Cash Application and Unique Customer Payment Behaviors

Cindy Scott:

I can’t feel like I’m making an entrance, this is weird. I was on this podium yesterday and it’s really high, it makes my heartbeat and it creaks so I won’t be moving around a lot. I am probably going to put my glasses on today so I can actually see what I’m talking about and my main goal after last night is to make sure that you stay awake so hopefully throughout this you’ll get a little bit of information that is helpful as well.

Cindy Scott:

So with that said, let’s talk about the Cash Application and BlackHawk. There it is, so I’m just going through the agenda real quick.

Cindy Scott:

We’re going to talk a little bit about BlackHawk because some of you have no idea what that is. We gonna talk about our cash application at BlackHawk after numerous acquisitions and not interrogating those acquisitions very well. We have a lot of unique payment scenarios that I think we’ll spend most of my time speaking to because I was unsure when we first were looking at HighRadius as a solution whether it could accommodate our weirdness. We’ll look at some of the results since we’ve implemented about a year ago on the cash application and then some of our lessons learned and best practices.

Cindy Scott:

So, a little bit about BlackHawk: headquartered and present in California, a little over a thousand employees across 24 countries, 25 or 26 now and we are not helicopters. So BlackHawk is the engine behind the settlement of gift card transactions. So when you go to your local grocery store and buy a Macy’s gift card, we are the financial settlement aspect of that gift card which gets activated at the POS system there at your grocery store and goes through our settlement system to give the grocery store their piece of the pie, Macy’s gets their cut and of course, BlackHawk keeps a little for providing the service. There’s a lot of complex transactions around the printing, distribution, merchandising, and marketing of those gift cards that are handled on our nation settlement side of the world. So, the settlement is all of the gift card activation stuff. Non-settlement is all the other trade-related stuff like the marketing, promotions, distribution and shipping kind of things.

Cindy Scott:

We also have an incentive side of our business that is growing pretty quickly. It’s around employee rewards where you know companies buy gift cards in bulk or they have a gift card catalog that you get rewards, you can build up points, go out and redeem those very gift card, things like energy companies that give rebates for you know installing solar and that kind of thing as well as we have several standalone platforms at this point where we’re working on consolidating them where they are branded company but they’re BlackHawk engines an example is with Kroger you can go on their site if you’re, for instance, a charity, school or church and order bulk gift cards and you’re ordering them from Kroger you’re getting Kroger gift cards but BlackHawk is doing all of the order activation shipment and redemption tracking for those gift cards. So that’s a little bit about BlackHawk.

Cindy Scott:

We have our global shared service center, it’s in Phoenix Arizona. We’re just now starting to consolidate that into a true shared service center. My team handles all of the AR billing and cash app for the U.S. and Canada side of the business. We have the international teams, which are still regional but that’s going to start coming our way too. So, we have a lot to look forward to. We needed a scalable solution that was one of the things that we kept in mind when we went with HighRadius. Our landscape when we first kicked off with High Radius, we had 10 business units across U.S. and Canada, some of them are the engagement platforms, some are of the settlement and some are of the trade that’s kind of gives you an idea about what I mean by business units and we had four full-time employees working on just cash application across those business units, process about 150000 payments per year. Many of those are very large payments because as you can imagine the number of gift cards especially during the holiday season that is bought at a Safeway or a Kroger is enormous. So that we’re talking like 5, 7 or 9 million dollar invoices on those. And 75 percent of our payments come in ACH and about 20 to 25 percent our check.

Cindy Scott:

We have a few wires that come in as well. One of our biggest challenges when we first embarked on this was we have payments that come in both to prepay for gift card orders, so there’s not really an invoice they’re saying, they’re just funding an account they’re going to then draw down against as they activate orders and then we also have the traditional invoice matching on our cash app. We wanted to be able to solve for both situations as best we could and that was one of my initial concerns, how can a system handle that when you’re getting cash and you don’t even know, there’s no order related to it, there’s no invoice related to it, how is a cash app system really going to help us in that situation and I found out how.

