Comparações e alternativas

Price Double Check vs an Excel spreadsheet

Última verificação 7 September 2026 · Esta página é disponibilizada em inglês.

Price Double Check vs an Excel spreadsheet: the sheet checks one supplier's invoice against one price list well, if someone re-types every line and matches codes by hand. We do the same comparison from documents you forward by email, nothing typed, many suppliers. The sheet wins on ad hoc analysis and cost; the service wins on the second price list.

The comparison below is between two ways of answering one question: is this invoice billed at the price list in force. Most purchasers who ask it already own the spreadsheet answer. They built it in an afternoon, it worked for the first supplier, and it slowly stopped being used. We describe why that happens, what a spreadsheet still does better, and how many minutes each way costs per invoice, with the figures we can stand behind.

The VLOOKUP sheet that works for one supplier

The standard build is two tabs. One holds the price list as it arrived, usually an xlsx the supplier attached. The other holds the invoice lines, keyed in by hand or pasted from a PDF. A VLOOKUP or XLOOKUP on the item code pulls the list price next to the billed price, a fourth column takes the difference, and a filter shows the lines that are not zero. Microsoft's own documentation for XLOOKUP describes exactly this: search for a value in one column and return the matching value from another, with an if_not_found text when the code is absent. Excel 2021, 2024 and Microsoft 365 have it; Excel 2016 and 2019 do not, which matters when the sheet is shared with a colleague on an older licence.

For one supplier with clean codes, this sheet is correct and fast to build. We would not argue anyone out of it. It breaks at the second price list, for three reasons that are not about Excel.

• The second supplier's price list has a different layout. The header is on row 7 instead of row 1, the code column is called NEW CODES, the price you pay is one of three price columns. The lookup range has to be rebuilt per supplier, so one sheet becomes one sheet per supplier.
• The price list changes. A new version arrives by email with a new effective date and no version number in the file. Someone has to notice, paste it over the old one, and remember that invoices dated before the effective date are checked against the old list. In the chain we validated on, three price lists covered eighteen months, and one price appeared on invoices seven months before it appeared in any file called a price list.
• The invoice lines still have to get into the sheet. That is where the time goes, and the next section is about it.

Where the time goes in a spreadsheet check

Comparing two numbers takes Excel no time. Everything around it takes a person's time, and the person is usually the purchaser, not a clerk.

1. Re-typing the invoice. A PDF invoice with thirty lines is thirty codes, thirty quantities and thirty unit prices. Copy and paste from a PDF loses the columns, joins a quantity to a price as one token, or splits a description over two lines with the line number on the second one. Every one of those is a thing we hit on real invoices, and every one is a keying error waiting to happen in a sheet.
2. Normalising the codes. The lookup is exact. A supplier's ERP prints 5116_1 on the invoice and 5116 in the price list, and the suffix is documented nowhere. One code in the price list arrives as a text cell with a tab in front of it and never matches the same number typed as a number. Each of these is a #N/A that someone has to chase, and the chasing is not recorded anywhere, so the next person chases it again.
3. Descriptions that do not match. When the code is missing, the fallback is to match on the description, and it fails. In our validation set, 335 of 442 invoice lines carried a description that did not match the price list wording for the same item. A fuzzy match on text is a guess; a purchaser cannot take a guess to a supplier.
4. Versions. Three PDFs of one invoice with the same number and the same date but three different totals, all in the inbox. A sheet holds whichever one was typed in last. Nothing in it says the other two exist, or that the credit note that followed cancels only one of them.
5. Rounding. A price list with four decimals against an invoice with two produces a column of differences of 0.004 that look like findings and are not. The sheet needs a rounding rule before it needs a lookup.

None of this is a fault of the person who built the sheet. It is the reason the sheet is opened for the big invoice at quarter end and not for the twelve small ones in between, which is where a price increase that was never announced gets through.

What Excel does that a service does not

There are things a spreadsheet does better, and a purchaser should keep it for them.

• Ad hoc analysis. Once the numbers are in a sheet you can pivot them by month, by category, by supplier, chart the trend, and answer a question nobody wrote down in advance. Our reply is a fixed table of differences per invoice and a digest of what changed between price lists. It does not pivot.
• No subscription. Excel is already paid for. Our Free plan covers 10 documents a month; beyond that Standard is €29 a month for 100 documents and Pro is €79 for 500. A purchaser with three invoices a month and one supplier has no reason to pay anyone.
• Your own rules. A sheet can carry a tolerance of your choosing, a margin column, a note per line, a colour for the supplier's account manager. We show every difference and judge none; we do not know your margin unless you keep it elsewhere.
• Documents we cannot read. A photographed invoice, a scan without a text layer, an Apple Numbers file: a person can read them and type them in. We say we could not read them and ask for the PDF or the xlsx.

What we do instead is narrower. You forward the invoice, the price list, the purchase order from Sent, the order confirmation, the credit note, to one address. We join them by item code within a supplier, check that the line sums equal the printed total to the cent before comparing anything, and reply with the lines that differ, the lines we could not match with their amounts, and what to send to get more. The code normalisation, the version detection and the effective-date logic are inside the service because they were the traps that broke our own first sheet. The method is written up step by step, and it is the same method a careful spreadsheet follows.

Minutes per invoice, honestly

The figures below are our own estimates for a thirty-line PDF invoice from a supplier whose price list is an xlsx, done by a purchaser who knows the supplier. They are not benchmarks from a vendor and they are not measured across many companies. They are the steps, timed on the chain of documents we validated on. Measured this way, Price Double Check vs an Excel spreadsheet is a question of minutes, not of features.

The independent numbers point the same way, with the caveat that they measure other things. Raymond Panko's survey of field audits of operational spreadsheets, published in the Journal of End User Computing and on arXiv, reports that 91% of audited spreadsheets contained errors, with cell error rates of 1.1% to 2.5% in the audits that caught most errors. Ardent Partners' Accounts Payable Metrics That Matter in 2025, as summarised by Medius, puts the average invoice exception rate at 22% against 9% for the best teams, and the average time to process an invoice at 17.4 days against 3.1. Those are AP departments, not a purchaser with a sheet; we cite them for the direction, not for the decimals.

Price Double Check vs an Excel spreadsheet, the recommendation

Price Double Check vs an Excel spreadsheet is settled by how many suppliers and how many versions you deal with. Keep the spreadsheet if you have one supplier, a stable code list and fewer than ten invoices a month, or if the thing you want is analysis rather than a check. Forward the documents to us if you have more than one supplier, price lists that change without a version number, invoices that arrive in several versions, or a sheet that nobody has opened since the last price increase. Many purchasers do both: the reply from us lists the differences per line, and that table pastes into the sheet without any typing, which is the part the sheet was bad at. If your alternative is an AP system rather than a sheet, the comparison with AP automation is a different page.

Forward one invoice and the price list to your address and compare the reply with what the sheet would have found, and how long it took.