Proof of delivery, read, matched
and ready to invoice.
Proof of delivery processing reads POD photos, scans and driver app records, matches each one to its order and checks it for a signature, a date, quantities and remarks such as damage or shortage. Clean PODs release the invoice. PODs with problems go to a person before the customer is billed.
Why is POD handling worth automating?
A delivery is not finished until the paperwork is. Many customers will not pay without a POD, and some contracts say so. Missing or unclear PODs delay invoices, start disputes and leave you without evidence when a claim arrives weeks later.
PODs arrive in mixed forms: signatures in a driver app, photos of a paper docket taken in a dark yard, a stack of dockets scanned at the end of the week, and PDFs emailed by subcontractors. Someone matches each to an order, checks it and files it. Automation does the matching and checking, and leaves the judgement calls to a person.
Is proof of delivery processing a job for an agent?
Tick what is true of your proof of delivery processing work.
Tick the statements that apply and the suggestion appears here: an AI agent, ordinary automation, or a person with software support.
What does the system read from a POD?
| Item | Why it matters |
|---|---|
| Order or consignment reference | Links the POD to the right job; often a barcode or a handwritten number |
| Signature present | Most contracts need one; the system checks it is there, not who signed |
| Printed name | Needed when a customer disputes who received the goods |
| Date and time | Proves on-time delivery; checks waiting time claims |
| Quantity received | Compared with the quantity dispatched |
| Remarks | Handwritten notes such as “2 cartons damaged”, “short 1 pallet” or “received unchecked” |
| Photos | Goods at the drop point; damage at delivery |
Handwritten remarks are the most valuable field and the hardest to read. AI models read handwriting far better than older text recognition, but not perfectly. A good rule: if there is any handwriting in the remarks box, a person sees it. A blank remarks box can pass automatically.
How are PODs matched to orders?
- Barcode or QR code on the docket: exact match, the most reliable.
- Printed reference read from the document: match to an open order, with a check on customer and date.
- Handwritten reference: match, then confirm with the delivery address and the date of the run.
- No reference: suggest likely orders from the driver, date and address, and ask a person to pick.
Print a barcode on every docket you produce. It is a small change that removes most matching errors. For subcontractor PODs, ask for your reference on their paperwork as part of the job instruction.
Batch scans need splitting first. A 40-page PDF from the depot scanner must become 40 PODs, and a two-page docket must stay together. The system does this before it reads anything, and flags pages it cannot place.
How much time could proof of delivery processing give back?
Put your own numbers in.
The starting values are an illustrative example, not a benchmark or a neu figure. Use your own volume, the time each one takes today and the share you expect the system to handle without a person.
Time saved is only part of the case: response times, missed work and errors matter too. The ROI guide covers how to set a baseline, and the cost guide what the build and running costs include.
What gets flagged, and what happens next?
| Outcome | Action |
|---|---|
| Clean: signed, dated, quantities match, no remarks | Mark delivered; release the invoice |
| Remarks: damage, shortage or “unchecked” | Hold the invoice line; alert the customer service lead; start a claim record |
| Quantity differs from dispatch | Hold; check against the loading record; a person decides whether to invoice in part |
| No signature or unreadable | Ask the driver or subcontractor for a better copy; hold the invoice |
| No POD after a set time | Chase the driver or subcontractor automatically; report open items daily |
Releasing an invoice can mean marking the order “ready to bill” in the TMS, or creating the sales invoice in Xero, QuickBooks or Sage with the POD attached. Attaching the POD to the invoice itself saves a round of queries from the customer’s accounts team.
What about international deliveries and CMR notes?
For international road carriage, the CMR Convention has the force of law in Ireland through the International Carriage of Goods by Road Act 1990. Note that one legal commentary states that movements between Ireland and the UK are outside CMR, so check how your own contracts treat that route with your solicitor or insurer.
Timing of remarks matters under CMR. Article 30 says that apparent loss or damage should be noted at delivery. For loss or damage that is not apparent, written reservations must reach the carrier within seven days of delivery, not counting Sundays and public holidays. A claim for delay needs a written reservation within 21 days.
So for CMR consignments, the system should read the remarks box on the consignee’s copy and alert someone on the same day. A claim deadline missed because a scan sat in a folder for a week is avoidable. This page is a practical guide, not legal advice.
What does this look like in numbers?
Take a hypothetical haulier with 1,000 deliveries a month. Each POD takes about 2 minutes to find, match to the job, check and file. That is about 33 hours a month of office time, often done in a batch at the end of the week, which delays invoices by days.
With automation, assume 75% of PODs are clean and pass with no one looking. The other 250 go to a person with the likely match and the problem shown, at about 1 minute each. That is about 4 hours a month. Invoices for clean deliveries can go out the same day the POD arrives.
These figures are illustrative. Your clean rate depends on docket design, photo quality and how many PODs come from subcontractors. Measure a month of current PODs first, then use the calculator on this page.
Where should you start?
Start with the POD backlog report: every delivered order without a POD, by driver and customer. That alone often shows where invoices are stuck. Then add reading and matching for your own drivers’ PODs, where you control the docket format. Add subcontractor PODs next. Once PODs are reliable, connect them to carrier invoice audit and invoicing. For other paperwork, see document processing.
Common questions
Do we need a driver app, or will photos of paper dockets work?
Photos work. AI models read a reasonable phone photo of a docket well, including most handwriting. Quality is the issue: dark yards, blur and folded paper all cause errors. A driver app gives cleaner data, but many firms run both for years. The system should accept both and ask for a better photo when it cannot read one.
Can it detect damage from photos?
It can flag photos that appear to show damage, such as crushed cartons or torn wrapping, for a person to look at. It should not decide a claim. Photo checks are useful as an early warning, and as evidence that goods were in good condition at delivery when a claim arrives later.
How long should we keep PODs?
Keep them at least as long as your contracts, insurance and claims periods need, and as long as your accountant needs them as support for invoices. Revenue’s general guidance is to keep business records for six years. Agree a retention period with your accountant and apply it to the POD store automatically.
Can customers get their own PODs?
Yes. Once PODs are matched to orders, customers can retrieve them by reference on a portal or by message. That removes a common daily request from the office. Limit each customer to their own documents, and do not show driver details beyond what the delivery needs.
Sources
- Irish Statute Book: International Carriage of Goods by Road Act 1990 (CMR in the Schedule, Article 30)
- Legal Guide Ireland: CMR carriage (Ireland and UK movements)
- Revenue: Keeping VAT records
We last checked these sources on 19 November 2026. Rules, fees and prices change. Check the source before you act.
Bring us the task.
Tell us what happens today, which tools are involved and where it goes wrong. On a free 30-minute review, we’ll work out whether an agent, an automation or neither is the right fix, and what a sensible first release would be.
Start a project.

Book a call or send us a message. We read everything.
tash@neu.ie
Working with teams in Ireland and worldwide.
Free 30-minute AI review.