The human impact of data transformation

About this episode

Todd Schnirel and Jennifer McCalicher from Airline Hydraulics discuss how they use data in their organization and the challenges they've overcome. They cover topics such as data sources, data formats, ERP systems, automation, and the human impact of data transformation.

Key takeaways:

• How to say 'yes' to the data. Data comes from various sources including suppliers, customers, and sales teams, and arrives in different formats such as text files, raw data, and JSON files and Airline wants to work with suppliers where they are.
• The use of data integration tools like Clover DX helps in managing and transforming data for e-commerce and inventory management.
• The human impact of data transformation is significant. It frees employees from manual data entry and allows them to explore new roles and opportunities within the organization.

AI-generated transcription

Matthew Stibbe (00:01.294)
Hello, welcome to Behind the Data with CloverDX. I'm your host, Matthew Stibbe, and today I'm with Todd Schnirel and Jennifer McCalicher from Airline Hydraulics, and we're going to be talking about how they use data in their organization and some of the challenges that they've overcome. Todd, Jennifer, welcome. Thank you so much for being on the show.

Todd Schnirel (00:23.224)
Thanks for having us.

Matthew Stibbe (00:24.846)
So let's start by learning a little bit about Airline Hydraulics. Tell me what you do.

Todd Schnirel (00:31.256)
We're a value-add distributor. We've been in business for 75 years. This is actually our 75th anniversary. So we've been around for a little while and seen a lot of changes in the industry, but our primary functions are distributing high-quality products to factory and mobile automation customers and then doing value-added work for those customers to put those products together in a meaningful way to help the customers enhance their systems or get their products to market faster.

Matthew Stibbe (01:06.542)
Amazing. And it's not just for airlines anymore, is it?


Todd Schnirel (01:09.24)
No, no. Actually, airline is air in a tube, which we started because our founder was selling air tools to the manufacturing facilities in Philadelphia after World War Two and selling, you know, literal air in a tube. Thus, airline. So that's the airline, not the airplane. So we get that a lot. But it is the airline that goes into a factory, not the airline that goes into the air.

Matthew Stibbe (01:41.934)
Thank you for clearing that up. That's really useful to know. OK, so Jennifer, I'm thinking here about the business. You've seen the film The Matrix, and at some point the real physical world turns into ones and zeros and into data. Tell me a little bit about your world of data. Where does data come from? Where does it go to? How does it flow through Airline?

Jennifer McCalicher (01:42.734)
Yeah, so we get data directly from our suppliers, which at this point, I think we're over 250, 280 plus suppliers. We also get data directly from our customers, and we get it in various methods. Some of it even comes from our sales team as well. We get it through email. Some of it is we actually have to go to the FTP sites to get the data. Some of it is transferred to us, like an API that we may have.

And the data comes in all different types of forms. Some of it's text files, some of it's just raw data. Some of it is, you know, maybe JSON files. So it comes to us in all different shapes, sizes. Some of it, you know, some of our suppliers are very small. They're still like the little mom-and-pop shops down the street that only have a couple hundred SKUs. And we have some suppliers that have close to a million SKUs. So I think our database in total right now is over three million SKUs that we manage in our database.

Matthew Stibbe (03:01.358)
Three million SKUs, and the video on your website shows an amazing warehouse with 12 million dollars worth of stock. So I'm assuming a lot of this is product information, inventory information and that sort of thing from different people's... managing supply chains in and out of the business, right?

Jennifer McCalicher (03:02.334)
Mm-hmm.

Jennifer McCalicher (03:16.51)
Yeah, so most of the data that my team manages every single day is more on making sure the item numbers are correct, making sure descriptions are correct, along with pricing, unit measures, and sizing. We get into weights and things like that because we do have, as you saw, a very extensive website where our customers can come on and place an order. And we try to get as much information as we have up front so that they know what kind of product they're getting and making sure it's the right product.

So yes, that video encompasses a lot of the SKUs that we manage and how big and small they are. Some of them are the size of a pen tip and other ones take half of our warehouse, I think.

Matthew Stibbe (03:56.942)
And Todd, what sort of systems inside the business are you operating? What sort of applications do you use to run all of that?

Todd Schnirel (04:04.92)
So we have an ERP system, an Epicor Profit21 ERP system that we've had for probably a little over a decade now. And it's not the cutting edge of technology, but steady and it does a good job for us. And one of the challenges that my team and Jen's team by extension has is how do we use that steady, traditional distributor ERP system and make it work with modern web tools and modern API interfaces. And we use a product called CloverDX to do a lot of that work. And it gives us the bridge to kind of get between a legacy ERP system and many of the modern applications that we have to deal with.

