Pick a replacement by how your team asks questions of data. If analysts write SQL and share charts, you need a query-first tool. If the wider business wants answers without writing queries, look at self-service or AI-assisted analytics, and if you only need live team numbers on a screen, a dashboard tool is enough.
Redash built its following on a simple loop: connect a database, write a query, turn the result into a chart and share it. Plenty of teams still work exactly that way. Others have outgrown it and want people outside the data team to find answers on their own. The eight tools below cover both camps. They're listed in no particular order, A to Z first, then sorted by the job each one does best.
Eight BI tools, A to Z
Every line reflects how the vendor describes its own product.
Cube
Best for: Teams that want metrics defined once on a semantic layer, then reused in AI answers and embedded analytics for staff and customers, per the vendor.
Geckoboard
Best for: Managers who want live KPIs from their everyday business apps shown on office TVs, chat tools, email and mobile, so performance is visible to everyone.
Holistics
Best for: Data teams handing AI-assisted, self-service reporting to the rest of the business; the vendor's line is about asking hard questions and getting reliable answers.
Klipfolio
Best for: People comfortable with spreadsheets who want automated data retrieval and dashboards they can present and share in many ways, says the vendor.
Mode
Best for: Analysts who want SQL, R, Python and visual analysis in one collaborative place, then share what they find with colleagues quickly.
Preset
Best for: Teams that like open source, getting dashboards, embedded analytics and self-service exploration built on Apache Superset, plus a chatbot, according to the vendor.
Sigma
Best for: Anyone building dashboards, apps and AI agents on governed warehouse data, working through chat, a spreadsheet, SQL or Python, as the vendor puts it.
ThoughtSpot
Best for: Organisations wanting governed AI analytics, with agents, automated insights and analytics embedded where people already work, per the vendor.
How they compare at a glance
| Tool | Approach | Good for |
|---|---|---|
| Cube | Semantic layer and analytics | One trusted definition of every metric |
| Geckoboard | KPI dashboards | Showing team numbers on screens and in chat |
| Holistics | Self-service BI | Data teams serving the wider business |
| Klipfolio | Spreadsheet-style dashboards | Spreadsheet users building shareable dashboards |
| Mode | Collaborative analysis | Analysts mixing SQL, Python and charts |
| Preset | Hosted open source BI | Teams who prefer open source dashboards |
| Sigma | Warehouse BI and apps | Spreadsheet-style work on warehouse data |
| ThoughtSpot | Agentic analytics | AI-led answers across the business |
Two questions that matter more than chart types
Every tool here can draw a bar chart. What separates them is how people ask questions of the data in the first place, and how dashboards are shared and kept up to date once they exist. Get those two right and the rest is detail.
How questions get asked
In a query-first tool, someone writes SQL or Python and everyone else reads the result. In a self-service tool, people click, filter or type a question in plain words, and the tool works out the query behind it. AI chat features now turn up across this whole category, so ask each vendor how answers are checked and who decides what a metric means.
How dashboards stay current
A dashboard nobody trusts is worse than none. Check how often data refreshes, whether a broken data source gets flagged, and where people see the numbers: in the tool itself, in Slack or Microsoft Teams, on a TV in the office or by email. Run one real report through your shortlist and see how much work it takes to keep it right for a month.
Who each one suits
Analysts who live in SQL
Mode combines SQL, R, Python and visual analysis in one shared workspace. Preset builds on the open source Apache Superset project, which will feel familiar if you liked writing queries and charting the results yourself.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Business users asking their own questions
Sigma gives people a spreadsheet-style way into warehouse data alongside SQL and Python. ThoughtSpot centres on AI agents and automated insights. Holistics pitches AI-assisted self-service for data teams that want to hand everyday questions to the business.
Teams that need one definition of each metric
Cube is built on a semantic layer, a single place where metrics are defined once and reused everywhere, and it also feeds analytics embedded inside your own product.
Teams that want numbers on the wall
Geckoboard turns data from apps such as HubSpot, Zendesk and Salesforce into live KPI dashboards for TVs, Slack and email. Klipfolio pairs spreadsheet-style formulas with dashboards you can share widely.
If open source matters to you, my list of open source BI tools goes further, and for pure display work see my guide to dashboard software.
Straight advice on choosing a BI tool
Don't start with the demo. Start with the five questions your team asks most often, and who asks them. If it's analysts asking, a query-first tool will feel like home and you'll be productive in days. If it's managers and sales leads asking, you'll want a self-service layer, and you'll need someone to define the metrics before anyone else touches it.
Be wary of judging AI chat features on a vendor's sample data. They look sharp when every column is tidy and named well. Try them on your own messy tables, and see whether the tool shows you the query behind each answer. If it doesn't, nobody will be able to check it, and that's a problem the first time a number looks wrong in a board meeting.
Questions to put to each vendor
- Which of our data sources do you connect to directly, and which need a workaround?
- Can someone who doesn't write SQL answer a new question on their own?
- Where is each metric defined, and who can change that definition?
- How do you flag a dashboard when its data stops refreshing?
- Can we share dashboards with people who don't have an account?
- How do we bring our existing saved queries across?
Do you make business intelligence or dashboard software? Tool makers can ask to be considered for a featured place on this list.
Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.
Frequently asked questions
Is there an open source option similar to Redash?
Yes. Some tools in this category are built on open source projects. Check whether you'd host the software yourself or use the vendor's hosted version, as that changes who maintains it.
Can non-technical staff use a BI tool without SQL?
Many can. Self-service and AI-assisted tools let people filter, click or type questions in plain words. Someone still needs to set up the data and define the metrics first.
What is a semantic layer?
It's a shared set of metric definitions that sits between your data and your dashboards, so revenue or active users mean the same thing in every report.
How do I move my saved queries?
Export your SQL, rebuild the most used charts first, and compare the new numbers against the old ones before you switch anyone over.
Do AI answers in BI tools get things wrong?
They can, especially when metrics aren't clearly defined. Ask how the tool shows the query behind an answer so someone can check it.
