Dashboards Are Not Dying. The Questions Are Changing.

Dashboards are not dying. They are good at one job and will stay good at it: answering a fixed, known question you check again and again, like today's revenue or this week's signups. What is changing is the other kind of question, the open-ended "why did this move," which used to mean filing a ticket and waiting for an analyst. That part is moving to an AI agent.

Key takeaways

• A dashboard answers questions someone already thought of and built a chart for. It is fast and reliable for those, and it will keep that job.
• The question that is shifting is the ad-hoc "why did this number move," the kind that today turns into a ticket and a two-day wait for the data team.
• Agentic analytics is an AI agent that investigates that open question end to end, then shows its work and how sure it is, so you can check it.
• Most teams need both, and Sundial gives you both. It has dashboard capabilities for the metric everyone watches, plus the ask-anything agent for the question nobody pre-built a chart for.

---

When should you still use a dashboard?

A dashboard is a printed map: accurate, but it only shows the routes the mapmaker drew. Someone picked the metric, set the filters, and shipped the chart. For a fixed KPI everyone watches, that is the easy, right place to work. You want today's revenue to load in one glance, the same way every time, with no surprises. A dashboard does that. Sundial has dashboard capabilities too, so the fixed views live in the same place as the agent.

What is changing is that more of the work moves off the chart and onto asking your data directly, in Slack or Teams, the way you would ask a human analyst, instead of hunting for the right dashboard.

The trouble starts when your real question is slightly different from the one the dashboard was built for. Say you want weekly active users among U.S. iOS users who are male and between 13 and 17. That can take four separate dashboards to stitch together, and there is no fast way to drill in. So you file a request, and you wait.

Which question is moving to AI?

The expensive question was never "what is the number." It was "why did the number move." A dashboard shows that conversion dropped. It does not tell you the drop is concentrated in one country, on one platform, after one product change. Finding that out is detective work: pull data from scattered systems, write SQL to test one hypothesis at a time, rule things out, repeat. That detective work has a method, and an analytics playbook encodes it so the agent runs the same rigorous steps every time.

At OpenAI, a sudden drop in conversion used to mean one to two full days of an analyst's time doing exactly that. With Sundial, the same investigation runs automatically and surfaces whether the decline ties to a particular country, platform, or user segment in seconds rather than days. The dashboard still shows the drop. The agent answers the "why."

The dashboard tells you the number changed. The slow, expensive part has always been finding out why, and that is the part moving from a ticket to an agent.

What is agentic analytics?

Agentic analytics is an AI agent that takes one open data question and investigates it from start to finish, the way you would brief an analyst and wait two days for the answer. You ask in plain language. The agent plans an investigation, runs the queries itself, checks its own work, revises, and hands back an answer with the reasoning attached. Sundial is an agentic analytics platform that does this on a company's own data. For business users asking questions, it is read-only by default: they get answers and cannot change the data.

It is not a dashboard you read, and it is not a chatbot that turns one sentence into one SQL query. One query answers a shallow question like "what was revenue last week." A question like "why did revenue drop last week" is a dozen queries plus the judgment to know which ones matter. Put plainly: text-to-SQL answers one query, and agentic analytics runs the whole investigation.

Here is how the three tools split, side by side:

The piece that keeps the agent grounded is a semantic layer, the core of the broader context layer. That is the place that holds the official definition of each metric, how your tables relate, and which source is the truth. Without it, the model guesses what "active user" means. With it, the answer reflects how your company actually defines things. The work splits across four agents, each doing a job a human analyst does without thinking: Quality (is the underlying data fresh and complete enough to trust), Modeling (what the metrics and entities actually mean), Analysis (the chain of queries that gets to why), and Storytelling (turning the result into something a decision-maker can act on).

Why is a fast answer not enough?

A wrong answer delivered with confidence is more dangerous than a slow one, because someone makes a decision on it before anyone catches the mistake. A plain text-to-SQL tool can write a query that runs cleanly, returns a confident number, and is measuring the wrong thing. You have no way to tell. That is why "it answers fast" is not enough on its own.

