--- title: 'Wait, you don't have a semantic layer yet?' AI... description: Metabase surveyed 73 data people during its AI Analytics Week and found 59% already use AI with their data, while only 4% have ruled it out. 44% want AI fully... canonical: https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: 'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse | daily.dev og:description: Metabase surveyed 73 data people during its AI Analytics Week and found 59% already use AI with their data, while only 4% have ruled it out. 44% want AI fully... og:url: https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj og:image: https://api.daily.dev/og/posts/bHZYzThsj.png og:image:alt: 'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # 'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse **[Metabase](https://daily.dev/sources/metabase)** · 10 min read · 0 upvotes · 0 comments ## Summary Metabase surveyed 73 data people during its AI Analytics Week and found 59% already use AI with their data, while only 4% have ruled it out. 44% want AI fully self-hosted, and 58% want to self-host or bring their own model. Correctness is the top blocker: 48% worry about wrong numbers and 37% can't tell how an answer was derived. 31% of poll voters had never heard of a semantic layer, 54% have something ad hoc, and only 15% call theirs mature. Demand for self-hosting runs inverse to AI experience, dropping from 69% among those who haven't started to 30% among daily users. Half of all respondents want AI agents, rising to 78% among regular users. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What percentage of data teams are already using AI with their analytics data? 59% of surveyed data people are already using AI with their data, based on a poll of 73 respondents during Metabase's AI Analytics Week. Only 4% had decided against adopting AI analytics, while 22% hadn't started yet. The sample was self-selected, anonymous, and skewed technical, so it isn't a scientific representation of the broader market. _Teams weighing when to adopt AI analytics can compare adoption patterns like these on daily.dev._ ### Why do AI-generated analytics answers give wrong numbers? Wrong numbers from AI analytics tools are often caused by missing context about which data to trust, such as having both an 'accounts_v2' and 'accounts_final' table with nothing telling the AI which one the company actually bills from. A semantic layer, a shared set of models, metrics, and definitions, prevents this by telling AI which data is correct and what it means. Without it, confident-sounding wrong answers become more common as a company and its data grow. _Developers debugging unreliable AI analytics output can track semantic layer practices on daily.dev._ ### Do teams adopting AI analytics prefer self-hosting or cloud deployment? Preference for self-hosting is strongest among teams that haven't started using AI analytics yet, at 69%, and drops to 30% among teams already using AI regularly, based on a survey split by experience level. Overall, 58% wanted to self-host or bring their own model, and 58% of poll voters said their data can't leave their network at all. _Engineers deciding between self-hosted and cloud AI deployments can weigh findings like these on daily.dev._ ## Similar posts on daily.dev - [The bottleneck moved: how to build a semantic layer for AI Analytics](https://daily.dev/posts/the-bottleneck-moved-how-to-build-a-semantic-layer-for-ai-analytics-appzsz9ve) · Metabase · 0 upvotes · 0 comments - [Building or Buying: The Agentic Analytics Dilemma](https://daily.dev/posts/building-or-buying-the-agentic-analytics-dilemma-jei830uta) · JetBrains · 0 upvotes · 0 comments - [Connecting AI agents to enterprise knowledge](https://daily.dev/posts/connecting-ai-agents-to-enterprise-knowledge-wjpvnq4xg) · MIT Technology Review · 0 upvotes · 0 comments - [Most Semantic Layers Were Built for BI: What a Semantic Layer for AI Requires](https://daily.dev/posts/most-semantic-layers-were-built-for-bi-what-a-semantic-layer-for-ai-requires-bpd6ufoso) · SD Times · 1 upvotes · 0 comments - [Essential ingredients for enterprise AI success](https://daily.dev/posts/essential-ingredients-for-enterprise-ai-success-7yea3ryew) · Stack Overflow Blog · 2 upvotes · 0 comments --- Tags: [#data-engineering](https://daily.dev/tags/data-engineering), [#bi](https://daily.dev/tags/bi) [View this post on daily.dev](https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse","url":"https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj"},"datePublished":"2026-10-28T13:00:10.393Z","dateModified":"2026-10-28T14:58:10.132Z","description":"Metabase surveyed 73 data people during its AI Analytics Week and found 59% already use AI with their data, while only 4% have ruled it out. 44% want AI fully...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/302375524cf6dcc36b3a84658cccf560?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/302375524cf6dcc36b3a84658cccf560?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Metabase","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Metabase","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/97c4b2294a914263b4f3b4a82a5f896b","url":"https://daily.dev/sources/metabase"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"data-engineering,bi","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Metabase","item":"https://daily.dev/sources/metabase"},{"@type":"ListItem","position":3,"name":"'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/wait-you-don-t-have-a-semantic-layer-yet-ai-analytics-readiness-pulse-bhzyzthsj#faq","mainEntity":[{"@type":"Question","name":"What percentage of data teams are already using AI with their analytics data?","acceptedAnswer":{"@type":"Answer","text":"59% of surveyed data people are already using AI with their data, based on a poll of 73 respondents during Metabase's AI Analytics Week. Only 4% had decided against adopting AI analytics, while 22% hadn't started yet. The sample was self-selected, anonymous, and skewed technical, so it isn't a scientific representation of the broader market. Teams weighing when to adopt AI analytics can compare adoption patterns like these on daily.dev."}},{"@type":"Question","name":"Why do AI-generated analytics answers give wrong numbers?","acceptedAnswer":{"@type":"Answer","text":"Wrong numbers from AI analytics tools are often caused by missing context about which data to trust, such as having both an 'accounts_v2' and 'accounts_final' table with nothing telling the AI which one the company actually bills from. A semantic layer, a shared set of models, metrics, and definitions, prevents this by telling AI which data is correct and what it means. Without it, confident-sounding wrong answers become more common as a company and its data grow. Developers debugging unreliable AI analytics output can track semantic layer practices on daily.dev."}},{"@type":"Question","name":"Do teams adopting AI analytics prefer self-hosting or cloud deployment?","acceptedAnswer":{"@type":"Answer","text":"Preference for self-hosting is strongest among teams that haven't started using AI analytics yet, at 69%, and drops to 30% among teams already using AI regularly, based on a survey split by experience level. Overall, 58% wanted to self-host or bring their own model, and 58% of poll voters said their data can't leave their network at all. Engineers deciding between self-hosted and cloud AI deployments can weigh findings like these on daily.dev."}}]} ```