web
You’re offline. This is a read only version of the page.
close
Skip to main content

Notifications

Announcements

No record found.

Community site session details

Community site session details

Session Id :
Microsoft Dynamics AX (Archived)

Counting best practices

(0) ShareShare
ReportReport
Posted on by

Hi,

We need to redo an inventory count. My question is what the best practice is in AX? Use cycle count plan or a counting journal? And how to start practically to avoid wrong data back from the counters.

J.

*This post is locked for comments

I have the same question (0)
  • Verified answer
    Brandon Wiese Profile Picture
    17,788 on at

    Good counting controls have very little to do with AX at all.  There are some AX gotchas, like the fact that the on-hand numbers includes Picked inventory.  That means if you have an outstanding picker, say for 100 of an item, but you haven't yet removed that 100 from the shelf, you'll be counting inventory that the system isn't including in the on-hand, and thus adding 100 in error.  And of course, good physical controls, such as not counting inventory that's on the dock but not yet received, or inventory that's on the truck already packing slip updated, as so not to count inventory that isn't part of your on-hand yet or anymore.

    As far as counting errors, which is to say errors that result FROM poor counting and not corrections to inventory that are a result of proper counting, there are simple things you can do.    This all assumes you're counting by hand.  I don't see how you can do a count by barcode scanner and find yourself having to do it over.

    Deploy your counters in teams of two, but NOT so that one can count while the other writes.  BOTH count, and when they agree on a number, then they write.  You want two independent counts on everything that agree.  Second, follow up all teams and double check their first 20, 30, or 50 counts yourself.  Some people just cannot count.  And by that I don't mean they can't add or multiply, I mean they just cannot pay attention to details, and they double count, or miss items, or mistake part numbers, or transpose their answers, etc.  

    You have some idea of the accuracy of your warehouse on-hand, let's say it's 99%, which means you have an idle error rate of about 1% at any point in time.  If you follow-up count a team and find 5 mistakes in the first 50 counts for that team, that's 10%!  That team is generating more errors in your inventory by counting poorly than they could possibly fix by counting at all.  Send them home.  Counting poorly is almost always worse than not counting at all.

    In my experience, the biggest mistake most companies make on inventory counting is to throw every employee on staff in the warehouse and assume that everyone can do it.  If they don't know your product numbers, they're far more likely to confuse like products with similar numbers.  If they're not familiar with the layout or physical controls, they're far more likely to count written-off inventory that should be ignored or count quarantined stock as if it were main stock.  The pressure to get it done quickly often outweighs the need to get it done accurately.

    The second biggest mistake companies make is to pretend they can do a controlled physical inventory all while shipping and receiving, that they can count with inventory in motion and keep it all straight.  Most of the time, they cannot.

  • Community Member Profile Picture
    on at

    Great answer. Thank you. Regarding the 'picked' I did not know. That implicates we need to cancel all work of the WHS module and cannot count that is under production.

    Technically, do we need the counting journal or the cycle/ spot count? And how does that  work in the warehouse without creating to much data entry which takes time and can also cause errors.

  • Brandon Wiese Profile Picture
    17,788 on at

    I'm not familiar with how the new WHS module affects on the on-hand numbers on counting journals.  You should definitely test that to understand how it works.  But it does illustrate the underlying problem, that counting is fundamentally comparing a real world number against a system generated number, and you need to make sure it's apples to apples.  You shouldn't count that inventory sitting on the dock you haven't received yet, that's obvious.  But other forms of inventory in motion are exactly the same problem but harder to see.

    I worked for a company that liked to post its packing slips early for customer pick up orders (before the customer had arrived), even though the inventory was still sitting on the shelf, presumably as a means of saving time or allocating the stock or whatever bad reason they had for doing that.  Do you think that ever resulted in a bad count, i.e. counting something on the shelf that the system had deducted already?  Of course it did.

    I was having a discussion a week ago about how to measure counting accuracy.  I don't mean the accuracy of your total on-hand quantity or value, which is how much counting as a function corrects your inventory.  I mean how much of your counting corrections are clearly errors in themselves.  How can you tell?  When you see a count adjustment of -100 for an item, and then 2 weeks later you see another count adjustment for +100 for the same item in the same place, that's a counting error followed by a counting correction.  You might as well have not counted at all.  Certainly in those 2 weeks your purchasing department could have reacted to that mythical shortage and it's possible you've spent good money replenishing stock you had all along.  

    I theorize that you could measure your counting error by measuring counting adjustments that offset other counting adjustments, like in my example.

    Anyway, I'm highly opinionated on this issue.  I counted and reconciled inventory counts for years and I've seen all the mistakes.

    I don't know what you mean by cycle/spot count?  Is that just a counting journal where you count by location/area rather than by item or on-hand?

  • Brandon Wiese Profile Picture
    17,788 on at

    Production is entirely another issue.  It's important to understand what Work in process really is to your system.

    For example, you have a production order that has already been picked or pre-flushed, which consumes the raw materials, and those materials are now sitting on the line somewhere.  You haven't reported as finished yet.  What's in process at that point is pure money, and not on-hand quantity you can count.  When you picked the components, the quantity was deducted and the offset in pure dollars (or whatever your currency happens to be) went into a WIP main account.  When you eventually do report as finished, quantity will be added and the offset in pure dollars drawn from the (presumably) same main account.  But while it's in process, it's not inventory anymore, and if you count it and compare that number to an on-hand number, that's a mistake.  It's a problem of physical controls, knowing what is inventory and what is not.

    Now note how easily the opposite problem occurs.  Say you have a production order with components in process, but you're back-flushing, and since you haven't reported as finished yet, the system proposes an on-hand number that includes inventory that might literately be in the machine and not countable.

    Finally consider how using barcode scanners to count solves none of these issues.

  • Community Member Profile Picture
    on at

    Hi,

    Thank you Brandon,

    We did a stock count pair-by-pair with workers already working at inbound and quality. Count independently and compare. Then recount again if a mismatch was there. For large quantities the difference was accepted due to time constraints and inability to count a lot of metal in a wooden box...

    Luckily we provided this instruction to search for items not on the counting list (blind list from Excel). It turned out that we listed not all locations, some locations had no transactions upon them since last count and those might have been missed. The workers had written down all these locations.

    Unfortunately, it appears they write down the count from a previous count although the plate indicating the quantity from last count was incorrect since other workers picked from them.

    Your help and input was highly appreciated. The count pair-by-pair was an eye-catcher.

    J.

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Responsible AI policies

As AI tools become more common, we’re introducing a Responsible AI Use…

Neeraj Kumar – Community Spotlight

We are honored to recognize Neeraj Kumar as our Community Spotlight honoree for…

Leaderboard > 🔒一 Microsoft Dynamics AX (Archived)

#1
Martin Dráb Profile Picture

Martin Dráb 4 Most Valuable Professional

#1
Priya_K Profile Picture

Priya_K 4

#3
MyDynamicsNAV Profile Picture

MyDynamicsNAV 2

Last 30 days Overall leaderboard

Featured topics

Product updates

Dynamics 365 release plans