A serious agent has to be checkable. What it can do depends on who is using it. For business users, the consumers asking questions, the agent is read-only by default, so they get answers without changing your data. Data practitioners use the same agent to do more: it is the one tool that can comprehensively help model the tables and run data-quality checks, so the team that owns the data can build and maintain the context layer everyone else relies on. Either way, it shows its work, the steps and the queries, so a data team can audit how it got there. It gives a confidence signal, so a decision-maker knows a solid answer from a rough estimate. And it leaves an audit trail, so the data team can see what the agent did across the company. You are judged on two things at once: how good the answer is, and how easily you can tell whether to believe it.

Will AI replace BI tools like Tableau, Looker, and Power BI?

No. The point is not to throw out dashboards. It is to stop using them for the question they were never good at. A dashboard view is the right place for a fixed KPI everyone watches, or for numbers that have to tie out exactly, like a financial close. An agent is the right tool for the ad-hoc, open-ended question that today becomes a ticket. This is not a choice between a dashboard and an AI analyst, and it is not "use a competitor's dashboard for the fixed KPIs and Sundial for the rest." Sundial does both. You get the fixed dashboard views and the ask-anything investigation in one place.

Mighty Networks shows the split in practice. Their Hosts, the people who run communities on the platform, earned over $500M in 2025, and the company had built hundreds of Looker dashboards. The dashboards had the answers. The team could not surface them fast enough to act. Sundial brought three insights that reset their product roadmap, like which Host behaviors in the first 30 days predict whether a community is still thriving a year later. As founder and CEO Gina Bianchini put it, "no CEO should be a victim of dashboard sprawl." Her team went from asking "where is that dashboard?" to prototyping what to build next.

That is the shift. Not the death of the dashboard, but a change in where the hard questions go. The fixed ones stay on the chart. The open ones move from a queue to an agent that does the legwork and shows you how it got there.

This changes the analyst's job too. It does not remove the human, but it transforms the role, and a team likely needs fewer analysts. The agent can do much of what analysts used to spend their days on: the repetitive pulls and the first-pass investigation. So the work shifts from reactively answering query requests to architecting the context, defining the metrics, the relationships, and the source of truth, so that everyone's questions get high-quality answers. Humans stay in the loop on judgment and the highest-stakes calls.

Common questions

Are dashboards dying? No. Dashboards still answer fixed, recurring questions well, like today's revenue. What is moving to AI is the open-ended "why did this number move" question that used to take an analyst hours.

What is replacing the ad-hoc data request? An AI agent. The "why did churn spike in this segment" question that used to become a ticket is moving to agentic analytics that investigates it directly.

Is agentic analytics the same as text-to-SQL? No. Text-to-SQL turns one sentence into one query and gives you no way to know if it is wrong. Agentic analytics runs a multi-step investigation, shows its steps, and signals how sure it is.

How does a data team audit AI-generated analysis? A serious agent shows its work, the steps and the queries, and leaves an audit trail. For business users asking questions, Sundial is read-only by default, so they get answers without changing your data. Data practitioners use the same agent to model the tables and run data-quality checks.

Is the agent read-only? It depends who is using it. For data consumers, the business users asking questions, it is read-only by default: they get answers and cannot change the data. Data practitioners use the same agent to do more, including modeling the tables and running data-quality checks, so they build and maintain the context layer the consumers rely on.

When should a company still use a dashboard instead of AI? A dashboard is the right place for a fixed KPI everyone watches, or for numbers that have to tie out exactly, like a financial close. You do not have to choose: Sundial has dashboard capabilities and the ask-anything agent in one place.

Does agentic analytics replace data analysts? No, but it transforms the role, and a team likely needs fewer analysts. It can do much of the repetitive pulls and first-pass investigation analysts used to spend their days on, so the job shifts from answering query requests to architecting the context, the metrics, the relationships, and the source of truth, while humans stay in the loop on judgment and the highest-stakes calls.

See what that looks like at Sundial.