Okay so we’re drowning over here. Hundreds of invoices coming in every month and our AP team is manually trying to match each one to a purchase order before we can cut payment. It’s slow, error-prone, and honestly it’s burning people out. Has anyone actually solved this with automation? Like, genuinely automated it — not just “here’s a spreadsheet template” kind of solved it?
Oh man, been there. PO matching is probably the most common AP headache I hear about, so you’re not alone.
The way I see it, there are a couple of directions you can go. The big enterprise platforms — SAP Ariba, Coupa — have matching logic baked in, but implementation is a whole project in itself. Months of config, significant cost, the works. That might be right for some orgs, but it’s not always realistic.
For more flexibility without the massive setup overhead, intelligent document processing tools have gotten genuinely good at this. We’ve used Lido for invoice extraction — it pulls out vendor name, invoice number, line items, quantities, amounts — and then you can pipe that into your ERP or even Excel to do fuzzy matching against your PO records. What I liked about it is that it doesn’t need templates. Different vendors send invoices in wildly different formats and it handles that pretty gracefully. Things like slightly different vendor name spellings or minor line-item description mismatches — the AI works through those instead of just failing.
In my experience, the sweet spot is setting up proper tolerance rules for three-way matching (PO, invoice, and receipt confirmation). A lot of teams start there, get comfortable, then loosen up to two-way matching over time as confidence grows.
Honestly? The ROI shows up fast. Within the first month for most teams. Worth piloting on a subset of invoices to prove it out before full rollout.
+1 on Lido. We made the switch around two months ago and it’s been pretty solid so far, knock on wood. Honestly what pushed us over the line was how well it plays with our existing spreadsheets — we didn’t want to rip out a bunch of stuff we’d already built and Lido basically just slotted in. Still early days but no major complaints.
That’s mostly right, but I’d actually push even harder on the scan quality point — people really underestimate this. We learned it the hard way during our implementation. We were blaming the software for weeks before someone finally looked at the actual scans we were feeding it and… yeah. They were bad.
Once we got a halfway decent scanner and bumped up to at least 300 DPI, our OCR accuracy went from sitting around 85% to hitting 96% basically overnight. Same software, same everything else. The hardware investment was like a few hundred bucks and it made more difference than anything else we tried. If you’re troubleshooting accuracy issues, seriously check this before you do anything else.
Same here, for what it’s worth. We’re a bit bigger, around 200 people, and we push through roughly 800 invoices a month. Tried Tesseract at first because the price (free) was obviously appealing and it seemed worth a shot. But our documents are all over the place — different vendors, different formats, some of them clearly scanned on a printer from 2009 — and the accuracy was just not there. Some docs we were lucky to hit 65%.
Moved to Lido and we’re pretty consistently above 95% now. Setup wasn’t instant but it wasn’t painful either.
Jumping in here because we’re in the exact same boat — QuickBooks is our system too and I spent way too long trying to figure this out. Short answer: it depends on the tool. Some have native QB connectors that work pretty seamlessly, others you’ll need to route through a middleware like Zapier or Make. We ended up going with a direct integration and it’s been mostly solid, though we had a rough first few weeks getting the chart of accounts mapped correctly. If you haven’t already, I’d recommend doing a trial run with like 20-30 invoices before you flip the switch on the full pipeline.
This is such a simple tip but honestly it made a huge difference for us too. We had invoices coming in through like four different people’s inboxes and it was a nightmare. One dedicated address fixed like 80% of our routing problems overnight. The only thing I’d add is make sure you set up an auto-reply so vendors know their invoice was received — cuts down on the “did you get my invoice” emails a lot.
Okay, so here’s a massive pro tip that I feel like nobody ever talks about when you’re diving into automation for PO matching. Before you even think about setting up any fancy tools or workflows, do yourself a huge favor: get a dedicated email address specifically for all your incoming invoices.
Seriously, make it something like accounts.payable@yourcompany.com or just invoices@yourcompany.com. Trust me on this one. It cleans up the entire pipeline so incredibly much, right from the start. You’re not sifting through a general inbox, everything is funneling into one spot, perfectly prepped for whatever automation system you’re using. Your future self will absolutely thank you for that bit of upfront organization.