Matthew Stibbe (04:56.974)
Well, obviously this is Behind the Data with CloverDX, so thank you for mentioning them. And I like that bridge metaphor. It's a useful way of thinking about it. They often talk about pipelines. So Todd, your job title is a director of technology. And very impressively, you've been there 18 years. So that's a good innings. I'm interested to understand or explore perhaps, a challenge that you've had to face with data in that time, perhaps a little bit more recently than 18 years ago, and kind of what the problem was and how you overcame it.

Todd Schnirel (05:37.144)
Yeah. So, the first one that comes to mind is, going back to kind of the 2016 timeframe where we decided that we needed to modernize our website and go from, kind of the traditional, digital brochure - here's who we are, here's what we're about, to an actual, e-commerce website. And, as Jen said, we represent over 3 million different products.

So, here we are kind of a smaller, at the time I think we were just over 200 employees, so a smaller company with limited resources trying to make an e-commerce site for you know three million products is quite a daunting task and those three million products come from over two hundred and seventy-five two hundred and eighty different vendors and all kinds of formats and so trying to communicate that onto a website so that a customer at either three in the afternoon or three in the morning can make an informed buying decision. We were looking at just mountains of dissimilar data and saying, okay, how do we take this and turn it into something that now becomes useful to us, ourselves and our customers? And so that is about the time that we found Clover and realized that it may be a tool that would allow us to meet our vendors and our customers where they were in their data journey.

Some of them are very sophisticated and they have restful APIs and they can do everything as JSON objects and they can array their data and all kinds of great stuff that makes it very easy and very fast for us to load. Others is ... mom and pops, and they want to email us a CSV file. And we want to sell their products as much as we want to sell the Fortune 100 company's products. So we had to come up with a way to, like I said, bridge that gap between an email CSV file and a RESTful API. When we were looking at options to do that and we found Clover, it became very apparent very quickly that this may be a tool that allows us to meet our partners where they were on their journey and still use and normalize their data so that we could use it on our website.

So, we're not done. It's a never-ending task to continue to update and optimize products for our website, but it's certainly significantly easier than it used to be when we were hand scrubbing Excel files and trying to upload them to databases. So I cringe to think of the way we used to do things compared to the way we do things now, and that we just would not be able to scale the way we do it now.

Matthew Stibbe (08:18.734)
Presumably the website has to have that data in a consistent structured data format to some sort of backend database to drive the e-commerce piece. So that's the destination for the, but you've just got this very heterogeneous wide source range of sources. That must be, what's the most challenging source of data that you have to deal with when you're getting that in from vendors, from suppliers?

Todd Schnirel (08:52.812)
I don't know about the most challenging, but the two most critical are the price and the availability or lead time, right? Because if you're trying to make a buying decision, those are the things that you really care about, obviously. And they're the things that change very dynamically or potentially could change very dynamically, right? The technical specifications, it's a quarter inch or, you know, it has 20 SCFM of flow, or it can do 3000 psi of pressure. Once we have those established, they don't really change. The product is what it is. But on any given day, we may have some in stock. Our vendor may have some in stock. There may be some in transit that we want to represent, giving our customer, especially over the last few years with all the supply chain issues, giving our customer accurate information about the price and the lead time so that they can plan their production accordingly is critical to us.

So getting that information, some of our vendors update it now on an every hour basis to us. They say, here's the information that we have. So we have direct connections through Clover to our manufacturers' ERP systems from our website that allow us to reflect their inventory in near real-time.

So that what used to be a phone call through our customer service representative to say, hey, I really need this pump. Can you call the factory and see how many they have and then call me back or send me an email or give me a quote, now turns into the customer searching that potentially in Google, landing on our website and seeing in real time a price and a quantity that can be delivered and being able to make a buying decision right there, saving not only their time, but saving us time and not having to do all the steps to help the customer decide if that product was going to make their timeline or not.

Matthew Stibbe (10:52.526)
Jennifer, as product data team manager, you've got to be cashing the checks that Todd is writing to all of these vendors. Yes, we can connect to it. Tell me a little bit about the challenges that you face making that a reality.

Jennifer McCalicher (10:52.862)
The biggest challenge is just like we brought before is all the different formats we can get them in. So we have to start with whatever they have. Sometimes it could be just a PDF. They're very challenging, like again, working with CloverDX to create PDF readers has eased that a little bit where we can take a 30-page catalog and pull out the data that we need and get it in our ERP system, where before we would be paying someone to sit there and actually read the PDF document and hand type everything in so that we can make sure we had accurate pricing to sell some of the smaller suppliers items that we would have.

