfolder - it's easier to measure the impact of Google updates and the content updates you made Top use case: create a segment of the most valuable pages (aka money pages) and send a separate performance report by them. And this is just a start. The next steps: - adding GA4 Insights, where reports by the same page segments will be presented too - adding options to log any important changes only for a specific segment or landing page - combining data by conversions from GA4 and clicks from GSC If GSC had a pro mode, it would look like this -> https://sitechecker.pro/ Would you like to try it, but don't have an account yet? Comment something, and I'll give you a free trial for 2 weeks (also make sure we are connected).

📊 GA4 Ecommerce Funnel Dashboard — Final Project This interactive dashboard was built using the public GA4 dataset in BigQuery and visualized in Looker Studio. It maps the full user journey — from session start to purchase — and highlights key business metrics: ✅ Visits ✅ Orders Started ✅ Purchases 🔍 Users can filter the data by: Landing page Source, medium, and campaign Device category, language, and operating system 📈 The dashboard includes 5+ visualizations, KPI scorecards, and follows UX best practices (5-second rule, clear labeling, color hierarchy). 🛠️ Tools used: BigQuery · SQL · Looker Studio 👩💻 Created by: Larysa Rehush

A new Power BI blog has been published: We are thrilled to announce that editing semantic models in the Power BI service is now generally available! This milestone unlocks end-to-end Power BI authoring directly in the browser, bringing core modeling parity between the web and Desktop experiences. Yes, that means Mac users can now model in Power BI, no Desktop required!

DAX + UDF = The React of Power BI https://lnkd.in/e2CwfyNm Hello everyone! I want to share with you my latest blog post, which explains how to use UDF and DAX in a "React" style in Power BI. What will you find there? - A complete video walkthrough to start your journey creating "React" like components in Power BI - Sample code for HTML KPI cards and a Table built with DAX-UDF I am a believer that we will soon start finding amazing "React" like components in the Power BI ecosystem! #PowerBI #UserDefinedFunctions #UDF #HTML #UI Daniel Marsh-Patrick, Injae Park, Pascal Kiefer 🥷, BIBB | UX driven BI