Order Backlog Dashboard

What this tool does

CM8-272 turns your open order book into a dashboard. You keep one row per open order — line level is fine if that is how you sell — with its value, the date it arrived, the date you promised, where it sits in production and who owns it. The tool shows the backlog as money: total value, value past its promised date, value at risk of joining it, value stuck waiting for material, and how the book is spread across customers and margin bands.

The backlog deserves that attention because of what it is: confirmed demand you have not yet delivered. Every row is a promise a customer is holding you to, and every row is revenue that has not yet become an invoice. Read one way, the order book is a promise ledger; read the other way, it is a cash forecast. A business that watches its backlog knows three or four weeks early what its delivery performance and its bank balance are about to do. A business that does not finds out from an angry phone call and a quiet month.

Everything runs inside this single file. There is no account, no upload and no network request of any kind, so your order values, customer names and margin positions — some of the most commercially sensitive numbers you hold — never leave the computer you are using.

Reading the delivery wall

The first chart buckets the open value by the week it is due: everything already overdue in a red bar at the left, then the next eight weeks. Most order books are lumpy, and the lumps are the point. A wall of value standing in week two means one of two things is coming: overtime, or conversations with customers about dates. Both are far cheaper planned this week than discovered in week two — overtime arranged in advance costs a premium; overtime discovered on the day costs a premium plus the jobs that were bumped to pay for it, and a conversation offered early is a courtesy while the same conversation forced late is an apology.

Use the wall forwards, too. Empty weeks four to eight are not a holiday — they are the sales pipeline arriving in production. If the wall slopes down to nothing, the workload cliff is visible eight weeks before the quiet hits the shop floor, which is exactly enough time to do something about it.

The risk bands, precisely

Every order carries one of three bands, computed the same way everywhere in the tool:

Note what “at risk” deliberately is not: it is not a schedule. An order due in five days that is in production may still be tight — the tool cannot know your cycle times. What it can know is that an order due in five days that nobody has started is in trouble by arithmetic alone, and that is worth flagging without asking anyone's opinion. Widen or narrow the window on the Settings tab to match your typical lead time: a machine shop with two-day cycles might use three days; a fabricator with three-week jobs might use twenty-one.

The delivery risk list on the Reports tab collects both bands, sorted by promised date, with the blocked reason and the owner alongside — it is the expedite meeting, pre-built.

Aging honestly

The age of an order is the days since you received it, and the average-age tile is the quiet health check of the whole book. A backlog that ages is not just slow — old orders are where problems hide. An order that has sat for ninety days has usually stopped moving for a reason nobody wrote down: a confirmation that was never sent, a spec question the customer never answered, a casting on a lead time everyone forgot, a price query parked in someone's inbox. The order book does not forget, but people do — and the customer certainly does not.

When the average age climbs month on month while the order count stays flat, work is entering faster than it is leaving and the promises are quietly stretching. Pull up the oldest five orders and ask one question of each: what, exactly, is it waiting for? Write the answer in the blocked reason field. An order with a named blocker is a task; an order without one is a mystery.

Customer concentration

The customer chart and table show whose money the backlog is. In the sample book, one customer holds forty percent of the open value — a common shape, and a double-edged one. That customer is simultaneously your most important delivery relationship and your largest single risk: their orders deserve the best service in the book, because losing them removes almost half of your forward workload in one decision you do not get a vote on. The right response is both things at once — service the concentration brilliantly, and diversify it deliberately. Neither substitutes for the other: diversification without service loses the anchor customer before the replacements arrive, and service without diversification just makes the dependency more comfortable.

The share column in the customer table gives you the number for the conversation. If one name is drifting up past a third of the book, that is a sales strategy discussion, not a production one.

Awaiting material — the purchasing handshake

The “awaiting material” tile is written for one meeting: the one between production and purchasing. It is the value of orders that production cannot touch until something arrives, which makes it the clearest possible statement of what purchasing's next phone calls are worth. A weekly routine of reading that tile together — here is what is stuck, here is what it is worth, here is what unlocks it — replaces the usual argument about whose fault the late order was with a shared list of what to chase. Put the specific shortage in the blocked reason field so the list names parts, not just orders.