Other than that, it is just making sure that we have the data. It's not always labelled the way it's supposed to be and knowing in this column, you know, this might be the lead time and this column might be the pricing and, you know, making sure we have accurate descriptions and those descriptions fit where we need them to fit. Cause some, you know, ERP systems like to limit the size of our fields and how much data we can show. So again, we, we've developed different areas to store the data in so that we can show more data on a website and give the customers a, you know, a paragraph of a description versus just a little 40 character like, here you go. This might be a valve. It might fit, you know.

Matthew Stibbe (12:20.238)
Yes.

Matthew Stibbe (12:24.174)
Can you give me an example of a specific problem that you've had to tackle, perhaps one of the more challenging integrations?

Jennifer McCalicher (12:33.022)
Yeah, so most of it has to do with time. Our one supplier is a very large supplier and they have almost a million parts. And when we would get a price file from them, they would give us all the data we needed. You know, they would have the description and item numbers and everything we need to calculate our pricing. But it was a close to a million dollar Excel sheet that we would get from them and to load it in our ERP system would take three to four days that we would have to batch it in in different amounts. Otherwise we didn't have enough memory. We'd be kicking users off the system sometimes because of the memory we'd be doing or it would freeze up.

Matthew Stibbe
So just to parse in the spreadsheet?

Jennifer McCalicher (13:09.758)
Well we had to take the spreadsheet, put it as a text file. Cause again, we're limited on the data we could put into our ERP system. It has to be a certain format. So we would have to take that 1 million line text file, like Excel file and make it into various text files that would process in about a 12 to 16 hour timeframe. So we could just keep running them around the clock and getting all the, all the items updated. And like Todd said, we stumbled upon Clover at one point and brought that in to solve some solutions.

That was probably the biggest solution because now we can manage the million records. I believe it takes us about three hours to process through the million records now. And it's a single file that we utilize, like a single graph within Clover that we utilize. We don't have to develop 25, 30 different sub processes just to get the information in.

Matthew Stibbe (13:57.358)
Building these sorts of data pipelines, it sounds to me like the big benefit for you is time saving, right?

Jennifer McCalicher (14:04.286)
It's a huge amount. We had a lot of staff who used to do data entry. I mean, that used to be a big job. Everyone was a data entry clerk. You don't hear so much about them anymore. There's a reason for that. But that's what they would do is they would sit there and just enter data all day long. And there are some times we still have to do it. But now we're touching it for things that need that human interaction or that thought behind it, not just take this and put this here, which no one wants to sit there and spend their days just copy and pasting all day.

Matthew Stibbe (14:35.79)
Yeah, indeed, indeed. My grandparents worked in a bank and writing out ledgers by hand. And I would have thought this one thing worse than typing data in, it's writing data in. So over the years that you've been solving these problems, what is the biggest learning that you've had, both of you?

Jennifer McCalicher (15:01.47)
For me, it's going to definitely just be just to try it. Like, you know, I think we've come to the conclusion, you know, Todd kind of brought up earlier is we can say yes, because we know there's a solution. It's just a matter of figuring out what's the best solution path to go down. You know, a lot of times before you would just say, no, we can't do that or no, we can't get that data or, you know, there's no way we can manage that. It's just too large. And now it's like, no, let's, let's do this. We can do it.

So, I think the biggest thing is just always, always give it a good try and just see if you come out successful on the other end because most of the time after some trial and error you will.

Todd Schnirel (15:40.76)
I think for me, this is going to be or sound like an odd answer for a, you know, conversation around data. But for me, the thing that I've learned is how amazing our people are. We are an employee-owned company. And Jen talked about all those people that used to be data entry. Right. And when we first brought Clover in, we first started automating these processes. A few of those people came to myself and our CEO and said, you know, okay, I know you're going to, you know, automate my job. I'm not going to be able to do this anymore. Could you give me six months? Could you let me know so I can start looking for another job? And I can honestly say the majority of the data entry people are still with Airline and they're in different roles. And it's been amazing to see what they can do for us when they get to use their mind as opposed to just their fingers on a keyboard. And they've been in different roles now. Some of them have gone on to do management. Some of them have gone on to do purchasing, our customer service. We've had some going to outside sales, all of these little talents and hidden abilities that our employee owners had that we just didn't realize because we had them head down at a computer typing away a needed job, but not a job that was adding value in the way that they're adding value now.

So for me, freeing up the time of the people by automating the data processes has led to exponential benefits throughout the rest of our organization because we free these people up and we've seen their passions and we've seen their abilities come through in all these amazing ways. And so it's interesting for me that the conversation around data has led to the blossoming of people.

Matthew Stibbe (17:11.822)
I love that. I love that story. And I love that you're an employee-owned company. You know, you could easily imagine a private equity company, we go yippee, we can fire a bunch of people. And there's a human cost to that. And it's, it's, it's, that's really encouraging.

