Why failed prints are easy to ignore
A failed print is not just a bad part in the trash. It already used material, printer time, electricity, and sometimes your attention. By the time you peel it off the bed or pull it out of a spaghetti mess, the cost has already happened.
Still, failed prints are easy to leave out of the numbers. The successful part is what gets sold, packed, photographed, or delivered. The failed attempt feels like an interruption rather than part of the job. So when people calculate profit, they often use the clean version of the print: one part, one filament estimate, one selling price.
That makes the numbers look better than the workshop actually feels. If a product sells well but fails often, the slicer estimate will keep saying it is profitable while the printer keeps spending hours producing parts you cannot sell.
Why failures matter financially
Every failed print consumes something. Filament is the obvious one, but it is usually not the whole loss. The machine was occupied, the bed was heated, the hotend was running, and another job could not use that printer during the failed attempt.
Some failures also consume labor. You may need to clean the bed, remove stuck material, cut away a blob, change a nozzle, dry filament, restart the job, adjust slicer settings, or inspect the printer before trusting it again. A failure that takes thirty seconds to clear is different from one that turns into an hour of troubleshooting.
This is why ignoring failures creates bad pricing decisions. A product with a high failure rate may look profitable when measured only by successful prints. Once failed attempts are included, the real margin can shrink or disappear.
The painful part is that the mistake is usually quiet. Nothing in the order total tells you that two earlier attempts failed. Unless you track it, the loss is spread across the day and hidden inside your time.
Cost per attempted print vs cost per successful part
The most useful metric is cost per successful part, not cost per attempted print. Attempt cost is still useful, because it tells you what one run consumes. But profit depends on what it costs to produce something sellable.
Imagine a part that costs $1.50 in material and machine cost per attempt, and sells for $6.00. If every attempt succeeds, the margin looks comfortable. If one out of every four attempts fails, the math changes. Four attempts produce three sellable parts, so the cost of the failed attempt has to be spread across the three successful ones.
That is the number that matters when pricing repeated work. The customer is not paying for attempts. They are paying for finished parts. If the process needs extra attempts to produce those parts, the finished parts have to carry that cost.
This also explains why a low failure rate can still matter at volume. A five percent failure rate may not sound serious, but across hundreds of prints it becomes a steady drain on filament, machine capacity, and attention.
Why farms need product-level tracking
Printer-level tracking is helpful, but it does not answer the whole question. If Printer A has three failures this week, you know Printer A needs attention. But if the same file failed on three different printers, the problem may not be the printers at all. It may be the model, the slicer settings, the filament, or the way the part is oriented.
A farm needs to group failures by product or file, not only by machine. That is how you find the parts that quietly waste time across the whole setup.
This matters when the same product is printed wherever capacity is available. Monday's batch might run on two enclosed machines. Tuesday's batch might be split between a fast printer and a backup printer. If you only look at each printer separately, the product-level pattern is hard to see.
The useful question is not just "which printer failed?" It is also "which product keeps causing failures, and what does that do to profit?" A reliable printer can still lose money on a badly tuned part. A troublesome printer can also make every product look worse than it really is.
What profit tracking should include
Profit tracking does not need to be complicated, but it needs the right pieces. You need the selling price, the quantity of successful parts, the real production cost, and the failed attempts that belonged to that product or order.
Real production cost should include material, machine time, power, maintenance assumptions, and any labor that is part of the workflow. The exact model can be simple, as long as it is applied consistently. A rough but consistent model that includes failures is usually more useful than a precise filament-only number.
Quantity matters because one failed attempt can affect a batch differently depending on the order size. One failure while making two parts is a very different problem from one failure while making fifty. Grouping cost by product and batch helps avoid overreacting to noise while still catching patterns.
Selling price matters because failures do not hurt every product equally. A high-margin decorative item may absorb an occasional failed print. A low-margin functional part may become pointless if it needs frequent reprints or too much cleanup.
Using failure data to improve production
Failure tracking is not only about pricing. It is also one of the fastest ways to find operational problems.
If failures cluster around one material, the filament may need drying or the profile may need work. If they cluster around one printer, the issue might be bed adhesion, extrusion, cooling, motion, a worn build surface, or inconsistent first layers. If they cluster around one file across several machines, the part itself probably needs attention.
The best failure records are simple but specific. "Failed" is better than nothing, but "warped corner," "support broke loose," "nozzle clog," or "part detached at hour three" is much more useful. Over time, those notes show which problems are random and which are repeatable.
A short review habit helps. Look at the parts with the most failures, compare them against profit, and decide what to do next. Sometimes the answer is tuning. Sometimes it is changing material, slowing the print, redesigning the part, increasing the price, or deciding the job is not worth offering anymore.
This is the kind of tracking My3DMonitor can automate for a Klipper-based farm, but the same logic works in a spreadsheet if the setup is small enough and the entries are kept consistently.
Final thoughts
Failed prints are part of production, not exceptions outside the accounting. They consume real resources, and they change the cost of the parts that do succeed.
If you only calculate successful prints, your profit will usually look cleaner than reality. The more printers you run, the easier it becomes for small losses to hide inside the daily workflow.
Tracking failures by product, file, printer, and batch gives you two benefits at once. Pricing becomes more honest, and the workshop gets easier to improve. You can see which jobs are worth repeating, which machines need attention, and which parts are costing more than they appear to.