Margin bands

Each order carries a margin band — a deliberate four-step judgement, not a costing exercise. The band does two jobs. First, the stacked chart crosses margin against delivery risk and answers a question worth asking every month: are the loss-makers also the late ones? Work that loses money and ships late is being paid for twice — once in price and once in goodwill — and it is usually the same legacy contract doing both.

Second, the “loss-making — review” band is a flag you set on purpose, which is why the tool never blocks it: recording a loss-maker honestly is the useful act. The right sequence is almost always ship it, then fix the pricing. The order in the book was accepted at that price and your name is on the promise; delivering it late or grudgingly turns a pricing mistake into a reputation problem. But the next order at that price is a choice. Use the band to make sure the renewal conversation happens before the next loss-maker lands in the book.

Confirmations discipline

The confirmation checkbox records that you acknowledged the order back to the customer with a date. It sounds like paperwork; it is actually the moment the promise becomes yours. An unconfirmed order is the customer's plan — the date in their system is whatever they asked for, and the first anyone discovers the difference is when it does not arrive. Confirming is the act that replaces their assumption with your commitment, and it is also your one clean chance to move a date you cannot meet before anyone has planned around it.

The average-age tile counts unconfirmed orders for exactly this reason. The discipline worth having is simple: nothing sits in the book unconfirmed for more than two working days. Either send the acknowledgement, or ring up and renegotiate the date — but do one of them.

The weekly backlog review

Fifteen minutes, same time every week, in this order:

1. Overdue first. Open the delivery risk list. For every overdue order: what ships this week, and who has told the customer? An overdue order the customer has not heard about is the worst item in the building.
2. The wall second. Look at the next two or three bars of the due-week chart against the capacity you actually have. If a bar will not fit, decide now what moves — the work, the hours or the date — and make the calls this week.
3. Material third. Read the awaiting-material tile with purchasing. Chase what unlocks the most value first, not what was ordered first.
4. Sweep the rest: confirm the unconfirmed, put a blocked reason on anything stalled, and glance at the average age. Close by adding this week's new orders so next week's meeting starts true.

FAQ

One row per order or per line? Whichever level you promise dates at. If a ten-line order ships as one delivery, one row. If lines ship separately with their own dates, give each line a row — the backlog is a list of promises, and each promise needs its own date.

What value should a partially shipped order carry? The open balance — the value not yet delivered. The backlog is what you still owe, so reduce the row's value as parts ship. The sample's partially shipped row carries only its remaining balance for exactly this reason.

When does an order leave the book? When the last of it is delivered — not when it is invoiced, and not when it is finished. “Finished — awaiting dispatch” exists precisely because finished-but-not-delivered is still a promise outstanding, and often an invoice you cannot yet raise.

Is backlog value the same as revenue forecast? It is the confirmed part of one. The backlog tells you what will invoice if you deliver to the promised dates; it says nothing about orders not yet won. Read the due-week chart as the floor of the next two months' invoicing, not the ceiling.

Should quotes or unconfirmed enquiries go in? No — only orders the customer has placed. A book padded with hopes stops being a promise ledger. The confirmation checkbox tracks whether you have confirmed back, which is a different thing from whether the order is real.

Saving your work

Orders, settings and the report header are written to this browser's local storage as you type, and the toolbar shows the time of the last save. That storage belongs to one browser on one computer: another browser, a private window, a second machine or a clean-up tool that clears site data will not have it.

Treat Export .json as the real save — one file containing everything, which Import .json restores anywhere. Export CSV gives you the order book for spreadsheet work. Reset asks twice, then erases everything this tool has stored. There is no undo. An order book with values and margin bands is commercially sensitive — treat exports accordingly.

Accuracy & disclaimer

The arithmetic here is simple and stated in full — sums of values, differences between dates — and the tool does it faithfully. Everything that matters sits underneath: whether every open order was entered, whether values are the open balances, whether statuses were updated when the work moved, and whether the promised dates are the ones the customer actually holds. The dashboard shows the promises that were recorded; an order nobody entered is still a promise, and it will still be late.

The risk bands are date arithmetic, not a production schedule, and the margin bands are your own judgement, not a costing. This is a record-keeping and prioritisation aid for running your order book — it is not an ERP, not a scheduling system, and not financial advice.