How distributors stop retyping emailed purchase orders into the ERP

It’s 7:40 on a Tuesday night and someone on your inside sales team is still at their desk. There are 23 unread POs in the orders inbox. Two are PDFs from a customer’s purchasing system. One is a spreadsheet with a tab nobody asked for. One is a phone photo of a printed form, taken at an angle on a warehouse table. The rest are typed straight into the body of an email: “Same as last time but 40 instead of 25, ship to the Mesquite store.”

Each one gets opened, read and keyed into the ERP line by line. That is the job, and it has been the job for years.

Where retyped orders go wrong

The typing itself is not the hard part. The hard part is everything a person has to know to type it correctly.

• Units of measure. The customer orders 10 “CS” and your item is stocked in eaches. Someone has to know whether a case of that item holds 6, 12 or 24. Get it wrong and you ship a fraction of the order, or many times too much.
• Part numbers that aren’t yours. The PO lists the customer’s part number, not your SKU. The cross-reference lives in the ERP if you’re lucky, in a spreadsheet if you’re not, and in one person’s head most of the time.
• Ship-to addresses. A customer with 14 locations sends the PO from headquarters. The ship-to is buried in a note at the bottom. The default address gets used instead.
• Prices that changed. The customer’s PO shows last quarter’s price. Nobody catches it until the invoice is disputed 30 days later.
• Orders that wait. A PO that arrives at 4:50 on Friday gets entered Monday. The customer wanted a confirmation the same day and starts calling to chase it.
• One person who knows everything. When your most experienced order entry person is out sick, orders slow down and errors go up. Everyone knows it, and nobody has written down what that person knows.

None of this is a training problem. It is what happens when a person is asked to act as the bridge between hundreds of customers’ formats and one ERP.

What retyping actually costs

The cost shows up in three places, and only one of them is on the payroll report.

The first is hours. Every order is opened, read, looked up and keyed. A short order with three lines takes a few minutes. A 60-line restock order from a big account takes far longer, especially if the part numbers need looking up. Month end and seasonal peaks bring overtime, or orders that sit until the next day.

The second is errors. A wrong unit of measure means a return, a reship, a credit and a customer who now checks every order you send. A missed price change means a dispute and a delayed payment. These rarely get tracked back to order entry, so they rarely get counted.

The third is speed. A customer who emails a PO at 10 a.m. and hears nothing by 3 p.m. calls. Someone stops what they’re doing to find the order, check whether it was entered and send a confirmation by hand. Multiply that by every customer who chases, every week.

You can put a number on the first one with your own figures. Count the people who enter orders, estimate the hours each spends on it in a normal week and use their loaded hourly cost. Here is an example for a mid-size distributor with three people who each spend about 10 hours a week on emailed POs. Swap in your own numbers.

People doing it

3

Hours each, per week

10

Loaded cost per hour

$32

Hours per month

130

Cost per year

$49,920

Hours back per month

65–91

An estimate from the numbers above, not a client result. It assumes 50%–70% of the work can be taken off your team; your audit checks that against your own numbers.

In that example, the team spends 1,560 hours a year retyping orders. That is before a single error, reship or chasing call is counted.

What happens instead, step by step

Here is what happens to a PO once order entry is automated. Your customers don’t change anything. They keep sending orders the way they send them today.

1. The order inbox is watched

The inbox your customers already send to is checked as mail arrives, including the PO that lands at 9 p.m. on a Sunday. Nothing sits unread because someone is at lunch or on vacation.

2. Each PO is read, whatever its format

A clean PDF, a scanned fax, a spreadsheet with merged cells, a phone photo or three lines typed in an email body: each one is read the way a person would read it. The order details are found wherever the customer put them, not in a fixed template. An email that says “same as last time but 40 instead of 25” is checked against that customer’s last order on file, and the change is flagged for a person to confirm.

3. The order details are pulled out

First the header: PO number, customer, bill-to and ship-to, order date and requested ship date. Then every line: the customer’s part number, description, quantity, unit of measure and price. Notes like “deliver to dock 3” or “call before delivery” are kept with the order.

