Tablumo demos — step by step

See Tablumo at work

Click-through demos, one use case each, showing Tablumo analyzing sample data. Click any screenshot to enlarge it; use ← → to step.

16 steps

Sales — pipeline analysis

From raw CSV to an interactive pipeline board: find the $4.75M whale skewing every average, name the rep whose deals keep slipping — then download the data and reproduce all of it yourself.

Play the demo → 16 steps

Marketing — campaign effectiveness

From a raw lead export to a live ROI board: name the $169k campaign that returned $4,650, vindicate the $2,400-a-seat dinner seats, and crown the $1,280 content guide that out-earned everything — then reproduce all of it yourself.

Play the demo → 16 steps

Finance — invoice & spend analysis

From a raw invoice export to a live payment-audit board: catch the 1,874,000 fat-finger that faked a #2 vendor, stop a double-imported batch before its 183,950 goes out twice, and put a number on the supplier paid 37 days late — then reproduce all of it yourself.

Play the demo →

In development

Operations — supplier delivery performance

On-time-rate trends, a degrading supplier and lead-time distributions.

In development

Customer support — ticket queue health

A backlog spike after a release, clustered SLA breaches, resolution-time outliers.

In development

HR — attrition & hiring funnel

Headcount trends, attrition clustered in one department, an interview-stage bottleneck.

In development

E-commerce — sales & returns

Revenue by category, a return-rate outlier product, seasonal patterns.

The sales, marketing and finance demos above are complete — 16 steps each, with downloadable datasets and guided do-it-yourself scripts. The remaining four use cases ship one by one; progress is tracked on the roadmap.

Try it yourself

Same data, same app, your machine. Download the exact CSV from the demo, open Tablumo, and follow the script — the planted surprises are all still in there.

Open Tablumo Download the dataset (1.4 MB CSV)

Drag sales_pipeline_demo.csv onto the Tablumo window (or click browse). When the “how do you want to start?” banner appears, pick Create from scratch.
You're in the Data view: 12,000 rows. Check the total at the bottom of the Amount column — that's your whole pipeline.
Click the Amount column header twice to sort descending. Look at row one. How much bigger is it than row two?
Open the Chart view in the left sidebar. Tablumo configures the first chart for you — total Amount by Owner. Who leads, and by how much?
Set X (group by) to Stage and Aggregation to Max. One stage towers over the rest — which one, and why?
Switch Chart to Histogram and set Column to Amount. Notice how one outlier stretches the whole x-axis and squashes 11,999 deals into the first bar.
Click and drag horizontally across the empty right half of the histogram, then release. The drill-down opens the rows in that range — how many are there, and whose deal is it?
Switch Chart back to Bar (Owner / Sum / Amount) and click the tallest bar. The panel shows every deal behind that number — find the big one in the total.
Back in the Data view, click + Add filter…, pick Stage, and untick everything except Proposal and Negotiation. Watch the row count change honestly.
Click the Close Date header once (ascending). The deals at the top have close dates in the past but are still open. Whose name keeps repeating?
Open the Dashboard view. The starter board builds itself — and your stage filter is still applied (see the chip top right).
Click any bar on the owner chart. Every other tile follows, and a chip names your selection. Click the chip's × to clear it.
Click + Add summary, then the tile's Configure table button, and group rows by Owner. You've built a live leaderboard.
Press Ctrl/⌘-K and type “chart”. Everything in the app is reachable from the palette.
Open Analyses (bottom of the sidebar), name your workspace, and save. Reload the page — then reopen it exactly as you left it.
That's the whole loop: raw CSV → sorted truth → charts → drill-downs → filtered focus → interactive dashboard. All local, all reversible, no formulas.

Tip: the dataset is synthetic, generated with planted, discoverable stories — a whale deal, slipping close dates — so the surprises above are really in the file you just downloaded. No real companies, no real people.

Try it yourself

Same data, same app, your machine. Download the exact CSV from the demo, open Tablumo, and follow the script — the planted surprises are all still in there.

