🎧 Listen to this article (14 min)
Most procurement teams do not have a purchasing problem. They have a follow-up problem. The purchase order goes out clean, the supplier receives it, and then nothing happens until somebody on the buying side picks up the phone and asks what is going on. That gap between "PO issued" and "PO confirmed" is where mid-market manufacturers lose the most time, and it is almost entirely staffed by people sending emails.
This is a guide to closing that gap. Not by adding a portal your suppliers will ignore, and not by ripping out your ERP, but by automating the specific follow-up loops that consume the most buyer hours and produce the least judgment.
The cost is rarely on anyone's budget line, which is why it survives for years. It shows up as headcount that feels permanently underwater and as delivery dates nobody trusts.
Start with volume. A manufacturer running 800 open purchase order lines with 60 active suppliers is generating a continuous stream of things that need chasing: unacknowledged orders, missing ship dates, quantity changes, split shipments, price discrepancies, and promised dates that quietly slip. According to Gartner, 50% of purchase order lines undergo changes after issuance, making real-time supplier visibility a procurement priority. Half your order book is moving, and manual follow-up is how most teams find out.
Now look at how the chasing gets done. A buyer opens the ERP, pulls a report of orders with no acknowledgement, and starts writing emails. Each email is short and mostly identical. The supplier replies a day or two later, sometimes with the ship date in the body, sometimes as a PDF attachment, sometimes with a confirmation that contradicts what was ordered. The buyer reads it, interprets it, and types the result back into the ERP. Then the cycle repeats next week.
Three things make this worse than it looks on paper. First, the work is unbounded. There is no version of manual follow-up where a buyer finishes. Second, it crowds out the work that actually requires a buyer, which is supplier negotiation, sourcing decisions, and exception judgment. Third, coverage is inconsistent by design. When a buyer is out or a month gets busy, follow-up is the first thing to slip, and the orders that go unchased are not the unimportant ones. They are just the ones nobody got to.
The downstream effect is planning error. A Deloitte supply chain study found that 70% of supply chain disruptions originate before materials leave the supplier's facility. If your only signal about supplier-side movement is an email thread somebody has to go read, you are learning about disruption late and reacting instead of planning. Production scheduling, inventory buffers, and customer commitments all inherit that lag.
Buyers reasonably ask why the ERP cannot just tell them. Whether your procurement team runs on SAP, Oracle NetSuite, Microsoft Dynamics 365, Epicor, or Infor, the ERP is an excellent record of what you intended and a poor record of what your supplier is actually doing.
The reason is structural. An ERP stores the PO you issued and the receipt you eventually booked. Between those two events, the authoritative information lives in your supplier's inbox and in their production schedule. Your ERP has no channel to it. The acknowledgement date field exists, but it only gets populated when a human reads a supplier email and types something in. So the field is either empty or stale, and neither state is a useful planning input.
This is why the classic remedies underperform. Built-in ERP supplier modules assume the supplier logs in, which most small and mid-size suppliers will not do for one customer. EDI works well for the handful of large suppliers who already have it and is not economically sensible for the long tail. Spreadsheet trackers are accurate on the day they are built. We wrote about this tradeoff in more detail in our comparison of ERP-agnostic PO automation versus built-in ERP modules, and the short version is that the constraint is not your ERP's capability. It is the missing communication layer between your ERP and supplier email.
For teams running Microsoft Dynamics 365, whether Business Central, Finance and Supply Chain, or Navision, Leverage AI integrates directly with your existing ERP environment to automate supplier PO confirmations, flag exceptions in real time, and surface OTIF data without custom development or ERP modification. The same pattern applies across the other platforms. We go deeper on the D365 specifics in our guide to Dynamics 365 procurement automation and PO visibility.
Not all follow-up is equal. Some of it is pure transport, moving a known question to a known recipient and parsing a predictable answer. Some of it needs a buyer. The automation case is strongest where the question is standard and the answer is structured, and these four loops account for the bulk of that volume.
Acknowledgement chasing. This is the highest-volume, lowest-judgment loop in procurement. An order goes out, and somebody needs to confirm the supplier received it, accepted the line items, and committed to a date. The question is identical every time. The escalation logic is simple: ask at day two, ask again at day five, flag to a buyer at day seven. There is no reason a person should be composing these.
Ship date confirmation and re-confirmation. An acknowledgement is a snapshot, not a guarantee. Dates move. The valuable loop is the recurring one that asks suppliers to re-confirm commitments as the promised date approaches, because a date that slips three weeks out is a planning adjustment and a date that slips three days out is an expedite and a customer conversation.
Reading and structuring supplier replies. This is the step teams forget to count. Sending the email is quick. Interpreting sixty replies that arrive as free text, forwarded threads, and attached PDFs, then getting the extracted dates and quantities into the ERP, is the actual labor. Automated parsing turns unstructured supplier responses into structured fields, which is what makes the rest of the loop worth anything. Our walkthrough of supplier email and PDF parsing for PO updates covers how that extraction works in practice.
Discrepancy detection. When a supplier confirms a quantity, price, or date that does not match the PO, that mismatch needs to surface immediately rather than at receiving or invoice matching. This is the one loop where automation should stop and hand off. Detection and routing are mechanical. The decision about whether to accept a changed date or push back is a buyer's call, and it should stay that way. Our PO exception management checklist lays out where those handoff lines belong.
