Ask buyers what frustrates them about their BI tool and the answer is rarely that it does not work. It is that it needs a specialist: business users can read the dashboards but often depend on an analyst or developer to build or change them — how much depends on the report, the data model behind it and who is building. Close behind are dashboards that look dated or cannot answer a new question without someone rebuilding them, slow loads and refreshes, and numbers that do not reconcile. Cost and licensing is the complaint behind most moves, and it is mostly a Tableau complaint. Most of the evidence comes from Power BI, Tableau and Looker users, and many of them rate their tool well while still naming a weakness — this page reports which complaints come up, not how unhappy users are. For the ratings and the pre-purchase view, see what BI buyers wish they'd known.
The six complaints — graded
Each complaint is classified by how buyers in these interviews typically respond: tolerating the issue, working around it, waiting for the vendor, or reconsidering the contract. These are observed responses, not fixed severity ratings; the same limitation may be manageable for one company and unacceptable for another.
Reports depend on specialists
The most common complaint across all three tools is the expertise needed to create or change reporting. Buyers describe business users who can consume dashboards but depend on specialists for more complex reports, data modelling or customization. The difficulty varies by implementation and task: one company moved from Power BI to Tableau to give its teams more independence, while other Tableau customers still describe needing developers. “To set up Tableau to report things that are actually valuable, you have to be a developer,” says one buyer; another that “Power BI is restricted to only people who know how to use it and manipulate it”; a Looker user describes “a high learning curve just in terms of needing LookML or SQL experience.” Most companies work around it, with a central data team building for everyone, exports to spreadsheets, or a second tool added alongside. It is also the reason most often given for moving to a tool business users can build in themselves.
Named for this: Power BI · Tableau · Looker · Sigma · Domo
Dated, rigid dashboards
Two complaints that travel together: dashboards that look old or are hard to format, and dashboards that cannot answer a new question. One buyer compares the two leaders directly: Tableau's visuals look more professional and up to date, “whereas Power BI seems old and outdated.” A Looker user finds “the visualizations are just a little bit antiquated.” On rigidity, one Tableau user says: “If a report isn't exactly what you need and you need to add a dimension or two metrics, that requires an analyst to go in.” Power BI draws this complaint most, usually by comparison with Tableau, whose visuals are the reference point. Buyers live with it, rebuilding charts in slides or exporting to spreadsheets.
Named for this: Power BI · Tableau · Looker
Slow to load and refresh
Long loads, refresh lag and time-outs. “Oftentimes it takes so long to load that I get distracted and forget what I'm doing while I'm waiting,” says one Power BI user. Looker users raise it more than anyone, and often tie it to their own builds: “The more complex the builds on top of Looker, the slower it gets.” Buyers mostly tolerate it, and many blame their data volume or modelling as much as the tool; it appears in moves, but never as the only reason.
Named for this: Looker · Tableau · Power BI · Sigma
Numbers don't reconcile or aren't trusted
The complaint about whether business users can rely on the numbers they see. Buyers describe dashboards that disagree, users querying the wrong data, and connections that do not always work as expected. The underlying causes vary: some problems involve report definitions or configuration, while others originate upstream, in data quality, modelling or governance. One buyer worries that users “might get results that are inconsistent with reality because they're querying incorrect data”; another says “The trust of Tableau data is sometimes a little leery.” Most buyers address it internally, through approved reports, consistent metric definitions, prepared warehouse tables and tighter governance, rather than by replacing the BI platform.
Named for this: Power BI · Looker · Tableau · Domo
Cost and licensing
The one complaint that clearly separates the tools: Tableau users raise it far more often than Power BI or Looker users. It takes two forms, the expense itself and seat models that decide who is allowed to build. “We moved away from it because the license costs are way too costly for us,” says one former Tableau user; another weighs Tableau against “the included cost for Power BI on our Microsoft license.” Cost is the reason named most often in completed moves in these interviews — most of them away from Tableau — and it is behind several planned ones.
Named for this: Tableau · Sigma · Looker · Power BI · Domo
AI and natural-language analysis fall short of expectations
An emerging complaint from buyers who want to ask questions in plain English rather than depend on predefined dashboards or specialist-built reports. Some describe the AI functionality available to them as limited or behind their expectations, while others are exploring AI tools connected directly to their warehouses or notebooks. These are buyer experiences with particular implementations, not an assessment of every vendor's current AI capabilities. “Any of these tools should be very accessible to play around with through natural language,” one buyer says; another that “Looker is still very far behind in what could be considered a good implementation of AI capabilities.” Tableau users raise it most. Some buyers wait for their BI vendor to improve its AI features; others experiment with notebook tools, internal assistants or AI capabilities in their data infrastructure.
