Inventory the Folder: The Fields That Identify a File

A file name is a label someone typed. It does not prove anything about what is inside. Open a typical reporting folder and you might find:

• Opex_June_final_v3.xlsx
• Opex_June_final_v3_REAL.xlsx
• Opex_June_final_FINAL.xlsx

These three could hold three different sets of numbers. They could hold the same numbers. One could be last month's file, renamed. The names cannot tell you which, because people choose names in a hurry and no rule governs them.

So you cannot tell from the folder listing which file fed the numbers. You need recorded facts about each file: who sent it, when it arrived, what period it covers and how its figures are stated.

That record is a file inventory. It is a list of every file in the folder, one row per file, with the same set of facts for each.

In this lesson you build a first inventory by hand. In the next lesson you will use AI to draft one faster. You need to know what a good row looks like first, so you can check what the AI gives you.

Your folder shows a modified date for each file. This is the last time someone saved a change to it. Opening a file and saving it again resets that date, even if no figure changed. Some copy and sync tools can change it too.

The date received is when the file actually reached you. It comes from a record outside the file: the email timestamp, the upload log on a shared portal, or a handover note.

Example:

• The business unit emails Payroll_Q2.xlsx on 3 July.
• On 9 July a colleague opens it to check a total, and Excel asks to save. They click Save.
• The folder now shows 9 July.
• A week later, someone sorts the folder by modified date and assumes this is the newest payroll file. In fact an updated version arrived by email on 7 July and is sitting further down the list.

The modified date only shows when someone last saved the file. It does not show when the data arrived or which version is current. Record the date received from the email, upload or handover, and note which one you used.

A useful inventory records the same seven facts for every file:

1. File name: exactly as it appears, typos included.
2. Folder location: the path, such as /Reporting/2024-06/Inputs/. Two files can share a name in different folders.
3. Owner or sender: the person or team who supplied it.
4. Date received: from the email, upload or handover.
5. Reporting period: the time the figures cover, such as Q2 2024 or June 2024 year to date.
6. Currency: such as EUR, USD or GBP. Use the three-letter code, not a symbol like $, which several currencies share.
7. Units: the scale the figures are stated in. Typical values are units (figures as shown), thousands or millions. A figure of 1,250 means 1,250 in units but 1,250,000 in thousands.

One row might read:

File name	Location	Sender	Received	Period	Currency	Units
Opex_June_final_v3.xlsx	/Reporting/2024-06/Inputs/	M. Lindqvist, Facilities	2 Jul 2024 (email)	June 2024	EUR	thousands

The first four fields tell you where the file came from. The last three tell you what its numbers mean.

The file name might say Sales_Q2_USD. Do not copy that into your inventory. Open the file and look for what the file itself states. Common places:

• Title rows at the top of a sheet: 'Sales by region, Q2 2024'
• Column headers: 'Amount (EUR 000)'
• Notes or footnotes below the table: 'All figures in USD millions'
• A cover or notes tab at the front of the workbook

Then record where you found each fact, in a short 'where found' note. For example: Summary tab, row 2 title: Q2 2024, EUR 000. This lets anyone check your entry in seconds without searching the whole workbook.

Watch for conflicts. The name might say Q2 while the title row says 'June 2024 year to date', which covers six months, not three. When the name and the content disagree, record what the content says, note where you found it, and flag the mismatch so the owner can confirm.

Sometimes the fact is not in the file. Freight_costs.xlsx has a column of numbers and no currency anywhere. The sender works in the UK office, so writing GBP is tempting. Don't.

Write unknown – ask owner and send the question. An honest gap shows up when someone sorts or filters the inventory. A confident wrong entry looks the same as a correct one, and nobody will look at it again until a total is wrong.

When you review an inventory, look for rows that need follow-up:

• Missing: a blank cell in period, currency or units.
• Marked unknown: waiting on the owner.
• Vague: 'Q2' with no year. '$', which could be USD, CAD or AUD. '000' with no note of where it was found. 'YTD' with no end month.

Each of these rows is a figure you cannot yet safely use. When the owner replies, update the row and note the source, such as 'confirmed by email, M. Lindqvist, 5 Jul'.

Two files called Sales_EMEA_Q2.xlsx sit in different subfolders. Without an inventory they look like duplicates, or like two parts of one total.

The inventory shows otherwise:

File	Location	Total shown	Currency	Units	Where found
Sales_EMEA_Q2.xlsx	/Inputs/Sales/	1,250	EUR	thousands	Header: 'EUR 000'
Sales_EMEA_Q2.xlsx	/Inputs/Archive/	980,000	USD	units	Header: 'USD'

If someone adds the two totals as shown, they get 981,250. That number means nothing, because it adds euro thousands to dollar units. File A is really EUR 1,250,000.

The inventory makes the difference visible before anyone adds them. You still need to find out whether these are two separate sources or two versions of the same data, and later lessons cover that. Converting currency uses a rate your organisation agrees, not one you pick. The inventory's job is simpler: show that the two files are not stated the same way.

Use the sliders to see how far apart a naive sum and a like-for-like figure can be. The exchange rate here is for illustration only.

Now build your own. Keep it as a simple spreadsheet table:

• One row per file. Two files with the same name get two rows.
• One column per field: file name, folder location, owner or sender, date received, reporting period, currency, units, and a 'where found' note.
• No merged cells, and no meaning carried only by colour. Merged cells break sorting and filtering, and colour can't be filtered reliably or read by everyone.
• Plain values: 'thousands', not 'k' in one row and '000' in the next.

Start small. Pick 5 to 10 files from the folder behind your current management pack. Open each one, read the facts from inside it, and fill in the row. Take the date received from the email or upload record.

When you finish, filter the period, currency and units columns for blanks and for 'unknown'. Those rows are your follow-up list.

Keep customer identifiers, account credentials and nonpublic reporting figures out of any tool your organisation has not approved. A hand-built inventory in your usual spreadsheet keeps everything where it already lives.