Logistics Fulfillment Backroom INF metric?

Joined
Sep 20, 2022
Messages
269
I know this is a long shot, but I am looking anywhere I can for this data. I'll probably be emailing a few different HQ partners, but want to make sure I've done my due diligence looking myself first.

When a TM pulls a batch like a One for One or Out-of-Stock, and they tell the myDevice that the item being requested at a given location is not there, this is an INF (Item not found). The system catalogues a backroom error, i.e. a ghost, and the item is deleted from the backroom location. This activity is well recorded in Greenfield datasets like Assets_Protection_EDABI.AP_BACKROOM_DPCI_PULL_BKSTK, StoreLogistics.MOVE_BRLA_FROM_BIGRED_MYDAYMYWORK, and surely others. With the right card and filters, its generally relatively easy to identify the source of the error.

By contrast, when a fulfillment TM picks a batch with items that have backroom locations, I can't find anything that record instances of them INF'ing the backroom location. But surely that is happening, right? The FF team, on the whole, interacts more with backroom than any other team. Over time, you'd expect them to encounter instances where the items weren't backstocked correctly or other times where the TM just didn't look hard enough. In my mind, it should create a ghost just like when a puller says item not found. Yet BRLA metrics are not impacted by FF TMs.

Now, I have really looked hard for this and recently found the following dataset: FlexFulfillment.BATCHING_METRICS_PICK_ITEM_DETAIL, which looks extremely promising when examining its dimensions and metrics. I spent time formatting a card to utilize it:
FF INF.JPG

I thought great, I've got it! But as I began researching each INF activity, what I realized was that it wasn't showing INF from the backroom. Instead what it's showing are instances where an item was INFd after first having some of the quantity fulfilled from the backroom. Useless!

What I want to see is any time the picked BR quantity differed from the expected BR quantity. If anyone has thoughts on this, I would be extremely grateful.
 
Last edited:
I believe a Flex TM is more likely to skip the item if it is not in the BR location in order to find it else where. They won’t necessarily “INF” the item if its not in a particular location, because the sky is the limit on where an item can be in the store.
However if a TM does hit all items scanned when in a BR location they should appear on the BR error report…
 
It's my understanding that if Can't Find is utilized that location (salesfloor or backroom) falls into the workload of Inventory Audit, usually the same day if the list hasn't been cleared yet. Perhaps there is a card in Greenfield that follows from that angle, separating between system-generated and TM-generated items.

I have had it happen where I INF'ed an item on the salesfloor while picking a cart and then a couple hours later audited that same item while completing the Inventory Audit task list.
 
However if a TM does hit all items scanned when in a BR location they should appear on the BR error report…

And yet... It doesn't!

Dataset 'StoreLogistics.MOVE_BRLA_FROM_BIGRED_MYDAYMYWORK' has column 'all_items_scanned_true_false'. You can make a card from it and filter that column to true. It can show you the backroom location, DPCI, and who scanned it.

That is a great way to look at the AIS of TM's doing Fills, but it doesn't pull any data from ePick activity.

I'm going to explore SigningLady's suggestion, but I feel like I've been walking through green fields for about 1000 miles and haven't come across this data.

Thanks for the ideas. Keep 'em coming. ;)
 
I think the issue lies with ePick. I think that (by design) ePick was never meant to generate backroom errors, only Audit tasks - unlike the old Move app or the current Fill tasklist. There was information on Workbench years ago explaining this - this was back when ePick forced you to AIS instead of skipping or hitting "can't find."
 
I think the issue lies with ePick. I think that (by design) ePick was never meant to generate backroom errors, only Audit tasks - unlike the old Move app or the current Fill tasklist. There was information on Workbench years ago explaining this - this was back when ePick forced you to AIS instead of skipping or hitting "can't find."
I think you're right. I wonder if this was due to them not predicting how popular FF would become. I do think it means that BRLA is wholly understated though. I wonder what the ratio is between the number of ePick backroom scans vs. the number of Fill scans, for a given time period. Obviously, highly dependent on the store. I think if I dig hard enough, I can find this. I have a backroom activity card that would just need a formula to run a card calculated field on those numbers, and should return the value. Alas, that will have to wait for another day.
 