The common thread is that automation replaces transport and transcription, not judgment. Teams that try to automate the negotiation get burned. Teams that automate the chasing and the typing free their buyers to do the negotiation properly.
The first change is boring and immediate: acknowledgement coverage goes to something near complete. Every order gets chased on the same schedule regardless of who is in the office. That consistency is worth more than the hours saved, because it means the absence of an alert becomes informative. Silence starts to mean "on track" instead of "nobody looked."
The second change is that your delivery data becomes usable. Once confirmations and date changes are captured as structured records instead of email threads, you can measure supplier promise accuracy rather than just final on-time performance. That distinction matters. A supplier who delivers late but tells you three weeks early is a planning input. A supplier who confirms on time and then misses without warning is a risk. Both look identical in a basic OTIF report, and they require completely different management. We covered this measurement gap in our piece on supplier OTIF tracking when ERP data is incomplete.
The third change is economic. Aberdeen Group research shows that automated PO tracking reduces operational costs by up to 30% for mid-market manufacturers. In practice that savings does not usually arrive as a headcount reduction. It arrives as the same procurement team covering substantially more order volume without adding people, and as fewer expedite fees and less premium freight because date slips got caught early enough to solve cheaply.
The fourth change is the one buyers notice personally. The job stops being reactive. Instead of opening the week with a report of everything unconfirmed and working down a list, a buyer opens the week with a short queue of genuine exceptions. That is a different job, and it is the one they were hired for.
The teams that get this working quickly all narrow the starting scope, and the ones that struggle try to automate everything on day one.
Start with acknowledgements only, on a single supplier segment. Pick the suppliers where you have the most open lines and the least reliable confirmation behavior, usually the small and mid-size vendors without EDI. Let the automated chase and parse loop run for a few weeks and measure one number: percentage of open lines with a confirmed date. That is your baseline and your proof.
Next, add re-confirmation on the same segment. This is where the planning benefit shows up, because you begin catching slips before they become expedites. Keep the escalation thresholds conservative at first. It is better to route a few unnecessary items to a buyer than to auto-accept a date change that should have triggered a conversation.
Then expand supplier coverage before expanding loop complexity. Adding two hundred more suppliers to a working acknowledgement loop is low risk. Adding change-order negotiation logic to sixty suppliers is not.
Two things to get right early. Keep the supplier experience in email. The entire premise is that suppliers respond to normal messages in the channel they already use, and any requirement to log in somewhere reintroduces the adoption problem you were trying to escape. And write down your exception routing rules before you turn anything on. Decide in advance what a buyer must see, what can be auto-recorded, and who owns each category. Automation makes whatever routing logic you have run faster, including the bad logic.
On measurement, resist judging the rollout on hours saved in month one. The durable metrics are confirmation coverage, average days from PO issue to confirmed date, percentage of date slips identified more than two weeks ahead, and expedite spend. Those move in a direction you can defend in a budget conversation. IDC projects that 60% of enterprise procurement teams will transition to AI-powered automation by 2025, and the teams already through it are the ones who instrumented the baseline before they started.
What does it mean to automate supplier follow-up?
It means software sends the routine chase messages, reads the supplier replies, extracts the committed dates and quantities, and writes them back as structured records. Suppliers keep responding by normal email. The automation handles the sending, the reading, and the data entry, and routes anything that conflicts with the purchase order to a buyer.
Do suppliers have to log into a portal or set up EDI?
No, and that is the point. Portal adoption fails with small and mid-size suppliers because logging into one customer's system is not worth their time, and EDI only makes economic sense for a small number of high-volume trading partners. Email-based automation works because it uses the channel suppliers already answer.
Will this work with our ERP?
It should, because the automation sits alongside the ERP rather than inside it. Leverage AI is ERP-agnostic and integrates with environments including SAP, Oracle NetSuite, Microsoft Dynamics 365, Epicor, and Infor. The integration reads open purchase orders and writes confirmed dates and exception flags back, with no ERP customization required.
How much of our follow-up should stay manual?
Keep judgment manual. Acknowledgement chasing, ship date re-confirmation, reply parsing, and discrepancy detection are mechanical and should be automated. Deciding whether to accept a pushed date, escalate to a supplier's management, or resource the order elsewhere is a buyer decision and should remain one.
How long before we see results?
Confirmation coverage usually improves within the first two or three weeks on the initial supplier segment, because the chasing becomes consistent immediately. The planning and cost benefits take a full order cycle to show up, since they depend on catching date slips early enough to avoid expedites. Most teams have a defensible baseline comparison within one quarter.
What is the difference between promise accuracy and OTIF?
OTIF measures whether the material arrived on time and complete. Promise accuracy measures whether the supplier's committed date held, and how early they told you if it did not. A supplier with poor OTIF but honest early communication is manageable. A supplier with decent OTIF who never warns you about slips is a hidden risk. You need both numbers, and you only get promise accuracy if confirmations are captured as data.
About Mary Chauvin
Mary is an Account Executive at Leverage AI, where she helps wholesale distributors and manufacturers automate purchase order follow-up and cut down on manual work. She works closely with procurement leaders to show how AI-driven PO tracking protects revenue and frees up teams to focus on higher-value work.