Bubble Or Not?
Is a stock's price running ahead of its fundamentals?
Is a US stock price running ahead of its fundamentals? Overlay any US-listed stock price against Revenue, Net Income, or EPS on one chart | verified SEC and Yahoo Finance data, with a plain-English auto-insight.
How it works
- Enter a ticker | any US-listed stock (AAPL, NVDA, or a smaller SEC filer). Unknown tickers are fetched live.
- Pick a fundamental | Revenue, Net Income, EPS | whatever the company files with the SEC.
- Read the overlay | price and the fundamental share one chart, plus an auto-insight in plain English.
Try an example
More tools on the Data Apps page, including the Stock Screener.
US-listed stocks (SEC filers). A non-US ticker without SEC fundamentals shows price only | never invented numbers.
What it reads
This report needs no account of yours. It runs on data autario already holds, so there is nothing to connect before you can read it.
- Public autario data: daily share prices plus the figures companies file with the SEC.
The questions it answers
These are the questions the report is built to answer. Each one is answered with stored numbers and a definition you can read, never with a guess.
- Is NVDA a bubble?
- How does AAPL P/E compare to its history?
What it computes, and which part of the report returns it:
- Valuation vs own history
- Where the chosen fundamental (e.g. P/E) sits in its own multi-year history: percentile, range, median, verdict.
- Multiple decomposition
- Exact split of a ratio move into the price move vs the earnings move.
- Price vs fundamental
- Share price overlaid against one SEC-reported fundamental over time.
How you use it
Three ways in, one set of numbers behind them:
- In the browser: open the app, pick a period, read the report. Nothing to model, nothing to build, no query to write.
- Through an assistant: the report is an MCP tool called bubble_or_not, so the numbers arrive inside the conversation instead of as a file to open.
- Over the REST API: the same numbers, as JSON, with an API key in a header. The datasets behind this one are part of the public catalogue, so you can go past the report and query them directly, filtered and ordered, or stream one as CSV.
This report has no read-only link of its own yet. What you can share is anything you build from its numbers: a chart has its own public address, and the underlying table can be exported.
What it never does
The limits below are the design, not a disclaimer, and several of them are enforced by a test that refuses the build rather than by a promise.
- It never writes back. Nothing here posts, spends, cancels, publishes or edits a single field in any account you connect. The requests a connector may make are a declaration in one registry file and a test reads that declaration, so a write verb cannot reach a provider even by accident.
- It never invents a number. A figure the provider did not publish stays empty, with the reason next to it, and an empty answer is recorded as an empty answer. A ratio with no denominator is empty rather than zero or infinite.
- It never mixes your data with somebody else's. Your tables are private by default, they are never merged into the public catalogue, and a report resolves only what the account asking for it owns.
- It never silently reshapes a table. When a provider renames or drops a column, the change is written to the dataset changelog and the table keeps its address, so a report that reads it can see what moved.
How often the numbers move
The report is computed when you open it, from stored runs. Nothing is re-derived and nothing is estimated, so the same window always produces the same numbers.
How to cite this
The numbers on this report come from named publishers and keep their provenance, so a report you hand on can be checked. A line that works in a document, a deck or a footnote:
Cite as: autario.com, Bubble Or Not? (accessed 2026).
Questions about the data
- Can autario change anything in the accounts I connect?
- No. Every source this report reads is read-only by construction, and a test refuses the build if a write verb ever appears in a connector's declaration. Nothing posts, spends, cancels or edits.
- Can an AI assistant read this report for me?
- Yes. autario speaks MCP, and this report is a tool called bubble_or_not. An assistant connected to your account calls it and gets the same numbers you see, already computed, in the parts it asked for. Nothing is re-derived on the way and no model writes any figure.
- What does it cost to read this report?
- Reading a report you already have data for costs nothing extra. The current plans and what each one includes are on the pricing page, which is the one place that carries the figures.
Read next
- Live data in Power BI and Excel: Point Power BI or Excel at a URL and keep your existing reports.
- Public data your agent can actually use: World Bank, IMF, Eurostat, OECD, WHO, FRED, ECB and SEC data normalized to one shape, addressable by entity code, over.
- All data apps: Every app that runs on autario, on the same shelves as the connectors.
- The connector catalogue: Every service you can connect today, each with its own page.
- How the platform works: What happens to data between a connector and a report: normalization, storage and the read paths.
- Power BI and Excel: The one link that lists all of your tables, and the two headers a refresh needs.
- The API and the MCP tools: The REST API and the MCP tools, with the exact calls.
- Security and data processing: Where this runs, what is encrypted and who can see what.
What autario is
autario is the reporting layer that is ready once you connect: your tools and public datasets land in one normalized layer that keeps its history, data apps run on top, and your AI, Power BI, the API or CSV read the same numbers. Underneath is data app infrastructure.
Best for: agencies that report across Meta Ads, Google Ads, GA4, Search Console and LinkedIn for several brands; merchants selling on Shopify and Amazon who want orders, ad spend and payouts in one table; teams tracking OpenAI and Anthropic spend; brands tracking how often ChatGPT, Claude and Perplexity mention them.