Backorder Notification Email: A Template AI Fills From Stock Data

A backorder email should say, in the first lines: which items are short, what ships now, the new date and how firm it is, and what the customer can choose to do. AI can draft the wording from a stock file, but the date and quantities must come from your spreadsheet, not from the assistant.

That split is the whole method. Your stock data holds facts: the order number, the quantity ordered, the quantity on hand, the supplier's confirmed arrival date. The AI turns those facts into polite, specific sentences and handles the variants (partial shipment, substitute, no date yet) without you rewriting the email each time. If you let it calculate or recall facts, it will occasionally write a date that looks right and is not.

Follow me on Instagram@sagnikteaches

What the customer needs in the first ten seconds

A buyer at a trade customer opens a backorder notice with one question: do I need to do something? Put that answer near the top. A good email gives six pieces of information, in this order.

Connect on LinkedInSagnik Bhattacharya

1. The order and the problem. Order number, the line or lines affected, in plain words ("20 of 48 cases of the 750 ml jars are short").
2. What is shipping now. If part of the order is in stock, say what leaves and when, so the customer can plan the rest.
3. The new date, and how firm it is. "Confirmed by our supplier for 16 November" is different from "expected mid-November".
4. The choices. Wait, accept a substitute, or cancel the short lines. One sentence each.
5. What happens to the money. Whether you charge for the shipped part now, whether freight changes, whether a substitute costs more or less.
6. A named person and a deadline to reply. If you need an answer by Thursday to hold the allocation, say it.

Leave out the long apology, the explanation of global supply chains and the marketing line at the bottom. A short, specific message reads as competent. A long one reads as an excuse.

Subscribe on YouTube@sagnikteaches

Step one: shape the stock file so every field has one job

The template can only fill what the file contains. Most wholesalers export from their stock or order system into a spreadsheet; give that sheet one row per short order line, with these columns. If your system cannot export the exact layout, build a helper sheet that pulls from the export with lookups.

Two columns carry most of the risk: eta_status and promised_date. Make the status a dropdown with three values only. Then calculate the promised date with a formula so nobody types it. In Excel, =WORKDAY(supplier_eta, 3) (with your own cell reference in place of supplier_eta) adds three working days to the supplier's arrival date for receiving, checking and picking, and a version wrapped in IF(eta_status="None","",WORKDAY(...)) leaves the cell blank when there is no confirmed date. Google Sheets has the same WORKDAY function, and both skip Saturdays and Sundays and accept an optional list of holiday dates as a third argument. Adjust the buffer to your real handling time; measure it on the last twenty inbound deliveries rather than guessing.

Never let a blank reach the customer. Add a check column that reads "OK" only if every field the chosen variant needs is filled. A merge that produces "Dear , your order of is delayed until " tells the customer nobody read it.

Step two: ask AI for the template, once, with the rules written in

Write the template in a single sitting rather than generating a new email for each case. Give the assistant the column names, the rules and your tone, and ask for a template with placeholders in square brackets. ChatGPT, Claude and Gemini all manage this; a free or standard paid plan is enough because nothing here needs the strongest model.

I run a wholesale business selling kitchenware to independent
retailers. Write backorder email templates I will fill by mail merge.
Placeholders use these exact names in double braces: {{customer_contact}},
{{order_no}}, {{item_name}}, {{qty_ordered}}, {{qty_shipping_now}},
{{qty_backordered}}, {{ship_now_date}}, {{promised_date}},
{{substitute_item}}, {{substitute_price_diff}}, {{reply_by}}.
Write four variants: (A) nothing in stock, confirmed date; (B) part
shipping now, rest on backorder with a confirmed date; (C) substitute
offered; (D) no date yet. Rules: first sentence says what is short;
second says what ships and when; never invent dates, quantities or
prices; never apologise more than once; plain, warm, under 130 words;
end with a choice (wait / substitute / cancel) and a reply-by date.
Sign off as "[your name], [company]". Subject lines under 60 characters.

Keep the placeholders exact, because your mail merge tool matches them character for character. Ask the assistant to list every placeholder it used at the end, and tick them against your column headings. A single typo, such as {{promised_data}}, is the most common reason a test merge comes out blank.

