Spending
Spending trends without the spreadsheet gymnastics
Every company has a categorised spend report. Almost nobody trusts it, because the categorising is done by a tired person at month end.
The standard spending report is built the same way everywhere. Export the transactions. Paste them into a sheet. Add a column called Category. Spend forty minutes filling it in from memory, guess on eleven rows, pivot the result, and send it round. Next month, do it again, slightly differently.
The output looks authoritative. It is mostly a record of one person's mood on the third of the month.
The three things going wrong
The categories drift. “Software” in March quietly becomes “Tools” in April and “SaaS” in May, so the trend line breaks without anyone noticing.
The same vendor lands in two buckets. A design agency is Contractors when it does production work and Marketing when it does a campaign. Both are defensible. Only one can be in the chart.
The report arrives after the decision. Month-end reporting is, by construction, about a month that is already over.
A spending report you can argue with on the twelfth is worth more than a perfect one that lands on the third of the following month.
Automate the sorting, keep the judgement
The useful split is not between manual and automatic. It is between the part that is bookkeeping and the part that is a decision.
| Let the tool do it | Keep doing it yourself |
|---|---|
| Matching a vendor to the category it had last time | Deciding what the categories are in the first place |
| Totalling, ranking and charting | Splitting one vendor across two categories when it genuinely belongs in both |
| Flagging a vendor that has never been seen before | Judging whether a rise is a problem or an investment |
| Carrying categories forward month to month | Retiring a category that has stopped meaning anything |
Once the first column is handled, the second column takes about fifteen minutes and is actually interesting.
Read vendors before categories
Categories are an abstraction you invented. Vendors are facts. When something looks wrong in a category, the fastest route to the cause is the vendor list underneath it — nine times out of ten, a category that jumped 40% is one supplier that invoiced twice, or an annual renewal that landed in a month that usually has none.
That is why it is worth keeping both views side by side rather than treating the vendor detail as an appendix.
Separate the repeating from the one-off
The most useful cut in a spending report is not by category at all. It is: which of this will happen again next month?
Payroll, rent, and cloud bills recur; they belong in the forecast as scheduled items. A trade show, a legal matter, a piece of equipment does not; it belongs in the history but should never be extrapolated. Mixing them produces a run rate that is wrong in both directions — too high in the months after a one-off, too low in the months before the next one.
The short version
- Hand-categorising is where most spend reporting quietly loses its credibility.
- Let software carry categories forward; keep the judgement calls human.
- When a category looks odd, go to the vendor list before you go to a meeting.
- Split recurring from one-off before you calculate any kind of run rate.
- Timeliness beats precision — a report nobody reads has no accuracy at all.
Category and vendor, side by side
Efiner sorts the last thirty days by category and by vendor on the same screen, and every table exports to CSV in one click.
See the platformKeep reading
Efiner Financial Analytics Inc., Austin, Texas. Guides and analytics published by Efiner are informational only and are not investment, tax, or accounting advice.