Cindy Scott:

Another thing that was on that side we don’t get clearly a lot of remittances for those pre-fund type of payments but we also have an interesting scenario in our settlement side of the business which is the largest portion of our business that several of our large customers, like Amazon, pays off of their data. And so, where we get this large payment with zero references. They don’t even reference a week number we just have to try to figure out where the payment applies and there’s a really in-depth reconciliation process that we go through where they have to transmit us their card level detail so that we can reconcile it to our systems’ card level detail and determine where there might be variances because things like commissions, they might have something set up as a 7 percent commission and it’s seven and a half in our system or they might have certain gift card UPC that is not set up in their system so they’re not paying us for them but they were activated because the activation system is separate, all kinds of fun reconciling scenarios and because those aren’t probably ever going to be applied at the invoice level. I shouldn’t say that we have ideas but it creates a challenge just to get the cash onto the account. I think for a long time two things were happening, one prior to my arrival at BlackHawk FIFO was like how they did cash application and it creates as if you’re in the cash up world or the AR world to the nightmares that they create we are unwinding a lot of historical issues because of FIFO. So, we knew we needed to get away from that. And I think had just assumed that was how it was going to be.

Cindy Scott:

And so nobody was thinking about you know looking at what the customer was providing in their remittances or whatever information they were providing that we could use to apply cash. So it was just like- oh it’s no invoice, no we’re gonna FIFO it. And thankfully we’ve gotten away with that and HighRadius has helped us do so. So I’m gonna walk you through just a couple of scenarios to see the kind of manuals that we were dealing with and what we were trying to solve for.

Cindy Scott:

And then I want to spend a little extra time that’s probably not on the slides just giving you some examples of our situations, so you can kind of see how it works in real life. First, I don’t have any slides to support it. So yeah you just have to listen to me and hopefully I’m clear. So this is in the prepayment scenario- as the customers go onto a portal they order their gift cards, book order, or whatever it is they send in the payment either through ACH or check. The invoices or payments would come in and then the individuals releasing the orders or fulfilling the request would have to go out and search, which you know they go onto the platform which is separate from our ERP and they would look at which order does this belong to. And one of the platforms we have, there are five right now that we’re dealing with. One does it, you can’t sort it. So you’re just scrolling through this page after page and in the holidays there were 56 pages of orders trying to find this amount and a name that kind of looks like that came out on that little blurb you get with the ACH remittance and it was manual and very time-consuming.

Cindy Scott:

Keep in mind releasing these orders quickly is paramount to the success both for us and for the customers using this especially on holiday because they’re giving these away as gifts and rewards. So anything that delayed those order release created a very unhappy customer and a stressed-out staff on my side. Once we were able to match it against the order that we have to go into the order system and manually release it, we’d have to do a manual journal entry to record it in our system. A lot of the N-word in this process. We have this working for one of our business units and we’re working on it. We’re kind of doing it by platform because every scenario the platforms are different because of the orders scenario and that fulfillment scenario is a little bit different but for the one that we have it working on the customers for the most part have been educated to include an order number on their remittance or at least their customer number. So if they’re just, for instance, pre-funding and they don’t have an order to offset they at least can put their customer number on the order.

Cindy Scott:

As it comes into ACH or through check, HighRadius can refer to that number and apply it to their account. So if it’s a pre-fund then it’s on their account to refund we have some processes in the background that reclassify those into a liability at the end of each period but for this process, it’s just getting it onto the account so that the customers’ orders can then draw down against that balance. One of the things that we learned as we implemented was we didn’t have to have the customer number if there was some kind of a customer reference the system trains itself or we can help it by either writing rules or through repetition, it knows that this St. Vincent de Paul order always comes in with this client reference number on the ACH , bank reference number or the name or we can somehow map it within HighRadius to then automatically apply to the customer’s account, no-touch which is a big deal.

Cindy Scott:

Then the funds are available for the drop-down we can run a report at HighRadius to then be able to fulfill those orders on the order site. Long term, we would like to be able to interface to the order site so that’s all automatic but that’s going to require some additional I.T. assistance. It just talks about the intelligence remittance capture that I just reference, so the payment comes in, the remittance information comes in and the system basically searches for something to make it match and then it’ll take anything that matches be it an order number invoice number that kind of thing that’s automatically matched. It flows through to post to the ERP if it’s not- we work through it as an exception. In the example I just gave as well we have customers who they might be paying for three orders and then they want the rest of their million-dollar payment to go towards funding the future orders.

Cindy Scott:

We’ve been able to write rules to say pay invoices first and then the rest goes to just on account as a prepayment. We’ve also been able to have the system if we don’t want their prepayments on that specific customer account we want it moved over to refund account for that customer. We can write those rules so that’s all automatic as well. This is just a flip through that I want to talk about the fun stuff. Now we’re going to talk about the payments that come in that reference something other than an invoice.

