Thirteen of fourteen models named Microsoft Power BI first on the direct prompt; one named Tableau. Both were named by all fourteen models and Microsoft Power BI carries 76 labels and Tableau 61, so the shares are not directly comparable.
By Microsoft, Vancouver, Canada, founded 1975. Named in four categories this edition.
Named in seven categories this edition.
Share is the count of first choices across the direct, paraphrase, budget and scale prompts over all fourteen models, for a mid-market B2B company; rank is within the category; every quote names the model and the prompt it came from. Both figures come from the dashboards and data visualization page.
Across every category in the November 2026 Edition, Microsoft Power BI and Tableau were named in the same answer 203 times, of the 281 answers naming Microsoft Power BI and the 244 naming Tableau. In those answers Tableau took the first choice five times and Microsoft Power BI ninety-seven.
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | ||||||
| Gemini 3.5 Flash | ||||||
| Perplexity Sonar | ||||||
| Grok 4.20 | ||||||
| Mistral Small | ||||||
| DeepSeek V4 Flash | ||||||
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | ||||||
| Kimi K2 | ||||||
| GLM 4.7 FlashX | ||||||
| MiniMax M2.5 | ||||||
| GPT-6 Luna | ||||||
| Muse Glimmer 30B |
Bold names in an answer are the products the judge labeled a first choice; a model naming several gives each of them that label. The full answer text for every row is in the record.
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Extremely popular and cost-effective, but watch for specialist needs (like building complex models), reconciliation errors, and AI that doesn't always meet hype.” Grok 4.20 · negative prompt · soft negative
“Power BI deeply locks you into Microsoft and can struggle with performance and governance when querying massive, complex datasets” Claude Haiku 4.5 · negative prompt · soft negative
“Power BI, for example, has meaningful differences between Free, Pro, and PPU, and sharing/collaboration can require paid licenses” GPT-5.4 mini · negative prompt · soft negative
“The strongest general-purpose choices are Microsoft Power BI, Tableau, Looker, Qlik, ThoughtSpot, Sigma, Sisense, Domo, Metabase, and Zoho Analytics.” Perplexity Sonar · comparative prompt · first choice
“Microsoft Power BI is the most-deployed mid-market BI platform globally... Power BI is the lowest-friction, cost-predictable default.” Muse Glimmer 30B · direct prompt · first choice
“Default / cost-effective for most mid-sized B2B: Microsoft Power BI. Strong Microsoft integration, Excel familiarity, low entry price” Muse Glimmer 30B · paraphrase prompt · first choice
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of seven in this category shown.
“Tableau documents cases where some composed data-source extracts can't be done on the web and require Desktop, which can be a maintenance or governance burden.” GPT-5.4 mini · negative prompt · soft negative
“Frequently cited for high costs and licensing... Dashboards can feel dated/rigid, require specialists for changes, and AI features underdeliver.” Grok 4.20 · negative prompt · soft negative
“The cost is significantly higher than competitors like Power BI... Tableau may be overkill and too expensive relative to the ROI.” Qwen 3.7 Flash · negative prompt · soft negative
“Tableau - Known for powerful visualization and analytics capabilities, strong community support, and scalability.” Claude Haiku 4.5 · direct prompt · first choice
“Choose Tableau when analysts prioritize sophisticated visual storytelling and exploratory analysis.” Perplexity Sonar · comparative prompt · first choice
“Choose Tableau if visualization quality and handling massive datasets are priorities” Kimi K2 · comparative prompt · first choice
Comparisons are drawn for the top eight products in each category, each against each. The output is the models' output; nothing here is a recommendation by the index.