Not a fan of tm assigned INF metrics at all. How is a fulfillment tm accountable for a truck not unloaded or softlines in repacks in a secondary storage waiting its turn to be worked out or mispicks or unlocated backstock or fake tied endcaps or location tracking DPCI's in the wrong spot unable to be scanned??? A metric... cool. NOT the "fault" of a fulfillment tm.
 
Not a fan of tm assigned INF metrics at all. How is a fulfillment tm accountable for a truck not unloaded or softlines in repacks in a secondary storage waiting its turn to be worked out or mispicks or unlocated backstock or fake tied endcaps or location tracking DPCI's in the wrong spot unable to be scanned??? A metric... cool. NOT the "fault" of a fulfillment tm.
I understand your point, but it is still pretty important in assisting to identify how to decrease the overall INF %. And it may help bring visibility to some of the issues you're raising. Trucks shouldn't be acknowledged until after they're unloaded. Are there deficiencies in our logistics timelines that need to be addressed? Do we need to validate that all FF TMs know where the secondary softlines storage is? Do TMs know how to identify and correct mispicks? The fake tied end won't hurt them unless there is no backstock or home location, and no other reasonable location to go and look.

Leaders should understand that no single INF event is a cause for a concern, but instead, trends over time. What is contributing to those trends? Is it a gap in the TMs training? A systemic issues which needs to be addressed? Something else? Assigning the error to the TM helps narrow that all down.

It seems like what you may not be a fan of, rather, is unsophisticated leadership?
 
Not a fan of tm assigned INF metrics at all. How is a fulfillment tm accountable for a truck not unloaded or softlines in repacks in a secondary storage waiting its turn to be worked out or mispicks or unlocated backstock or fake tied endcaps or location tracking DPCI's in the wrong spot unable to be scanned??? A metric... cool. NOT the "fault" of a fulfillment tm.
To play devils advocate I often find items that fulfillment tms don't find. It's not that it's a fulfillment tms fault that there inf is high as there are many factors, but if a TM is consistently higher then other tms I want to know.
 
I think you're right. I wonder if this was due to them not predicting how popular FF would become. I do think it means that BRLA is wholly understated though. I wonder what the ratio is between the number of ePick backroom scans vs. the number of Fill scans, for a given time period. Obviously, highly dependent on the store. I think if I dig hard enough, I can find this. I have a backroom activity card that would just need a formula to run a card calculated field on those numbers, and should return the value. Alas, that will have to wait for another day.
This just based off my own personal experience, but it's possible they didn't fully grasp how quickly Fulfillment moves through the backroom and allowing those TMs to audit backroom locations had an unintended consequence of large numbers of locations being locu'd. If you were a backroom TM or a DBO doing your pulls, you would spend the time to scan everything in the location before hitting AIS to ensure that everything remained located, but Fulfillment TMs on a deadline would not. At my store, at least, they would hit AIS without scanning anything at all, causing a lot of issues for the backroom and BRLA.

That, and ePick being a fundamentally different process than Move.

The information you're looking for is definitely tracked and kept, it probably just hasn't been made available through a dataset yet.
 
I understand your point, but it is still pretty important in assisting to identify how to decrease the overall INF %. And it may help bring visibility to some of the issues you're raising. Trucks shouldn't be acknowledged until after they're unloaded. Are there deficiencies in our logistics timelines that need to be addressed? Do we need to validate that all FF TMs know where the secondary softlines storage is? Do TMs know how to identify and correct mispicks? The fake tied end won't hurt them unless there is no backstock or home location, and no other reasonable location to go and look.

Leaders should understand that no single INF event is a cause for a concern, but instead, trends over time. What is contributing to those trends? Is it a gap in the TMs training? A systemic issues which needs to be addressed? Something else? Assigning the error to the TM helps narrow that all down.

