Executive Briefings

Pet Products Maker Decides Manual TM System Was a Dog

A conversation with Roy Gallagher, director of business systems for Petmate Inc., a leading producer and marketer of pet products, based in Arlington, Texas.

From very small beginnings in 1961, Petmate Inc. has grown to be a market leader in the world of pet supplies, generating $300m in revenue and operating 1.5 million square feet of office, manufacturing, and distribution facilities. With brand names Petmate and Dogloo, the company is the nation's leading manufacturer and marketer of plastic kennels, pet carriers and structural foam doghouses, as well as a leading supplier of cat products, soft pet bedding, and feeding and watering systems. Gallagher recently headed the company's implementation of Oracle Transportation Management, which eliminated manual processes and increased efficiencies across Petmate's logistics operations.

Q: Petmate is quite a success story. Can you tell us a little about the company's history?

Gallagher: Until July of this year, Petmate was known as Doskocil Manufacturing Co. It was named after the company's founder, Ben Doskocil, who 45 years ago started building wooden dog kennels in his garage with about $500 worth of tools. These were kennels that the airlines would use for dogs that were being transported. Today, Petmate is a $300m company with 80 to 85 percent of the market share for kennels and shelters. Our products now are mostly made of injection-molded plastics. In the 1990s we started focusing on supplying the big-box stores like Wal-Mart, Petco and PetSmart. Everything we do now is pet related, from kennels and shelters to feeding and watering systems to different kinds of containment products. And we recently bought Aspen Pet Products, a leading supplier of leashes and collars. So changing the name of the company to Petmate Inc. was really to more accurately reflect what we do.

Q: Is your manufacturing mostly done in Arlington?

Gallagher: Yes, we manufacture about 75 percent to 80 percent of what we sell here in Arlington. We have 1.2 million square feet under roof here and we make simple, injection-molded plastic products-things like cat pans, water bowls, kennels and dog houses. A lot of people know us through our Dogloo brand, which are dog houses that look like igloos.

More complex products, like electronic feeding and watering systems, are made in China. When it comes to deciding whether to manufacture something at our plant here or overseas, we do a typical cost-benefit analysis. If we think we can get something made in China and fill the supply chain efficiently, that's what we do. If not, we make it here. The company we recently acquired, Aspen Pet, is based in Denver but they source everything overseas. About 95 percent of their product comes out of China.

We feel fortunate that we have been able to continue building a lot of our products here. One of the benefits of local production is the ability to actually build based on orders as they come in. We try to ship everything within five days of getting an order so there are a lot of times when we are building product on the second or third day and then shipping out on the fifth day. That requires a lot of flexibility in the schedule because you have to move orders around and line up the shipments to make sure there is a match between manufacturing's needs and your customer's needs, but it helps keep inventory down, which is an important driver for us.

We have a primary distribution center here in Arlington and a secondary center in Denver, but we also have several satellite facilities around the country to keep the distances down to our customers. This is particularly important with some of our products, such as bedding, which is expensive to ship because it doesn't weigh anything, but is bulky and large.

From here we ship mostly truckloads to our customers' DCs-we might ship 48 or 54 pallets of Dogloo shelters, for example. In Denver, they do a lot of cross-dock because they handle small products like collars and leashes, which are so small and so low value that you don't want to pick them, so we cross-dock them for shipment to individual stores.

Q: Before you implemented Oracle's transportation management system, was shipping a manual process?

Gallagher: Yes, it was completely manual. We had some electronic reports but a lot of manual intervention was required. Oftentimes, someone would actually have to go eyeball a shipment to see if he thought it would fit on a truck. As you can imagine, that was not always accurate. Sometimes we would get ready to load up a truck and would find that we either had more product than would fit on the truck or not enough to fill it and we would have to send the truck out with a lot of air. Neither of those are situations that you want to have.