4. Every line is matched against your records

This is where most of the work is. The customer is matched to the account in your ERP. Each line is matched to your item master, using the customer’s part number cross-reference when they use their own numbers. The unit of measure is converted to how you stock and sell the item. The price on the PO is compared with the customer’s price list or contract price.

5. The checks your team makes today are run on every order

Your team already has rules, even if they aren’t written down. Is this ship-to on file for this customer? Is the quantity far outside what they normally order? Is the item discontinued or on allocation? Does the PO price match ours? Those rules are written down during setup and run on every order, every time, including at 4:50 on a Friday.

6. Clean orders go into the ERP and the customer hears back

An order that passes every check is entered as a sales order in your ERP, or staged for one-click approval if you prefer to keep a person in the loop. You choose, and you can choose differently for different customers. The customer can get a confirmation the same day the PO arrives, without anyone writing it by hand.

7. Anything unclear goes to a person, with the question highlighted

When something doesn’t match, the order goes to your team. They see the original PO side by side with the draft order and the exact problem marked: this part number has no match, this ship-to isn’t on file, this price is lower than the contract price. They answer the question, and the order goes in.

Why the odd cases matter more than the touchless rate

You will hear vendors talk about a “touchless rate”: the share of orders that go through with no person involved. It is a reasonable thing to track, but it is the wrong goal.

The goal isn’t zero people. It is people touching only the orders that need judgment. A new customer part number. A ship-to that isn’t on file. A quantity that looks like a typo. Those are the orders where your team’s experience matters, and today they’re buried among dozens of routine orders that don’t need anyone.

What makes this work over time is what happens after a person answers. When your team maps a new customer part number to your SKU, that mapping is saved. When they confirm a new ship-to, it’s added to the customer’s record. The same question isn’t asked twice. Over time, fewer orders need a person, because the answers your team gives are kept and reused.

That is also how the knowledge stops living in one person’s head. Every answer becomes a rule anyone on the team can see.

What it works with

It runs on the tools you already use. That means the ERP you already run, with the customers, items and prices already set up in it. It also means the order inbox your customers already send to. There is no new software for your team to learn and nothing new for your customers to sign up for.

If some of your larger customers already send orders by EDI, they stay on EDI. That channel works and there’s no reason to touch it. This handles the rest: the long tail of customers who send PDFs, spreadsheets and typed emails. Some of them may never set up an EDI connection with a supplier your size.

Distribution runs deep in North Texas. Take the Great Southwest Industrial District. The City of Grand Prairie says much of the district’s roughly 80 million square feet lies in Grand Prairie. It calls the city a well-established distribution center. If you distribute from there or anywhere else in Dallas–Fort Worth, your orders likely arrive in more formats than your team would choose.

What changes for your team

Orders are entered the same day they arrive, not the next morning. Customers get confirmations without having to ask, so the chasing calls stop.

Your order entry people stop being typists. They spend their time on the orders that need a decision, on customers who need attention and on problems they can actually fix. Month end no longer means a backlog of keying.

Errors from retyping drop away, because nobody is retyping. Wrong units, missed price changes and default ship-tos get caught by the checks before the order reaches the warehouse.

And when your most experienced person takes a week off, orders keep moving.

For more on how this fits the rest of a distribution business, see our automation guides for distributors. For what the work looks like as a project, see order automation for distribution and manufacturing.

Where to start

Don’t start with every customer at once. Start small and measure.

1. Pick one inbox. The shared orders inbox most of your POs arrive in.
2. Pick your top customers by volume. Their formats are the ones your team types most often, so they’re worth getting right first.
3. Measure time per order today. Have your team time a normal week of order entry for those customers. Note the errors and the chasing calls too.
4. Agree a target. Same-day entry, confirmations sent without a chasing call, fewer reships: whatever matters most to you.
5. Go live and widen. Your first workflow goes live within 14 days of access to your tools. Once a customer’s orders run clean, add the next one.

It starts with a free 20-minute audit call. We look at how orders reach you today and where the hours go. You get a written plan within 48 hours and a fixed-scope quote after the audit.