Named for this: Tableau · Power BI · Looker · Sigma
Who leaves, and why? Completed moves in these interviews run in both directions between Tableau and Power BI, and some companies move to newer cloud tools; one went from Tableau to Power BI and then decided on Looker. Cost and licensing is the reason named most often, followed by wanting something business users can build in themselves and consolidating onto one vendor's ecosystem — most often Microsoft's. More companies are planning or evaluating a move than have made one, and most complaints are never acted on at all.
Want this checked against your own stack?
Start my checkup →The stories behind the complaints
The counter-current: the tool is often not the problem
More than in most categories, BI buyers put part of the blame on their own data. “I don't necessarily blame the tool. That could be on our end too,” says one Power BI user; a Tableau user says “it's not a matter of Tableau weakness, it's a matter of data weakness because frankly, what goes in comes out.” Trust, where it exists, attaches to the BI layer: “We really do trust our Power BI numbers, starting with our CEO.” Vendor support, a top complaint in other categories, barely comes up. And the incumbents have staying power of their own: Power BI is often bundled with the company's Microsoft contract — “So Power BI is already included in what we pay for” — while Tableau is held in place by years of built dashboards and contracts already running. Before you conclude the tool is the problem, separate what the vendor controls — speed, visuals, AI, price — from what depends on your warehouse, your data model and who is allowed to build.
What this means — whether you run one or you're buying one
If you already run a BI tool, most complaints are worked around rather than acted on: buyers centralize building in a data team, export to spreadsheets, push modelling into the warehouse, and wait for AI features. Cost is different, because a licence model that limits who can build is the complaint most likely to end in a move. If you're evaluating one, our recommendations — not practices buyers describe — are to price the seats for the people who will build, not just those who will view; to have a business user, not an analyst, build a report during the trial; to test load and refresh times on your own data volume; and to decide who owns the data model before choosing the tool that sits on top of it. For cost, see what buyers actually pay for a BI tool; for where buyers go, Tableau alternatives, Power BI alternatives and Looker alternatives.
Common questions
What are the most common complaints about BI tools?
Among the Power BI, Tableau and Looker users in these interviews, six complaints recur: building or changing reports often depends on analysts or developers rather than business users; dashboards look dated or cannot answer a new question without an analyst; tools are slow to load or refresh; numbers do not reconcile or data does not connect; cost and licensing; and AI and natural-language features that fall short of what buyers expect. Depending on specialists is by far the most common, though how much varies with the report, the data model and the user. Vendor support, a top complaint in other categories, barely comes up.
What are the biggest complaints about Power BI?
Power BI users in these interviews most often say it needs someone trained in the tool to build anything useful, and that its visuals look dated and are hard to format, usually by comparison with Tableau. Slowness and numbers that do not reconcile come up too. Cost rarely does: buyers often note it is already included in their Microsoft contract, which is also a common reason it stays.
What are the biggest complaints about Tableau?
Tableau users in these interviews most often say it needs a developer to build valuable reports, and they raise cost and licensing far more often than Power BI or Looker users do. They also say it is slow at times and that its natural-language and AI features fall short of what they want. Its visuals, by contrast, are the reference point other tools are compared against.
What are the biggest complaints about Looker?
Looker users in these interviews most often say it needs LookML or SQL skills to build anything, and they raise slowness more than users of any other tool here, often linking it to how complex the models built on top of it have become. Dated visuals and numbers that do not reconcile come up too.
Why do companies switch BI tools?
In these interviews cost and licensing is the reason named most often in completed moves, followed by wanting something business users can build in themselves and consolidating onto one vendor's ecosystem. Moves run in both directions between Tableau and Power BI, and some companies move to newer cloud tools. Most complaints, though, are worked around rather than acted on: many buyers blame their own data and modelling as much as the tool.
Get a checkup on your tech stack
A physical for your marketing stack, built from hundreds of verified buyer interviews: what matters most to you, a shortlist for your stack, and your next steps.
Start my checkupFree · no password · your answers are anonymized before they inform research like this.