So we saw a lot of opportunity in moving to an automated system. With the increases in the cost of fuel, our transportation expense had just been going up and up and we felt that a transportation management system definitely would help us reduce costs. A lot of times we pay for the freight, but even when the customer is paying, they want us to keep that number as low as possible. Since our products don't have a lot of value, transport really affects the total cost. A truck full of plastic dog houses obviously doesn't have the same value as a truck full of televisions, so the freight expense added on top really gets to be painful for the people who are buying our product. So we needed some kind of systematic approach to manage our orders and shipments so we would be less likely to have trucks leaving here that were less than full. Before, we just didn't have the visibility we needed. Now, if a less-than-truckload shipment comes up going to Kansas City, for example, we can see that if we bring this other order up a couple of days we can consolidate them into a full truckload, or we can combine it with other orders so that the truck can make a stop on the way to Kansas City to drop off a shipment. That saves money for everyone. That was the kind of capability we wanted.

Q: You already were using other solutions from Oracle?

Gallagher: Yes. From an IT perspective, integration is the biggest challenge, so we really focused on solutions vendors that were either tightly integrated with Oracle or very easy to integrate with Oracle. It was after we started this process that Oracle purchased G-Log (Global Logistics Technologies), which was one of the products that our distribution people were very interested in. So with that product in their portfolio, it made sense to go with Oracle both from a solutions standpoint and from an integration standpoint.

Q: Can you walk me through the process?

Gallagher: We receive orders from our customers through EDI into Oracle Order Management. Through an interface, these orders are pulled into Oracle Transportation Management and our planners look to see which ones are truckload and which are less-than-truckload. The truckloads are pretty straight forward and easy to manage. But the planners are looking for opportunities to reduce costs-locations and types of orders that can be combined, perhaps by moving one order a little forward or another a little further out. After making the planning decisions, they schedule shipping dates, push out that requirement and create a tender to a carrier, which can either accept or decline the offer. Tenders go out to our carriers based on how we usually ship to specific parts of the country. When a tender is accepted, we set our schedule for when those orders will be picked up and that is sent back into Oracle. This information then goes to our warehouse management system, which will make sure those orders are picked, packed and lined up on the shipping dock at the appropriate time.

One of the key elements for us is the load configuration that is part of Oracle TM. A lot of our products do not stack very well, such as the Dogloo, which is the dog house that is shaped like an igloo. We can stack about 10 of those on a pallet, but the pallet does not go floor to ceiling in the truck and we can't stack anything on top of the pallet. On the other hand, a lot of our kennels come two to a box and those boxes can be stacked all the way to the top. We have other products that are boxed and other doghouse products that stack differently. The system knows all of our box dimensions and which things can and can't be stacked, so it is able to calculate how many trucks any given shipment will need and how much product we can get in the truck. That's very useful for us.

We also are in the process of implementing the freight payment and billing module of Oracle TM. This will close the circle for us because it will ensure that the invoice we get from the carrier actually matches with what we had agreed to pay on the front end. That's something we are looking forward to because we currently are paying a third-party to do our freight audits.

Q: Have you quantified the savings enabled by the transportation system?

Gallagher: I can't give you exact figures, but we definitely have cut our transportation expense. We know we are sending out trucks to our customers that are fuller and better configured. A truck may make two or three stops, but we are not sending out two or three semi-empty trucks, so we are saving money there. Also, we have significantly reduced the workload in our traffic department. So far, we have freed up about 75 percent of one full-time position, which means this person can be working on other tasks. As we get better at using the technology, we think we will free up more time. We are getting more efficient at this every day.

In this whole project, the biggest thing that we didn't realize going in was the amount of new technology in the Oracle transportation project. I have worked with Oracle's database tools and in Oracle development environments in the past and we have a lot of people on staff who have been users of Oracle's e-business suite for some time, so we are very comfortable and familiar with Oracle. But BEPL (Business Execution Process Language), which is the new interface layer, is new technology to us. It is part of Oracle's SOA-based Fusion technology and is basically a little engine that sits between two separate systems and creates the interface, but you don't have to hard code it. It is very configurable and maintainable and essentially self-documenting because it is all built into the system. It is also very flexible because you don't have to reprogram when you need to change something. So we had a bit of a learning curve on that, but we see a lot of value in it and are really happy that Oracle chose this technology. Integration with other systems is very important so you can do things like call a particular piece of information that you need. And every system is a little different, but this lets you configure the interface to fit your business suite and your processes. It has really helped us a lot with the implementation.

