For financial analysts

Reconstruct any chart from a research report

Pull quarterly figures, earnings projections, and sector trends out of analyst PDFs. From screenshot to CSV in under a minute, with the precision your model needs.

Today

What you're stuck on

Sell-side reports hide the table

You can see the chart but not the underlying values. Calling the desk for the dataset every time burns goodwill and slows the model.

Eyeballing pixels for hours

Tracing 8 quarters of competitor EBITDA across a PDF chart is guesswork. The error bands are wider than the differences you're modeling.

MNPI restricts external tools

Compliance won't approve uploading material non-public figures to a third-party server. Most chart-digitizing tools are off the table.

Workflow

From report to working model

  1. 01

    Screenshot the chart from the analyst report or earnings deck. PDF excerpts, slide exports, Bloomberg captures — all work.

  2. 02

    Drag axis lines onto the time-series tick marks. Enter the period labels (Q1 2024 / FY23 / etc.) once and they apply across the categorical axis.

  3. 03

    Export to CSV with one click, or copy directly to your clipboard to paste into Excel. Series stay separately labeled.

Why it fits

Built for the way you actually work

Time-series and quarterly axes

Q1/Q2/Q3/Q4, monthly, annual — the categorical-axis support handles every reporting cadence used in equity research.

Combo and dual-axis charts

Revenue bars + margin line on a secondary axis? Net debt vs. leverage ratio? Calibrate the left and right Y-axes independently and extract each series against its own scale.

Crop out what you can't share

Trim the screenshot down to just the plot before you extract — ticker symbols, deal codenames, and report footers stay out of the image. Manual extraction never leaves your browser; with AI it's the cropped region that's sent, for a single read, and it isn't retained.

Coverage

Chart types you'll handle

Time-series lineQuarterly columnAnnual columnStacked barDual-axisLog-scale

Combo charts with two Y-axes (e.g. revenue on left, margin % on right) are first-class. Set each Y-axis independently — the right-side axis uses the same calibration workflow as the left.

"Used to spend an hour redrawing a single chart in Excel. Now I drop in the screenshot, calibrate the axes, click each bar, and have the CSV in a few minutes."
ER

Equity Research Analyst

Investment Bank

FAQ

Common questions

Does any of my chart data leave my machine?
Manual extraction runs entirely in your browser — the image never uploads anywhere, important for MNPI and confidential research. Use it inside the firewall, on a laptop, doesn't matter. AI auto-extraction is opt-in; only then is the cropped image sent to the AI provider for a single read, and it isn't retained.
Can I crop out confidential labels before extracting?
Yes. The first step lets you crop the screenshot down to just the plot area, so ticker symbols, deal codenames, and report footers never make it into the image. With manual extraction nothing leaves your browser anyway; if you opt into AI, only the cropped region is sent for a single read.
Does it handle quarter-over-quarter labels (Q1 / Q2 / Q3 / Q4)?
Yes. Use the categorical X-axis option and tell it how many ticks the chart has (8 for two years of quarterly, 12 for monthly, etc.). Type the period labels once — they apply across the axis.
Can I extract from Bloomberg or FactSet screenshots?
Yes. Any chart image works — PNG, JPEG, PDF screenshot. Sources don't matter; only the visual quality of the screenshot does.
Does it handle dual-axis combo charts?
Yes. You can calibrate left and right Y-axes independently. The left axis values stay in their data series, right axis values get their own — keeps the two scales separate in your CSV.