It seems like what you may not be a fan of, rather, is unsophisticated leadership?
While I will agree that it should be used as a tool to identify problems, it is a tool that is used to single out fulfillment tm's and not the other issues. Come review time, it is the metric attached to the tm that will be looked at. Nobody will be saying "Yeah... that's the day the truck team had 4 call ins" or "FDC got delivered to the wrong store that day". Nobody will be saying that a lax reshopper dumped a pile behind the diapers. To some of your other points, if counts aren't updated, etc. guests can order items that we haven't had delivered in weeks but have an OH of 1. I have seen orders placed where the OH is 0 and 6 are expected tomorrow. A guest who orders 10 of the same item that isn't available counts as 10 INF's and if a tm isn't savvy enough to skip rather than "can't find" at a particular occasion (thinking that will trigger a fill to that location) but is able to pick from another location, every can't find counts as an INF rather than as a source of information and that would include those fake ties. It isn't the fulfillment team's job to train the salesfloor about mispicks or to really even understand what it is, but they will "pay the price" for it happening. Wrapped pallets of softlines in the steels or in an outside storage area are not reality picks for a team on a timer and I'm pretty sure that if you understand fulfillment or are in a leadership role you understand that and this is a devil's advocate point with no basis in real time. Your points are valid, but they are not a reason for a team member to take the hits from the opportunities that other teams have. You are right; I am not a fan of unsophisticated leadership, but a bigger fan of not using metrics that are not solely due to a single tm's performance to evaluate that performance. (Just for reference, while not on the fulfillment team, I help out several days/week, train fulfillment tm's, and have the highest pick numbers with an INF rate below 3%. The above are observations and have been discussed with the leadership team.)
 
Last edited:
To play devils advocate I often find items that fulfillment tms don't find. It's not that it's a fulfillment tms fault that there inf is high as there are many factors, but if a TM is consistently higher then other tms I want to know.
Once you know, do you delve into the reasons and follow up on that or simply use it to coach the fulfillment tm. telling them that the number is too high? Do you go back to the sf leaders and let them know where problems lie so that they can also follow up with their teams?
 
Once you know, do you delve into the reasons and follow up on that or simply use it to coach the fulfillment tm. telling them that the number is too high? Do you go back to the sf leaders and let them know where problems lie so that they can also follow up with their teams?
When I ran fulfillment I spent more time on it. I know for me it was often a training issue and some bad habits. Tms relying on pictures more then reading labels or not having enough time to properly look for an item where the most common reasons. If the reasons for high inf is that the products are stuck in the backroom for a month then there isnt much anyone can do about it. My core group of gm leaders work pretty closely together so many mistakes found are often remedied pretty quickly.
 
While I will agree that it should be used as a tool to identify problems, it is a tool that is used to single out fulfillment tm's and not the other issues. Come review time, it is the metric attached to the tm that will be looked at. Nobody will be saying "Yeah... that's the day the truck team had 4 call ins" or "FDC got delivered to the wrong store that day". Nobody will be saying that a lax reshopper dumped a pile behind the diapers. To some of your other points, if counts aren't updated, etc. guests can order items that we haven't had delivered in weeks but have an OH of 1. I have seen orders placed where the OH is 0 and 6 are expected tomorrow. A guest who orders 10 of the same item that isn't available counts as 10 INF's and if a tm isn't savvy enough to skip rather than "can't find" at a particular occasion (thinking that will trigger a fill to that location) but is able to pick from another location, every can't find counts as an INF rather than as a source of information and that would include those fake ties. It isn't the fulfillment team's job to train the salesfloor about mispicks or to really even understand what it is, but they will "pay the price" for it happening. Wrapped pallets of softlines in the steels or in an outside storage area are not reality picks for a team on a timer and I'm pretty sure that if you understand fulfillment or are in a leadership role you understand that and this is a devil's advocate point with no basis in real time. Your points are valid, but they are not a reason for a team member to take the hits from the opportunities that other teams have. You are right; I am not a fan of unsophisticated leadership, but a bigger fan of not using metrics that are not solely due to a single tm's performance to evaluate that performance. (Just for reference, while not on the fulfillment team, I help out several days/week, train fulfillment tm's, and have the highest pick numbers with an INF rate below 3%. The above are observations and have been discussed with the leadership team.)

Individual TM infs do tell you a lot though in large numbers. Fairly basic statistical analysis is all that's needed. Over a large enough sample size of units attempted, all of the bad breaks and unlucky batches will even out between TMs. They all run into the same shit eventually. Statistically, you would expect that given enough units attempted, your pickers should all regress to the mean. Essentially, they should all be fairly close to the store average for the time period. But, there might be TMs who are significantly below or above the average. Figuring out why they are is important.
 
every can't find counts as an INF rather than as a source of information and that would include those fake ties
INF is only calculated based on the quantity of items you picked vs the quantity of items needed. It has nothing to do with multiple locations. "Can't Find" will only generate an INF when it is the last existing location on your batch. Example: If you need one pack of out-of-stock batteries that are tied to 6 different sidecaps around the store, you will eventually still have to "can't find" all 6 locations, but it will only generate an INF for the one pack you needed - not 6.
 