From very small beginnings in 1961, Petmate Inc. has grown to be a market leader in the world of pet supplies, generating $300m in revenue and operating 1.5 million square feet of office, manufacturing, and distribution facilities. With brand names Petmate and Dogloo, the company is the nation's leading manufacturer and marketer of plastic kennels, pet carriers and structural foam doghouses, as well as a leading supplier of cat products, soft pet bedding, and feeding and watering systems. Gallagher recently headed the company's implementation of Oracle Transportation Management, which eliminated manual processes and increased efficiencies across Petmate's logistics operations.

Q: Petmate is quite a success story. Can you tell us a little about the company's history?

Gallagher: Until July of this year, Petmate was known as Doskocil Manufacturing Co. It was named after the company's founder, Ben Doskocil, who 45 years ago started building wooden dog kennels in his garage with about $500 worth of tools. These were kennels that the airlines would use for dogs that were being transported. Today, Petmate is a $300m company with 80 to 85 percent of the market share for kennels and shelters. Our products now are mostly made of injection-molded plastics. In the 1990s we started focusing on supplying the big-box stores like Wal-Mart, Petco and PetSmart. Everything we do now is pet related, from kennels and shelters to feeding and watering systems to different kinds of containment products. And we recently bought Aspen Pet Products, a leading supplier of leashes and collars. So changing the name of the company to Petmate Inc. was really to more accurately reflect what we do.

Q: Is your manufacturing mostly done in Arlington?

Gallagher: Yes, we manufacture about 75 percent to 80 percent of what we sell here in Arlington. We have 1.2 million square feet under roof here and we make simple, injection-molded plastic products-things like cat pans, water bowls, kennels and dog houses. A lot of people know us through our Dogloo brand, which are dog houses that look like igloos.

More complex products, like electronic feeding and watering systems, are made in China. When it comes to deciding whether to manufacture something at our plant here or overseas, we do a typical cost-benefit analysis. If we think we can get something made in China and fill the supply chain efficiently, that's what we do. If not, we make it here. The company we recently acquired, Aspen Pet, is based in Denver but they source everything overseas. About 95 percent of their product comes out of China.

We feel fortunate that we have been able to continue building a lot of our products here. One of the benefits of local production is the ability to actually build based on orders as they come in. We try to ship everything within five days of getting an order so there are a lot of times when we are building product on the second or third day and then shipping out on the fifth day. That requires a lot of flexibility in the schedule because you have to move orders around and line up the shipments to make sure there is a match between manufacturing's needs and your customer's needs, but it helps keep inventory down, which is an important driver for us.

We have a primary distribution center here in Arlington and a secondary center in Denver, but we also have several satellite facilities around the country to keep the distances down to our customers. This is particularly important with some of our products, such as bedding, which is expensive to ship because it doesn't weigh anything, but is bulky and large.

From here we ship mostly truckloads to our customers' DCs-we might ship 48 or 54 pallets of Dogloo shelters, for example. In Denver, they do a lot of cross-dock because they handle small products like collars and leashes, which are so small and so low value that you don't want to pick them, so we cross-dock them for shipment to individual stores.

Q: Before you implemented Oracle's transportation management system, was shipping a manual process?

Gallagher: Yes, it was completely manual. We had some electronic reports but a lot of manual intervention was required. Oftentimes, someone would actually have to go eyeball a shipment to see if he thought it would fit on a truck. As you can imagine, that was not always accurate. Sometimes we would get ready to load up a truck and would find that we either had more product than would fit on the truck or not enough to fill it and we would have to send the truck out with a lot of air. Neither of those are situations that you want to have.

