Blog · Wallet share and penetration
The complete whitespace calculation on a grid of ten accounts by five product lines, small enough to check by hand: the filled cells from the ledger, the norm penetration and median spend per line from the accounts that hold each, which empty cells are expected, which are returning lines, which cannot be bought, the value per cell, the value per account, the roll-up and the identity, so a reader can reproduce every figure and then run it on their own export.
A whitespace grid is a table of accounts against lines, and on ten by five every cell can be read. This page works the whole method: the filled cells, the norm per line, which empties count, the three kinds of empty, the value, the roll-up and the identity.
Trailing twelve months' revenue. Blank is empty.
| Account | A | B | C | D | E |
|---|---|---|---|---|---|
| 1 | $120k | $40k | $18k | ||
| 2 | $310k | ||||
| 3 | $22k | $9k | $14k | $6k | |
| 4 | $90k | $30k | $20k | ||
| 5 | $60k | $25k | $10k | ||
| 6 | $200k | $55k | $30k | $15k | $8k |
| 7 | $45k | $12k | |||
| 8 | $80k | $25k | $18k | ||
| 9 | $150k | $22k | $12k | ||
| 10 | $35k | $10k | |||
| Filled total | $1,112k | $181k | $129k | $55k | $14k |
Ledger for these accounts: $1,491,000. Filled cells sum: 1,112 + 181 + 129 + 55 + 14 = $1,491,000. Identity holds.
| Line | Accounts holding | Penetration | Expected (≥ 50%)? | Median spend among holders |
|---|---|---|---|---|
| A | 10 | 100% | Yes | $85,000 |
| B | 7 | 70% | Yes | $25,000 |
| C | 6 | 60% | Yes | $21,000 |
| D | 4 | 40% | No | $13,500 |
| E | 2 | 20% | No | $7,000 |
From last year's ledger and the account master:
| Cell | Kind |
|---|---|
| Account 2, line B | Returning: bought $52,000 last year |
| Account 9, line B | Returning: bought $20,000 last year |
| Account 5, line B | Cannot buy: account type excludes B |
| All other empties in B and C | Never bought |
| All empties in D and E | Not expected; ignored |
Expected empties valued at the line's median spend.
| Account | Empty expected cells | Value | Returning |
|---|---|---|---|
| 2 | B (returning), C | 25,000 + 21,000 = $46,000 | 1 |
| 9 | B (returning) | $25,000 | 1 |
| 1 | C | $21,000 | 0 |
| 7 | C | $21,000 | 0 |
| 10 | C | $21,000 | 0 |
| 5 | B is cannot-buy; none | $0 | 0 |
| 3, 4, 6, 8 | none | $0 | 0 |
| Total | $134,000 |
Account 2: the largest account and the most whitespace, with a returning line. Its B cell is the first call on the grid.
Whitespace value, segment = $134,000 Σ filled cells = $1,491,000 = ledger
Every empty counted. Twenty-three empties valued at their line medians: $360,000 of whitespace, most of it in lines few accounts want.
Cannot-buy unmarked. Account 5 pitched line B.
Returning not shaded. Account 2's B treated like account 1's C.
Grid not reconciled. A ledger line with no product code missing from the filled cells; the identity would catch it.
The same grid, the norm per segment cell with its count, the three kinds from history and the master. Covirage runs it on the ledger and the product list every month. The whitespace analysis guide covers the method, and the whitespace grid reading guide covers reading it at scale.
At ten it has to be, and it is the same at ten thousand: the norm for a segment comes from the accounts in it that hold the line. On a real base the cell would have thirty or more members and the norm would be greyed if it did not. Here the arithmetic is the point.
Half the accounts, stated. A line held by fewer is not expected of the ones that lack it, and its empties are ignored. Line E, held by two of ten, produces no whitespace on this grid.
The account bought the line before and stopped. The reason may be gone. It is the easiest recovery on the grid, and it is shaded so the rep sees it first.