INF is only calculated based on the quantity of items you picked vs the quantity of items needed. It has nothing to do with multiple locations. "Can't Find" will only generate an INF when it is the last existing location on your batch. Example: If you need one pack of out-of-stock batteries that are tied to 6 different sidecaps around the store, you will eventually still have to "can't find" all 6 locations, but it will only generate an INF for the one pack you needed - not 6.
Not what I meant. If there are no cookies and the guest wants 10 cookies, that's 10 INFs.
 
INF is only calculated based on the quantity of items you picked vs the quantity of items needed. It has nothing to do with multiple locations. "Can't Find" will only generate an INF when it is the last existing location on your batch. Example: If you need one pack of out-of-stock batteries that are tied to 6 different sidecaps around the store, you will eventually still have to "can't find" all 6 locations, but it will only generate an INF for the one pack you needed - not 6.
It used to misreport in MPM. It's been fixed for a while now though.
Not what I meant. If there are no cookies and the guest wants 10 cookies, that's 10 INFs.
Maybe we should have an additional INF metric, by DPCIs cleared, not units.
 
Target won't do that. INF exists to represent missed sales (OPU) and misplaced inventory. If you INF'd 10 cookies, you failed to sell/locate 10 cookies, not 1 cookie.
And somehow a fulfillment tm is accountable for that. Yes, there will be follow up to the other areas, yet we will never see a report that Susie Domestics didn't backstock that item correctly or that Timmy Stocker didn't catch a mispick and it resulted in a direct sales loss and Boy-o-boy! You are better than that! Still not convinced that tm assigned INF metrics should be a thing. INF metrics? Absolutely. Laying "blame" on a tm who doesn't ever touch the item until it is requested by a guest who isn't even in the store? No.
 
And somehow a fulfillment tm is accountable for that. Yes, there will be follow up to the other areas, yet we will never see a report that Susie Domestics didn't backstock that item correctly or that Timmy Stocker didn't catch a mispick and it resulted in a direct sales loss and Boy-o-boy! You are better than that! Still not convinced that tm assigned INF metrics should be a thing. INF metrics? Absolutely. Laying "blame" on a tm who doesn't ever touch the item until it is requested by a guest who isn't even in the store? No.
Corporate knows that, at least to an extent. It's why there isn't an official card on Greenfield that gives TMs their own BRLA. They understand that the errors that are "assigned" to TMs are just those the TM found but didn't cause. It's also why INF has a threshold.

If something has an OH greater than 0, it should either be on the floor where it's located or in the backroom, also located. If it isn't, then the system failed the TM looking for it. The DC didn't send it, Inbound didn't charge it out or sort it properly, GM didn't push it (right), AP didn't track it, Flex/GS didn't RTS it, etc... The list goes on and on before even reaching the Fulfillment TM, but it's ultimately easier to track who INF'd it than it is to track all those other things, much less
"fix" them. That's why the Fulfillment TM takes on the burden of blame. It's easier to coach one TM than it is to coach the entire store for not 100% completing their processes.
 
Corporate knows that, at least to an extent. It's why there isn't an official card on Greenfield that gives TMs their own BRLA. They understand that the errors that are "assigned" to TMs are just those the TM found but didn't cause. It's also why INF has a threshold.

If something has an OH greater than 0, it should either be on the floor where it's located or in the backroom, also located. If it isn't, then the system failed the TM looking for it. The DC didn't send it, Inbound didn't charge it out or sort it properly, GM didn't push it (right), AP didn't track it, Flex/GS didn't RTS it, etc... The list goes on and on before even reaching the Fulfillment TM, but it's ultimately easier to track who INF'd it than it is to track all those other things, much less
"fix" them. That's why the Fulfillment TM takes on the burden of blame. It's easier to coach one TM than it is to coach the entire store for not 100% completing their processes.
Corporate may know that but it's the lead that writes the reviews that remembers WHO they talked to about INF and more likely than not will not recall that day's circumstances. My mind is not changed because it's "easier".
 
Back
Top