Todd Schnirel (17:23.872)
Yeah. And for us with that history, it's allowed people to trust us. So they bring us other problems. They bring us other areas where they say, hey, I'm wasting time here. I know I'm doing this. I'm processing. We're working on a project currently to automate the processing of payments for invoices that's currently still being done manually.

And they've come to us and the accounting department has come to us and said, hey, we think this is an opportunity where we could use some of the tools you guys have developed to improve this process, improve the quality, reduce the speed. And we know we can then use our time to do more valuable things.

So it's really allowed that trust to develop from the organization where these ideas come from the grassroots and come up to us. So we don't have to go looking for them anymore. Yeah, they come to us, you know, almost more than we can handle on a given day. But that trust has been built, has allowed that to continue and then grow exponentially.

Matthew Stibbe (18:20.238)
We're approaching the end of our time together, but actually I'd like to dive into that upcoming or work in progress challenge of automating the invoice processing. What do you think the biggest data challenge is going to be there?

Jennifer McCalicher (18:27.81)
The biggest one we're dealing with actually now is it's very similar. The bank send us the information in all different formats. So we have to get it into, you know, a standardized format so that we can get it into our ERP system. So once again...

Matthew Stibbe
To do the invoice payment matching?

Todd Schnirel (18:54.2)
Mm -hmm.

Jennifer McCalicher (18:55.262)
...So once again, we're, we're utilizing Clover to go through the data and find the outliers and find the ones that match and, you know, update our ERP system so that the person who was sitting there and clicking boxes can go and do something more productive. And, and that, you know, brings value to the company, you know, even though this did before, like Todd said, it was a much needed position, but there's so much more that other people can do. So why have them sit there and do that? But yeah, it's the, it's the standard that nothing, nothing's the same, you know, from one business to the next.

Matthew Stibbe (19:24.878)
Was it Tolstoy who said all happy families are alike and all unhappy families are different? I don't know. But I'm going to give you a superpower. All right? This is the last question. You can travel back in time and give past Todd and past Jennifer one data-related tip that you've learned. So you're not allowed to give stock tips. It's data-related tips.

Matthew Stibbe (19:51.406)
What would you say to your past selves that you've learned that would encourage and help them?

Jennifer McCalicher (20:01.342)
I would say I'm just trying to think because it's kind of the same. Like it's just seeing how data has evolved. Like, you know, since I was hired to just literally work on spreadsheets, that was my first job with Airline was take the price file, get it in. And that was it. And to see where we are now versus then, like, all I can kind of just say is just like the stick with it, keep speaking up, you know.

The sky's the limit type of a thing. Technology is a great thing, especially in this field. Just keep trying and learn different things. Don't be afraid to learn. I guess that's probably my biggest thing is don't be afraid to learn. When something else comes in, don't be afraid of the automation part, because we all know a lot of people are afraid of that too. But it's just an exciting time that we're currently in and seeing how the company is growing and what we're doing and, you know, all the technology that we've brought in just, I think, what, the past five years, probably even. So, you know, we've been with Clover for seven, but it's just, it's definitely a technical world and it's kind of a fun place to be. So don't be afraid of it. Cause I think, I think everyone was afraid of data before, you know.

Matthew Stibbe (21:16.43)
I think that's a very encouraging sentiment. Todd, can you add to that?

Todd Schnirel (21:20.984)
I'm not sure I can beat that. So I'll just, I'll pile on a little bit and I would say the same is don't be afraid. And I think that's really apropos now as AI comes into the book, right? And there's so much data behind AI and there's so much buzz around AI and it's great, but it's also scary. And is the data real or is there AI hallucinations and where is this coming from? 

You know, if I'm giving my past self advice and kind of my current self is like Jen said, don't be afraid, like take baby steps and, and you can figure it out. We get to say yes to things. we have the tools to say yes to things. So we can take those small bites and we have tools to verify the data and we have tools to make sure that what we're seeing, whether it's from a traditional source or from an AI source is, is accurate and is correct and is of a high quality. So I think Jen's spot on, don't be afraid. Say yes and then see where that journey takes you.

Matthew Stibbe (22:27.438)
I love that. I think as we're bringing this conversation to a close, it's just been fascinating to learn about Airline and about your journey with, from spreadsheets and manual data entry to this wonderful world of automation. But actually the human story behind that and the consequences and I'm very happy that we've ended on a positive note. So Todd, Jennifer, thank you very much for joining me today and for sharing all of that but actually the human story behind it.

Todd Schnirel (22:57.784)
Thank you.

Matthew Stibbe (22:59.182)
And that brings this episode to a close. If you'd like to learn more and get more practical data insights about CloverDX, please visit cloverdx.com/behind the data. And thank you very much for listening. Goodbye.