Four filled-in variants, using one made-up customer order

Take a kitchenware wholesaler with a retailer's order for 48 stainless pasta pots. The stock file shows 28 on the shelf, 20 short, the supplier confirms the next container on Thursday 12 November, and the wholesaler allows three working days after receipt. The weekend does not count, so the formula lands on Tuesday 17 November, not Sunday the 15th. These are illustrative, as is everything below. The sample outputs are illustrative of what the prompt above returns, lightly edited.

Variant B: part ships now, rest on a confirmed date

Subject: SO-10482: 28 pots ship today, 20 follow on 17 November

Hi [first name],

We're short on 20 of the 48 stainless pasta pots (8 litre) on
order SO-10482. The other 28 ship today, Tuesday 10 November.

Our supplier has confirmed stock arriving on 12 November, so the
remaining 20 will leave us on Tuesday 17 November.

Your options: keep the order as it is, swap the 20 for the 9 litre
pot at $1.20 more a unit, or cancel the 20 and we'll adjust the
invoice. Could you let me know by Thursday 12 November? If we don't
hear from you, we'll ship the 20 on the 17th as planned.

[your name], [company]

Check what the assistant did well: the outcome is in the subject line, the sentence about what ships comes second, and the default ("we'll ship on the 17th") is clear. Check what it gets wrong in some runs. It sometimes adds a line such as "We truly apologise for any inconvenience", which is filler, and sometimes writes "the remaining 20 pots will arrive on the 17th" when you only ship them that day. Arrival depends on your carrier; promise dispatch.

Variant A: nothing in stock, supplier confirmed

Subject: SO-10490: pasta pots delayed to 17 November

Hi [first name],

The 8 litre stainless pasta pots on SO-10490 (24 units) are out of
stock with us. Our supplier has confirmed a delivery on 12 November,
so they will ship from us on 17 November.

If that date doesn't work, we can offer the 9 litre pot in the same
range at $1.20 more a unit, available now, or cancel the line.
Please reply by 11 November so we can hold the allocation.

[your name], [company]

Variant D: no date yet

This is the one to treat with the most care. A good version says: "Our supplier has not given us a delivery date yet. I will email you again by Friday 13 November with an update, whether or not I have a date. Meanwhile, you can wait, swap to [substitute] or cancel the line." The commitment to a date for the next update is a promise you can keep. Add a line in your process that someone in the team really does send that Friday email; otherwise this variant creates a second broken promise.

Variant C: a substitute offer

A substitute offer needs three facts a buyer will check: pack size, price difference and whether it changes shipping weight. Have the stock file hold those as columns rather than relying on the assistant to compare product descriptions. If the substitute comes in a different pack quantity (say a case of 6 instead of 12), the quantity needs recalculating, and a person should approve the offer. Give the assistant the comparison, not the job of finding it.

Step three: test with five real rows before sending any

Pick five order lines that cover your messiest cases: a full shortage, a partial shipment, a line with a substitute, a line with no date and one on a key account where a call is better than an email. Run the merge into a draft folder, not the outbox. Read each one aloud as the buyer would.

Use this short checklist on each draft.

• Do the quantities add up (shipping now plus backordered equals ordered)?
• Does the promised date fall on a working day, and is it later than the supplier's ETA?
• Does the subject line show the order number and the outcome?
• Does a blank field appear anywhere, even a comma or a space where a value should be?
• Would you be comfortable if the buyer forwarded this to their boss?

When a draft fails, fix the cause, not the draft. If the weekday is wrong, the formula is wrong. If a name is missing, the customer record is incomplete. If the tone is off, ask the assistant to rewrite the template with a sentence of feedback, and re-test the same five rows.

Step four: choose how the emails go out

Three routes work for a small wholesaler, in rising order of effort.

If the volume is growing, an automation tool is the next step, and the purchase order and supplier email tutorial covers the mirror image of this problem: chasing the supplier who caused the shortage. Whatever you choose, send from a shared address, such as a sales desk mailbox, so replies reach whoever is on duty rather than one person's inbox.

A worked day: eleven short lines, nine customers

Suppose a wholesaler of kitchenware with a three-person sales desk exports the morning's allocation run and finds 11 short lines across 9 customers, from 140 lines in total. Eight of the lines have a confirmed supplier date. Two have an estimate. One has no date at all.

