Turn analysis into shared business knowledge.
Seneca lets the people who understand the business build live analytics with their own AI assistant.
It keeps what the team learned, along with the evidence and corrections behind it, and gives that knowledge to the next person and their AI before new work begins.
Talk with usThe problem
Business understanding gets people only part of the way to an answer.
A marketing manager understands her customers, her campaigns, and the decision in front of her. She wants to compare how often customers buy again across acquisition channels, giving each customer the same amount of time to come back, and to see whether discounts change the answer.
Today that question needs SQL, cohort calculations, and a specialist who knows the dashboard tool.1 The technical work, not her understanding, is what stands in the way.
AI assistants can now do much of that technical work. But the analyses they produce often end up as files in a chat or a folder, with nowhere to live, no live data, and no record of what was assumed. When a colleague asks the same question three months later, the earlier work and the corrections that improved it are lost.2
How it works
People clarify what they mean. The AI handles the mechanics.
Ask with the assistant you already use
The marketing manager asks her AI assistant for the comparison. Seneca gives it the company's established definitions, the right data, warnings about incomplete periods, and any related analysis the team has already done.
Settle the questions that matter
The assistant asks one consequential question: should a customer count toward the channel that first brought them in, or the most recent one? It explains that it will compare each customer's first 90 days. She decides; it builds.
Publish a live analysis, not a frozen file
The result gets a stable address. Readers see current results under their own data access, how fresh the data is, the assumptions in plain language, and whether anyone has reviewed it. When data changes, a proposed rewrite becomes a new version for review instead of silently changing an approved conclusion.
Start the next question from what the team already learned
Months later, a finance analyst asks a related question using a different AI tool. Before it queries anything, Seneca hands his assistant her analysis, its assumptions, and the correction her team made. If marketing and sales explain the result differently, both explanations stay side by side with their evidence, rather than the latest opinion becoming the official answer.
What a reader sees
Every figure says what it assumes. Every trust signal stands on its own.
A Seneca analysis reads like an audited statement. Figures come first, each assumption has a numbered note beside the figure it affects, and the source, freshness, checks, and human review are shown separately, never blended into one score.
| Acquisition channel1 | Customers | Bought again within 90 days2 |
|---|---|---|
| Search | 4,210 | 31.4% |
| Social | 2,880 | 22.9% |
| Referral | 1,140 | 38.2% |
| All channels | 8,230 | 29.4% |
- 1Customers are counted in the channel that first brought them in, as decided by the marketing team. A later correction moved partner sign-ups out of Referral.
- 2Every customer gets the same 90 days. Customers who joined in the last 90 days are left out because their window is not complete.
- Source
- Orders, certified
- Data as of
- Today, 6:00 a.m.
- Checks
- 2 of 2 passing
- Review
- Reviewed
The channel assignment and the 90-day window are appropriate for comparing channels. This analysis does not show why Social customers return less often.Reviewer, Finance ยท October 6, 2026
How it fits
Seneca works alongside the tools you already have.
- Your data stays put
- Queries run in your existing warehouse, such as Snowflake. Seneca does not copy your datasets into a system of its own.
- Your AI assistant
- People use the assistant they prefer. Seneca supplies it with the company's definitions and past work; it does not replace it.
- Your existing BI tools
- Start with one recurring analysis that is awkward to build today. Nothing needs to be replaced to begin.
- Your access rules
- Each reader sees results under their own data access, and explanations carry access controls just like the data behind them.
Where we are
Early, and building with analytics leaders.
Seneca is a working prototype built on public sample data. It is not yet a hosted service, and we have no customers to quote. We would rather say that plainly than suggest otherwise.
We are looking for analytics and business leaders who are already seeing AI-built analyses appear in their teams and want a better place for them to live. If that sounds like you, we would like to hear what you are seeing.
Talk with usFounder
Jake Minette
Founder, Seneca Analytics
Jake has spent 20 years in business intelligence, analytics, and data science, turning data into stories that help people make decisions. He started Seneca after watching AI assistants produce complete, useful analyses that had no good place to live and no way to carry what people learned into the next question.