← DAX Labs / Case studies
Spreadsheet automation project — interactive case study

Atlas.
An operations workbook that does the work.

DAX built quotes, cash plans and invoices from one controlled workbook engine. Change a value below to inspect how the delivered formulas recalculate every dependent result.

8 × 16 grid 8 functions 3 templates 0 libraries
type in it
01 Delivered operations workbook

Controlled inputs. Reliable outputs.

Inspect DAX's quote, cash-plan and invoice views. Change a quantity, rate or cost to trace the recalculation through the finished system.

AUTOMATIC SHEET change a number, update the totals
engine live
booting
choose an example
A1 value
Editable spreadsheet grid, columns A to H, rows 1 to 16
this template

0cells
0formulas
0.0 mslast recalc
0errors
recalculation order
functions the parser knows
SUMAVERAGEMIN MAXCOUNTIF ROUNDABS + − × ÷ ( ) %> < >= <= = <>
engine log
every value on this grid is computed in your browser by code written for this page · no spreadsheet library, no server
precedent cells selected cell negative · flagged passing check corner tick = formula cell
how we build

A workbook is software, so we treat it like software. Inputs live in one place and are the only cells you touch. Calculations are formulas, never numbers typed over formulas. Every rate, tax percentage and threshold sits in a named assumptions block, so changing the tax rate is one edit and not forty. Validation stops bad input at the cell instead of poisoning the totals three tabs later. And you get the map — which cell feeds what — because a file nobody can maintain is a file you'll pay to rebuild.

02 the offer

Fixed price, fixed days, a file that does the work.

Three scopes. Each one ships a workbook that calculates on its own, refuses bad input, and comes with the notes to keep it alive after handover.

market-calibrated pricing
Working Template
$792 days
  • One sheet, built around your actual numbers
  • Live formulas — totals, tax, margins, running balances
  • Conditional formatting on the values that matter
  • Delivered as .xlsx and Google Sheets
  • 1 revision pass
live above → the quote-builder template
most commissioned
Calculating Workbook
$1994 days
  • Multi-tab workbook: inputs, calculations, output
  • Dropdowns and data validation so users can't break it
  • Cross-sheet formulas with a documented dependency map
  • Locked calculation cells, unlocked input cells
  • Print-ready output tab (quote, invoice, or report)
  • 2 revision passes
live above → the cashflow chain B → D → E → summary
Spreadsheet System
$4497 days
  • Full model: scenarios, assumptions block, audit checks
  • Automation via Apps Script or macro-free formulas — your call
  • Import routine for your exports (CSV, bank, POS, CRM)
  • Error trapping and a self-test tab that flags broken links
  • Written documentation + a recorded walkthrough
  • 30 days of fix support
live above → the recalc order + error handling panel
add-ons Extra sheet or tab $35 Apps Script / VBA automation $95 Dashboard tab with charts $120 Rebuild of an inherited file from $150
delivery formats — .xlsx default · google sheets · csv · macro-free build on request · pdf print layout
03 the gap

What goes wrong when you hire for this.

Four failure modes anyone who has commissioned spreadsheet work will recognise — and how this studio is built so they don't happen.

01 · handover

"It worked on their machine. I added three rows and half the totals stopped counting them."

the dax answer

Ranges are built to grow. Totals reference whole structured ranges or tables rather than a lucky selection of eight cells, so a new row lands inside the sum instead of outside it. Before delivery, we add rows, delete rows, and paste a block in the middle — if a total flinches, it isn't finished.

02 · hard-coding

"The tax rate was typed straight into forty formulas. When it changed, the whole file had to be redone."

the dax answer

Constants live in one assumptions block and nowhere else. Rates, thresholds, discounts and fees are cells you can point at, and every formula reads them. Change the number once and the file agrees with itself. You'll notice the demo above does exactly this — the tax line reads a rate, it doesn't contain one.

03 · validation

"Someone typed 'N/A' into a number column and the whole model quietly returned garbage for a month."

the dax answer

Input cells get real validation: number-only, list-only, date ranges, minimums. Bad entries are refused at the cell with a message that says what's allowed. Where a value can legitimately go missing, the formula handles the blank on purpose instead of pretending it's a zero. Try typing a word into a number column in the rig above — it flags immediately.

04 · fragility

"The macro died on my colleague's Excel, and there was no documentation to work out why."

the dax answer

We ask what your team actually runs — Excel on Windows, Excel on Mac, Google Sheets, or all three — and build for that before writing a line of code. Most jobs don't need a macro at all, and we say so. When automation is genuinely the right answer, it ships with documentation, a plain-formula fallback, and a walkthrough recording so the file survives the person who commissioned it.

commission

Send us the file you dread opening.

Free look at what you have now, a fixed quote, and a working sheet before you pay the balance.

← Back to case studiesTell us what you need →