The coordinator pastes the 11 rows into the template sheet. The formula fills the promised date for the eight confirmed lines. For the two estimates, eta_status is set to Estimated, which picks a variant saying "expected" rather than "confirmed". For the one with no date, the status is None, and the check column stays red until the coordinator types the date she will next update the customer. Two customers have two short lines each; she merges those into one email by hand-editing the first draft, because two emails about the same order look careless. Fixing the files and reading the drafts takes about 25 minutes. The same job, writing each email from scratch at three or four minutes each, would run nearer 40 minutes and would be less consistent. The saving is real but modest on a single day. The larger benefit is that the replies are predictable: customers know what you will say, and the sales desk spends the afternoon on the replies that need judgement.

The coordinator also tracks three numbers for a fortnight: how many customers replied, how many chose to cancel and how many phoned. If phone calls rise after you change the template, the email is probably missing something the buyer needs.

Before and after: the same message with and without the stock facts

Here is the kind of backorder email a busy desk writes without a template:

Hi, unfortunately the pots you ordered are out of stock at the moment
so there'll be a delay. We'll let you know when they're in. Sorry for
any inconvenience. Thanks.

It names no order, no quantity, no date and no choice. The buyer has to write back to ask what is late, how late and what you plan to do, and that exchange costs both sides more time than writing the email properly would have. The filled-in version above answers all those questions up front. It is also easy to automate, because every sentence maps to a column.

When the assistant makes up a date: a failure and how to catch it

A tempting shortcut is to paste the raw stock export into a chat assistant and say "write a backorder email for each of these". For small batches it often works, which is the danger. In a longer run, an assistant may fill a missing delivery date with a plausible one, carry a quantity over from the wrong row, or write "arriving next week" because the supplier note said "mid-month". None of these looks wrong when you skim the email.

Catch this with a reconciliation step: after the draft is generated, ask the assistant to list every date and number it used beside the source row, in a table, and compare the table against the sheet yourself. Better still, use a mail merge so the assistant never touches the numbers at all. When an AI model does the merge, you are trusting it to copy accurately; when a spreadsheet does it, you are not.

A second failure is subtler: tone drift when the same email goes to a key account. A three-year customer with a 40-line weekly order does not want a form email about one short pot. Mark those accounts in the sheet, hold their drafts, and phone first. The email then becomes written confirmation of the call.

Subject lines that get opened and filed

A backorder email is read twice: when it arrives and when someone searches the inbox for the order number a week later. So the subject carries three things: the order number, the outcome and the date. Compare these three, all about the same order.

Ask the assistant for three subject options per variant and pick the shortest one that still carries the order number. Keep them under about 60 characters so phone screens do not cut off the date. Do not use urgent-looking words such as "URGENT" or exclamation marks, since the news is not the buyer's emergency and the words push the email towards spam folders.

Handling quantities that do not split cleanly

Real stock files have awkward cases that a simple template mishandles. Four are worth a rule in the sheet before you start merging.

• Case quantities. If you sell in cases of 12 and the buyer ordered 5 cases (60 units), a shortage of 10 units means two cases are affected, not ten. Hold a pack_size column and compute shortages in cases, then let the email say "2 cases short".
• Multi-line orders. If three lines of one order are short with different dates, merge them into one email with a small table (item, shipping now, follows on). Ask the assistant for a template with a repeating block, but check how your mail merge tool handles repeating rows, since many handle only one row per email. A workaround is a helper column that joins the lines into one text cell.
• Made-to-order or customised items. A line with the customer's own branding is not interchangeable and should not carry a substitute offer at all. Add a substitute_allowed column with yes or no, and have the variant pick itself from it.
• Prepaid orders. If the customer has paid in full, say what happens to money for the short lines (credit note, refund to the original method, held against the backorder) in terms your accounts team will honour. This is a business decision to make once, then write into the template.

The first time you handle each of these, a person should read the draft. After a dozen clean ones, the rule is stable enough to run without a second look.

Where this fits with the rest of order handling

Backorder notices are the end of a chain. Upstream, orders arrive as emailed purchase orders that someone keys in; the tutorial on reading emailed orders into a wholesaler's system covers that part. Alongside, substitutes only work if your price list is consistent, which is where the wholesale pricing tool comparison helps. Treat the three as one flow: order in, allocation, notice out. Fixing only the last step means the sales desk still spends the morning on re-keying and price checks.