Cindy Scott:

I’ve heard about this in a couple of other sessions but you probably all have customers who pay by sales order or P.O. or load number or a million different references other than your invoice number or they embed your invoice number in some strange number that they have in their system or part of your invoice number.

Cindy Scott:

This was very manual we had typically the cash app individual would have to just posted on an account because he couldn’t spend the time searching for which invoices this cash was to apply to. Then my analysts had their analysts, we’re having to go out and search for the P.O. number that was referenced or see if they had known cheat sheets of the algorithms used to create the invoice number that would match to our system and a lot of time was spent just applying cash, instead of resolving issues. We’ve been able with kind of really simply creating rules in HighRadius that will it knows these customers are paying by P.O. number it knows what that looks like and how to how when we transmit our open AR file to HighRadius it contains all these reference field that has P.O. number. It has failed order number it has whatever other information that could potentially be used as a reference number by the customer that flows into the system and HighRadius is able to match that to payment no touched by the cash app team freeing up a lot of time on my analysts to resolve issues and not just be applying cash on their invoices.

Cindy Scott:

This has been a big win especially when I referenced the settlement side where once we really got to have the time to dive in and see what the customers were provided with their payments, we did see that a lot of them there was our invoice reference number in their remittance it was just embedded in some other string of numbers and we were able to carve those out and apply a lot of that settlement cash at an invoice level as well which was pretty much unheard for a lot of our larger customers.

Cindy Scott:

I think we will continue to have those customers that, if they’re paying off their data, there isn’t that unique reference number. But I think we can also get to some rules and analysis that will get us closer to being able to less reckon time more cash application time by the cash app folks. One of the other big wins we the OCR had, So if a check comes in the high radius OCR is reading the remittance because it’s kind of a copy of a copy so it first went to the bank and then it came into HighRadius. Sometimes the quality is not great unless we can convince either the customer or the bank to send through a higher quality image. We’re kind of stuck with it.

Cindy Scott:

Initially it was our invoices start BHINV and it was reading things like 8HINV. And so it wouldn’t be a match or B1NV and it wouldn’t be a match. We’ve been able to identify those exceptions and train the system for this customer when it looks like a ‘1’ it really means it’s an ‘I’ and it’ll auto-apply it saved a ton of time and I mean every week we’re learning new rules that we can write and get the cash to apply, successfully. The other thing on that slide that I just presented the manual journal entries has gone away. That all is automated as well through HighRadius. So it’s truly no-touch.

Cindy Scott:

Well I’ll get to the results in a minute. So just kind of recapping on. You know if you have a slow inaccurate cash application it creates all right. First of all, it creates potential credit limit issues, folks your cash in applied it looks like they’re hitting their credit because Kathy’s cash is aging at a different rate than the invoices, you for the order release scenario we’re not getting orders released until people are getting their cards on time. They’re really unhappy when they don’t get those cards on time. And then we’ve got a lot of wasted resources on our team trying to find where to apply the cash versus working on resolution or other issues even strategic initiatives and that type of thing. So that Intelligent Automation as I touched on, is it understands the customer behavior. I can’t sell this enough, I was a skeptic, I’m old I don’t understand the technology that much. I had to kind of grow up with it. It truly is the system learns as you go and what the system doesn’t learn. It makes it very easy for you to analytically either through reports or through data review determine what can be addressed. You see the patterns they think stick out like sore thumbs and you’re able to resolve those issues at a root cause level instead of just Band-Aid you know time after time.

Cindy Scott:

The result is a faster cash application, more accurate cash application, less frustration on the customer, less frustration on my team, the ability to allocate our resources differently and leverage them and cross-train them on other areas so that we have backup and support for things like vacations. She now goes regularly. If we don’t get the this is just speaking to what I already said if we don’t get the cash applied there is a negative effect not only on downstream but also upstream processes because you can’t, if you are difficult to do business with the customers do less business with you, right. So if we’re forever delivering Kroger’s charity organizations cards late they’re gonna find someone else to do that business for them. So it’s been key in ensuring ongoing sales growth and customer retention as well as just making sure that we have the right solution in place. Some of the things that we are also just starting to leverage are our kind of auto correspondence. If we have a payment that comes in that doesn’t have a remittance, the system can send a request for that remittance and it’s not in AR analyst or their cash app individual trying to reach out and ask for that remittance.

