BonoficiGrade your business
Menu
Lesson 4 of 5 · 6 min read

Document the exceptions: a process documentation template

How to find and document the exceptions in a workflow: six common types, using contradictions between teams, an evidence rule and an exception log template.

To document a workflow properly, record the exceptions alongside the happy path: what goes wrong, how often, what people do about it and who decides. Find them by asking the people who do the work about the last time something broke, and by comparing what different teams say about the same hand-off. Keep them in an exception log with an owner and a review date, and record only what someone doing the work has told you.

The happy path is half the process

Most process documentation describes the happy path: the order arrives complete, the stock is there, the price matches, the invoice goes out. On a good day that is what happens. But the time, cost and frustration in a process usually sit in the other cases: the order with no PO number, the delivery that arrives short, the customer who needs it tomorrow.

Those cases are handled by the layer of habits around your systems: the spreadsheet, the phone call, the colleague who just knows. If your documentation leaves them out, three things happen. New starters learn them by making mistakes. The person who knows leaves and takes them along. And any automation you build stops at the first exception and hands the work back with no context.

Common types of exception

TypeWhat it looks likeTypical workaround (example)
Missing dataOrder without a PO number; invoice without a delivery reference; new customer with no accountEmail the customer and park the order in a folder
Numbers that don’t matchInvoice price differs from the order; quantity delivered differs from quantity orderedCheck with the buyer, then approve or query (see the three-way match guide)
Urgent customerSame-day delivery request; a key account jumping the queuePhone the warehouse directly and skip normal order entry
Late or short supplierGoods not arrived, partial delivery, substitute productSplit the order, ship what is there, track the rest in a spreadsheet
System down or slowERP unavailable, portal timing out, overnight integration failedTake orders in Excel and enter them later
Policy edge casesDiscount above the usual limit, credit limit exceeded, return outside the windowAsk a named manager, often by chat

Not every exception is worth documenting. A practical rule: write it down if it happens at least monthly, if it costs real money when handled wrong, or if only one person knows how to handle it.

How to find the exceptions

Ask about the last time, not the usual

“How do you process an order?” gets the happy path. “Tell me about the last order that didn’t go smoothly” gets the exception. Ask the people doing the work, one to one, about specific recent cases: what happened, what they did, who they asked, where they wrote it down and how often that kind of thing comes up.

Look for contradictions between teams

The best signal of an undocumented exception is two teams describing the same hand-off differently. Sales says delivery dates are confirmed in the order confirmation. Logistics says they confirm them by phone the day before, because the dates in the confirmation are often wrong. Neither is lying, but the step between them has no owner, and that is where work gets done twice or not at all.

When you find a contradiction, do not pick a winner. Write down both versions, with who said each, and resolve it with both teams in the room. The resolution usually names an owner for the step, which is the fix in itself.

Follow the side channels

Shared spreadsheets, chat groups, sticky notes, an Outlook folder called “pending”: each one exists because the official process did not handle something. Ask what goes into them and why.

The exception log

An exception log is a simple table that sits next to the process map or SOP: one row per exception, linked to the step where it happens. It turns “what we do when it goes wrong” from folklore into something you can review, train people on and, later, automate.

Exception log template

Process: [e.g. Customer order, email to invoice] · Log owner: [role]
Last reviewed: [date] · Next review: [date]

IDStepExceptionHow oftenWhat people do nowWho decidesCost if handled wrongSource (who, when)Status
E-013. Enter sales orderOrder email has no PO numberWeeklyEmail the customer; hold the order in a “Pending PO” folderSales coordinatorCustomer rejects the invoiceSales coordinator, [date]Confirmed
E-025. Pick and packSpreadsheet shows stock the warehouse cannot findSeveral times a weekPicker tells sales by chat; sales phones the customer to offer a substituteSales manager approves substitutesLate delivery, complaintWarehouse lead, [date]; sales says it is rareContradiction: open
E-037. Raise invoiceDelivered quantity differs from the orderMonthly[ ][ ][ ][ ]Unconfirmed

Status values: Unconfirmed (nobody doing the work has said it yet) · Confirmed · Contradiction: open (two versions, both recorded) · Resolved (owner and rule agreed; update the SOP)

Who owns the documentation, and when to review it

Documentation goes stale because nobody owns it. Give each process a single owner, usually the lead of the team that does most of the steps, whose job is to keep the map, the SOP and the exception log true. Not to write everything, but to make sure it gets updated.

Review it when something changes, and on a calendar:

  • When a system changes: a new ERP module, a renamed field, a new supplier portal.
  • When a new exception turns up more than once.
  • When someone new follows the SOP and gets stuck.
  • When a customer complaint or audit finding traces back to the process.
  • Otherwise every six months, by walking one recent real case through the documents.

Keep a short change log at the bottom of each document: what changed, when and who said so. That is the same evidence rule, applied over time. The SOP template in lesson 3 already has one.

With the exceptions written down, you can see which steps follow rules and which need judgement. That is the starting point for moving from a process map to automation.

For an outside view of where your exceptions are likely to cost the most, the free Bonofici grader reads your company’s public pages and shows where to start.

Start with what your team already knows.

Book a 30‑minute call.We come back with the first three tools worth building, in writing.

Not ready for a call?

See where your company stands and which jobs are worth the most, from your public pages.