Open Tablumo Download the dataset (1.1 MB CSV)

Drag marketing_campaigns_demo.csv onto the Tablumo window (or click browse). When the “how do you want to start?” banner appears, pick Create from scratch.
You're in the Data view: 12,000 leads, one row per lead — channel, campaign, cost, and what each lead became.
Click the Cost per Lead column header twice to sort descending. Which campaign owns every top row, and what does a single one of those leads cost? Check the column footer for total spend while you're there.
Open the Chart view in the left sidebar. Tablumo configures the first chart for you — total Cost per Lead by Channel. Where does the budget actually go?
Flip Y (measure) to Revenue. Watch the channel ranking — does the biggest spender stay on top?
Now set X (group by) to Campaign. Two surprising campaigns lead the revenue ranking — note them both, they star in the next steps.
Switch Chart to Histogram and set Column to Cost per Lead. Almost everything lives under $200 — so why does the axis run past $2,400?
Click and drag horizontally across the empty right end of the histogram, then release. The drill-down opens the rows in that range — how many are there, and which campaign are they all from?
Switch Chart back to Bar, set it to Campaign / Sum / Cost per Lead, and click the tallest bar. That's the year's biggest campaign budget — keep its total in mind and compare it with the revenue chart from step 6.
Back in the Data view, click + Add filter…, pick Status, and untick everything except Customer. How many of 12,000 leads actually became revenue?
Click the Cost per Lead header once (ascending). Leads that cost under a dollar sit next to five-figure revenue — which channel keeps repeating?
Open the Dashboard view. The starter board builds itself — and your customers-only filter is still applied (see the chip top right).
Click any bar on the channel chart. Every other tile follows, and a chip names your selection. Click the chip's × to clear it.
Click + Add summary, then the tile's Configure table button, group rows by Campaign, and set the value column to Revenue. There's your live campaign leaderboard — who's on top, and what did that campaign cost?
Open Analyses (bottom of the sidebar), name your workspace, and save. Reload the page — then reopen it exactly as you left it.
That's the whole loop: raw lead export → spend vs revenue → outliers named → flop sized → sleeper crowned — on a live, filterable board. All local, all reversible, no formulas.

Tip: the dataset is synthetic, generated with planted, discoverable stories — an expensive flop, an organic sleeper hit, gold-plated dinner seats — so the surprises above are really in the file you just downloaded. No real companies, no real people.

Try it yourself

Same data, same app, your machine. Download the exact CSV from the demo, open Tablumo, and follow the script — the planted surprises are all still in there.

Open Tablumo Download the dataset (1.3 MB CSV)

Drag supplier_invoices_demo.csv onto the Tablumo window (or click browse). When the “how do you want to start?” banner appears, pick Create from scratch.
You're in the Data view: 12,000 invoices, one row per invoice — supplier, category, amount, due and paid dates. Check the total at the bottom of the Amount column: that's your year of spend.
Click the Amount column header twice to sort descending. Look at row one, then at the rows right under it — same supplier, same amount, one digit apart. Suspicious?
Open the Chart view in the left sidebar. Tablumo configures the first chart for you — total Amount by Supplier. Note who's #1 and #2; you'll want that ranking later.
Set X (group by) to Category. Where does the money actually go?
Switch Chart to Histogram and set Column to Amount. Almost everything lives under 100k — so why does the axis run to 1,900,000?
Click and drag horizontally across the empty right half of the histogram, then release. The drill-down opens the rows in that range — how many are there, and from which supplier?
Switch Chart back to Bar (Supplier / Sum / Amount) and click the second bar. Count the invoices in the panel and compare their amounts. What would this supplier's total be without the odd one out — and where would that put them in the ranking from step 4?
Back in the Data view, click + Add filter…, pick Category, and tick only Office Supplies. Sort by Amount descending. Read the top rows in pairs — compare the references, dates and amounts. How much money is about to leave twice?
Remove the category filter (the × on the chip), then filter Payment Status to Paid late. How many of 12,000 invoices were paid late?
Open the Chart view: set Bar, X = Supplier, Aggregation = Average, measure = Days Late. One supplier towers — by how much?
Open the Dashboard view. The starter board builds itself — and your paid-late filter is still applied (see the chip top right).
Click the tallest bar on the supplier chart. Every other tile follows, and a chip names your selection. Click the chip's × to clear it.
Click + Add summary, then the tile's Configure table button, group rows by Supplier, and set the value column to Amount. There's your late-payment league table — who tops it, and does the total match the chip?
Open Analyses (bottom of the sidebar), name your workspace, and save. Reload the page — then reopen it exactly as you left it.
That's the whole loop: raw invoice export → sorted truth → outlier named → duplicates caught → lateness quantified — on a live, filterable board. All local, all reversible, no formulas.

Tip: the dataset is synthetic, generated with planted, discoverable stories — a slipped-digit invoice, a double-imported batch, one chronically late-paid carrier — so the surprises above are really in the file you just downloaded. No real companies, no real people.