Wording that protects you without sounding legal

Use dispatch dates, not arrival dates, since carriers are outside your control. Say "confirmed" only for dates your supplier has confirmed in writing, and "expected" for estimates. Say what happens if you hear nothing, but only if you will really do it. Do not promise a price hold on a substitute beyond what your terms allow.

If your trading terms deal with delays, cancellations or partial deliveries, have the assistant draft the email to fit the terms, but check consumer or trade rights with a qualified adviser rather than relying on a draft. Do not paste your full trading terms into a free consumer plan; use the relevant clause or a summary.

Keeping the template honest over time

Review the template every quarter against the replies you actually receive. Customers often tell you what was missing: "which pack size?" or "is freight still free?". Add that fact as a column and a sentence. Keep a short changelog, one line per change, so you can see which edit moved the reply rate. Compare the wording of your best-performing variant with the others and fold the winning phrases back in.

Finally, fix the cause as well as the message. If the same ten lines are short every month, the shortage is a buying problem. Feed three months of backorders into a spreadsheet and ask the assistant to rank items by how often and for how long they were short, then use that list to change reorder points. The tutorial on comparing supplier quotes with AI helps when the answer is a second source. A backorder email is a good apology; fewer backorders are better.

Backorder email questions from wholesalers

How soon should I send a backorder email?

As soon as you know the line is short, ideally when the order is confirmed or picked, not on the day it was due. Customers can work around a late delivery if they hear early. A short email sent the same morning is better than a perfect one sent three days later.

Should AI choose the promised date?

No. Calculate the date in your spreadsheet from the supplier's confirmed arrival plus your handling days, then let AI write around it. Assistants are poor at date arithmetic and cannot see your supplier's confirmation, so a date it produces is a guess.

Can I send the emails automatically?

Yes, once a few batches have been checked by a person. Start with a mail merge you review before sending. Move to an automatic trigger only for the simplest case, such as a full backorder with a confirmed date, and keep human review for substitutes and anything on a key account.

What if the supplier gives no date?

Say so plainly, give the date you will next update them, and offer a choice: wait, substitute or cancel. A promise you cannot keep costs more trust than an honest 'no date yet'.

Further reads

• AI Order Entry for Distributors: Buy Software or Build Your Own? — Order entry for distributors: buy software or build your own.
• Builders Merchant Software: Trade Accounts, Quotes and Where AI Fits — Trade accounts and quotes where AI fits in a merchant.
• How Small Shops Can Use AI to Forecast Stock and Reorder — Reorder points that prevent the same lines running short.
• AI Supplier Management: Track Prices, Lead Times, and Risk — Track the supplier lead times your promised dates depend on.
• Best AI for Ecommerce Operations: Orders, Stock, Returns, Suppliers — Orders, stock and returns as one operating picture.
• Credit Control Software for Small Distributors: AI Features to Compare — Keep payment chasing separate from stock apologies.
• How to Use AI in a Bakery Business: Wholesale and Standing Orders — Wholesale bakery workflow: one standing-order sheet, a change form for cafés, AI that reads messy messages, and reminder texts that get answered.
• Minimum Order Quantity for Wholesale: Tiered MOQs Small Retailers Accept — A tiered MOQ model with margin maths, an exceptions policy and a way to test the rules against your own order history before you publish them.
• Wholesale Nursery Availability List: AI Builds and Emails It Weekly — A weekly routine that turns a grower's stock sheet into a priced availability list, emails it to trade buyers and marks lines as sold.
• Selling Produce to Restaurants: Availability Lists and Chef Emails With AI — A weekly availability list template and four chef email scripts for growers, with AI turning your harvest notes into the list.
• Subscription Box Curation: Match Items to Surveys With AI — A 120-subscriber craft box walked through: survey fields, item tags, matching rules, stock shortfalls, manual overrides and a margin check per box.
• Wholesale Application Form: Vet New Retail Accounts With AI — The fields a wholesale application needs, how to check them, and a 12-point scoring sheet that AI completes before you decide.
• ABC Analysis for a Distributor: Rank SKUs and Customers With AI — Rank a distributor's SKUs and customers into A, B and C classes in a sheet, then attach a service rule to each class.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: Microsoft Excel and Google Sheets function documentation for WORKDAY and IF; ChatGPT, Claude and Gemini vendor pages for plan details. The email wording is illustrative.