Data and ownership
- Does our data have to be copied into your platform? Copies mean another place to secure and another version of the truth. Prefer tools that can query your systems in place.
- Can it run on our own servers? For regulated or privacy-sensitive businesses, a cloud-only answer may end the conversation. Ask whether the self-hosted product is the full product.
- Does the product send usage data home? Ask plainly. The honest answer should be a plain yes or no.
People and permissions
- Can we control which rows and which columns each person sees? "Sales sees their own region; nobody but finance sees salaries" is a normal requirement, not an enterprise add-on.
- Do those rules hold in downloads, scheduled reports, embedded views and AI answers? A rule that only applies on the dashboard is not a rule.
- Is there an audit trail? Who viewed, changed and shared what.
Speed and scale
- What happens when our tables reach tens of millions of rows? Ask for the performance story and, crucially, whether it costs extra.
- Is there a separate "capacity" or "engine" charge? Some pricing looks simple per seat until data grows.
Building and living with it
- Can a non-technical person build a dashboard? Watch someone from your team do it in the trial, not the vendor.
- How are metrics defined? "Revenue" should be defined once and used everywhere; otherwise every dashboard drifts.
- Will you build the first dashboards for us? Some vendors offer a done-for-you service; others point you to consultants.
- What does leaving look like? If your data never left your systems, leaving is easy. Ask anyway.
Our answers, in order: no copies required; yes, the full product self-hosts; no call-home; yes to row and column rules; yes, everywhere; yes; the engine ships with every plan; no capacity charge; yes; defined once; yes, our BI developers will; and your data was never ours. The longer version.