How deep do you go with standard cost variances

[Elliott.sam]

There have been some regular variances showing up between our standard and actual costs mostly on the materials side, but a bit in labor too, just enough to make me question where to just say jack on this kind of matters and move on or whether I should start poking around, If youre dealing with standard costing regularly, how do you decide what is worth investigating? like are there specific thresholds for you, or do you go case by case based on context? not turning into a variance detective or something lol

 

[SlightlyLost]

There have been some regular variances showing up between our standard and actual costs mostly on the materials side, but a bit in labor too, just enough to make me question where to just say jack on this kind of matters and move on or whether I should start poking around, If youre dealing with standard costing regularly, how do you decide what is worth investigating? like are there specific thresholds for you, or do you go case by case based on context? not turning into a variance detective or something lol

It's really about balancing the time spent investigating against the potential savings or insights gained. Most companies use a two pronged approach, like
Dollar thresholds which is your first line of defense. You set a specific dollar amount (e.g., "$1,000 variance") or a percentage (e.g., "variance greater than 5% of standard cost"). Anything below that threshold gets ignored or rolled up into a general "unexplained variance" bucket. This prevents you from chasing pennies.
Also contextual exceptions. Even with thresholds, some variances always warrant a look, like new products/processes, persistent small variances, and high volume items
It really saves you from getting lost in the weeds if you automate the flagging of anything over your threshold and then manually review based on context

 

[chillywilly]

Yeah Ive been there. Best way is to set a simple threshold, like a % or dollar amount, and only dig in when a variance crosses that line. I also look at trends... so if something keeps popping up, even if small, its probably worth a quick check. No need to chase every little blip unless it starts adding up or points to a bigger problem

 

[doodlebun]

It's really about balancing the time spent investigating against the potential savings or insights gained. Most companies use a two pronged approach, like
Dollar thresholds which is your first line of defense. You set a specific dollar amount (e.g., "$1,000 variance") or a percentage (e.g., "variance greater than 5% of standard cost"). Anything below that threshold gets ignored or rolled up into a general "unexplained variance" bucket. This prevents you from chasing pennies.
Also contextual exceptions. Even with thresholds, some variances always warrant a look, like new products/processes, persistent small variances, and high volume items
It really saves you from getting lost in the weeds if you automate the flagging of anything over your threshold and then manually review based on context

Tthanks for breaking it down! Question for contextual exceptions though, do you track those separately or flag them in the same system?