Sign in

Blog · How-to guides

Excel or an analytics tool: when a spreadsheet is enough, and the four signs it is not

A fair account of what a spreadsheet does well for sales analysis, the point at which it stops, and the four signs a team has passed it: the same measure computed two ways by two people, a monthly refresh that takes days, a roll-up that does not reconcile and nobody knows why, and a number in a deck that nobody can trace. What changes when the measures are computed from stated definitions on every upload, and what does not need to change at all.

The short answerA spreadsheet is enough while one person computes the measures, the refresh takes an hour, the roll-up reconciles and every number can be traced to its rows. Four signs say it has stopped being enough: the same measure has two values because two people built it two ways; the monthly refresh takes days of copying and checking; a total disagrees with finance and nobody can say why; and a number in the board deck cannot be traced to the rows behind it. When those appear, the fix is not a bigger spreadsheet. It is stated definitions, a mapping done once, identity checks on every upload, and tables that cite their rows. The spreadsheet stays for what it is good at.

A spreadsheet built the first coverage report at most companies, and it was the right tool. At some point it stopped being the right tool, and nobody noticed because the report kept appearing. This guide gives a fair account of where a spreadsheet is enough, the four signs it has stopped, and what changes when it does.

Where a spreadsheet is enough

Condition Why it works
One person computes the measures One definition, in one head
Under a few hundred accounts Formulas hold; nothing times out
Refresh takes under an hour The person remembers the steps
The roll-up reconciles The SUM ranges are right, this month
Numbers are traceable The person can point at the cells

For a founder-led sales team with an analyst, this is most of the first two years.

The four signs

One measure, two values. Two people built coverage. One counted emails; one did not. Both are in decks. The argument in the meeting is about which spreadsheet, and it is the definition that was missing, not the tool.

A refresh that takes days. The steps are in one person's head, the SUM ranges have to be extended, the pivots refreshed, the subtotals stripped, and the totals compared by eye. Days, every month, and a person who cannot take a holiday in the first week.

A total that disagrees with finance. Region does not equal the sum of teams; the ledger says $4.18m and the sheet says $3.96m. Nobody knows which SUM range, which hidden row, which overwritten formula. An afternoon, sometimes a week, and the number is patched rather than fixed.

A number nobody can trace. The board asks where 27 percent came from. The sheet has been copied, pasted as values, and emailed. The rows are gone.

What changes

Spreadsheet Computed tables
Definition in the formula Definition written, versioned, cited
Columns mapped by hand each month Mapped once; reused
Roll-up checked by eye Identity checked on every upload; failures listed
Number in a cell Number citing its table, rows and version
Refresh in days Refresh in minutes; the person reads six validation lines

What does not change

The exports, the person, the questions. The ledger and the CRM produce what they always did. The analyst drops the same files. The measures are the ones the spreadsheet was trying to compute, now with the definition on the page.

A worked comparison

Month Spreadsheet Computed
Refresh time 3 days 20 minutes
Reconciliation issues found by hand 6 0; 3 exceptions listed automatically
Measures with two values 2 0
Board questions about provenance 4 0; every table cites

Where it goes wrong

A bigger spreadsheet. The four problems, larger.

A dashboard tool without definitions. The four problems, with charts.

The spreadsheet thrown out. It was the right tool for exploring, and it still is.

Moved before the definitions exist. The tool computes a measure nobody agreed.

The spreadsheet stays; the arithmetic moves

Covirage takes the same exports the spreadsheet did, maps them once, checks the identities on every upload, and computes from stated definitions with a citation on every figure. The Excel analysis solution describes the setup, and the five checks guide covers what to run on the spreadsheet in the meantime.

Questions people ask

What is a spreadsheet good at?

Exploring a question once, building a model with assumptions the reader can change, and presenting a small table. It is fast, everyone has it, and for a hundred accounts and one analyst it is the right tool. The problems start with repetition, multiple hands and scale.

Is an analytics tool just a bigger spreadsheet?

The useful kind is not. The difference is that definitions are written and versioned, the mapping from columns to roles is done once and reused, the roll-up identity is checked automatically, and every figure cites its rows. A dashboard tool without those is a bigger spreadsheet with charts, and it has the same four problems.

What does not need to change?

The exports. The ledger and the CRM produce the same files they always did. The person who dropped them into the spreadsheet drops them into the tool. What changes is what happens after the drop.