You have to have some master data set up for email addresses and that type of thing, but if you have that you can let the system do that kind of tedious task for you. And I already talked about the predefined strategies. One of the things I talked about the OCR, we also have customers that will pay multiple components of their business with us on a single check. So there’s going to be various customer numbers involved and we’ve written rules so the system knows which invoices go to which customer number.

Before we would get you know this million-dollar check from Kroger and fifty thousand would go here and twenty thousand there and eighty thousand there and maybe another thousand we didn’t know what to do with it based on and we would do these manual what we called C moves or cash moves which were basically journal-ed entries to move them from this customer to this customer on our ERP. So then the cash could be applied. The system can automate that based on that invoice numbering it knows invoices that are you know 10 digits belong to the settlement team.

Cindy Scott:

Invoices that belong to the trade customer account and so on and that allocation happens automatically. Most therefore then most of the cash app on that account happens automatically. We used to spend probably half a day posting Safeway payments and now we don’t touch Safeway payments. So it’s been a really big win especially on our larger customers that have multiple lines of business with us. Just to recap. Automatically aggregating remittance information so pulling it in from your emails from customer portals from off the bank stubs doing that without having you know the remittance laying on one side of the desk and or the email up and highlighting on the bank on the invoice or checking spreadsheets that kind of thing is a thing of the past. Having that remittance pulled into the system and automatically matched to the payments is a big win, number one.

And then having that single source of information to find the invoice number so again like I said we had our opening our file go out with as many reference numbers as we can drum up and then the system being able to identify the matching reference number based on the remittance has saved a lot of time as well. The automatic matching when it’s even when it’s just one for the one you got an invoice number and you have a payment number that was still a manual process and now it is not. . So that kind of thing. These are big wins for us. So what we win.

Since we implemented HighRadius we had a hit rate before because we weren’t automated. So I guess you could say we had a zero percent hit rate and we kicked it off with about a 30 percent hit rate initially just because we were still working through a lot of those rules and the customer mapping and you have alternate payers where you know that’s the situation where a customer sends in one check for multiple, child customers. We were able then within probably six months to get it. We got a 25 percent increase in that hit rate. The time that we’ve saved on just the cash application team itself, we have two full time employees now, earlier we had four full-time employees on those 10 business units. We’re now down to two and both of those are also doing other tasks as well so they’re able to support one of them as backup support for the deductions team. The other one is backup support for the order release team and it’s only on high volume days that they’re really truly full-time cash app individuals and then, we improved productivity is around the AR analysts not having to spend their time tediously we pay our analysts pretty good so them checking boxes to apply cash in that system was not a good use of our resources, so freeing up that time has enabled us to actually grow the amount of work that my team does without growing headcount at all and we actually still have some bandwidth and most of that is has to do with the automated cash application.

Cindy Scott:

Just real quick because I think we’re running up on time but I want to leave open for some questions, some lessons learned, don’t assume, ask a lot of questions don’t do like I did I just don’t think HighRadius could work for the weird scenarios we have so many issues and so many customer problems it make him work and even if it’s not 100 percent we’ve gotten we’ve got a huge gain from it. Don’t implement in silos and that means both from a process standpoint so we implemented cash app and deductions separately and we’re kind of learning, we should have been a little more cognizant of the impact one to the other after because we did deductions first we implemented cash apps without really testing the impact to the deductions and there was an impact that we had to kind of scurry through and also from a process standpoint we had customers that we needed to communicate with to get them to send in better remittances or we talked to them through their image qualities things like that. Like I mentioned doing rigorous testing just tests like crazy we paralleled for I think two weeks just to get everyone comfortable because if you mess up the cash you’re going to mess up a whole lot of lives, so that worked out well for us.

Cindy Scott:

I think everything else I’ve covered so I’ll open it up now for any questions that you all have.

Cindy Scott:

I’m not sure I understood your question.

Cindy Scott:

Okay yeah.

Cindy Scott:

Yeah for some customers. So for the most part faithful stopped probably early 2017 and it probably took us most of the 2017 to unravel and fix some of those nightmares and it honestly ended up in you know some write-offs and in refunds to customers that maybe weren’t theirs but it was just such a mess it was like we needed to start with a clean slate. We do have some lingering on some of our larger customers Amazon that we’re still working through but they’re partnering with us on that because it’s part of they had skin in that game as well.

Cindy Scott:

Anyone else? You guys are easy and you’re still awake. So thank you very much. All right. Everyone have a good day.