So we saw a lot of opportunity in moving to an automated system. With the increases in the cost of fuel, our transportation expense had just been going up and up and we felt that a transportation management system definitely would help us reduce costs. A lot of times we pay for the freight, but even when the customer is paying, they want us to keep that number as low as possible. Since our products don't have a lot of value, transport really affects the total cost. A truck full of plastic dog houses obviously doesn't have the same value as a truck full of televisions, so the freight expense added on top really gets to be painful for the people who are buying our product. So we needed some kind of systematic approach to manage our orders and shipments so we would be less likely to have trucks leaving here that were less than full. Before, we just didn't have the visibility we needed. Now, if a less-than-truckload shipment comes up going to Kansas City, for example, we can see that if we bring this other order up a couple of days we can consolidate them into a full truckload, or we can combine it with other orders so that the truck can make a stop on the way to Kansas City to drop off a shipment. That saves money for everyone. That was the kind of capability we wanted.

Q: You already were using other solutions from Oracle?

Gallagher: Yes. From an IT perspective, integration is the biggest challenge, so we really focused on solutions vendors that were either tightly integrated with Oracle or very easy to integrate with Oracle. It was after we started this process that Oracle purchased G-Log (Global Logistics Technologies), which was one of the products that our distribution people were very interested in. So with that product in their portfolio, it made sense to go with Oracle both from a solutions standpoint and from an integration standpoint.

Q: Can you walk me through the process?

Gallagher: We receive orders from our customers through EDI into Oracle Order Management. Through an interface, these orders are pulled into Oracle Transportation Management and our planners look to see which ones are truckload and which are less-than-truckload. The truckloads are pretty straight forward and easy to manage. But the planners are looking for opportunities to reduce costs-locations and types of orders that can be combined, perhaps by moving one order a little forward or another a little further out. After making the planning decisions, they schedule shipping dates, push out that requirement and create a tender to a carrier, which can either accept or decline the offer. Tenders go out to our carriers based on how we usually ship to specific parts of the country. When a tender is accepted, we set our schedule for when those orders will be picked up and that is sent back into Oracle. This information then goes to our warehouse management system, which will make sure those orders are picked, packed and lined up on the shipping dock at the appropriate time.

One of the key elements for us is the load configuration that is part of Oracle TM. A lot of our products do not stack very well, such as the Dogloo, which is the dog house that is shaped like an igloo. We can stack about 10 of those on a pallet, but the pallet does not go floor to ceiling in the truck and we can't stack anything on top of the pallet. On the other hand, a lot of our kennels come two to a box and those boxes can be stacked all the way to the top. We have other products that are boxed and other doghouse products that stack differently. The system knows all of our box dimensions and which things can and can't be stacked, so it is able to calculate how many trucks any given shipment will need and how much product we can get in the truck. That's very useful for us.

We also are in the process of implementing the freight payment and billing module of Oracle TM. This will close the circle for us because it will ensure that the invoice we get from the carrier actually matches with what we had agreed to pay on the front end. That's something we are looking forward to because we currently are paying a third-party to do our freight audits.

Q: Have you quantified the savings enabled by the transportation system?

Gallagher: I can't give you exact figures, but we definitely have cut our transportation expense. We know we are sending out trucks to our customers that are fuller and better configured. A truck may make two or three stops, but we are not sending out two or three semi-empty trucks, so we are saving money there. Also, we have significantly reduced the workload in our traffic department. So far, we have freed up about 75 percent of one full-time position, which means this person can be working on other tasks. As we get better at using the technology, we think we will free up more time. We are getting more efficient at this every day.

In this whole project, the biggest thing that we didn't realize going in was the amount of new technology in the Oracle transportation project. I have worked with Oracle's database tools and in Oracle development environments in the past and we have a lot of people on staff who have been users of Oracle's e-business suite for some time, so we are very comfortable and familiar with Oracle. But BEPL (Business Execution Process Language), which is the new interface layer, is new technology to us. It is part of Oracle's SOA-based Fusion technology and is basically a little engine that sits between two separate systems and creates the interface, but you don't have to hard code it. It is very configurable and maintainable and essentially self-documenting because it is all built into the system. It is also very flexible because you don't have to reprogram when you need to change something. So we had a bit of a learning curve on that, but we see a lot of value in it and are really happy that Oracle chose this technology. Integration with other systems is very important so you can do things like call a particular piece of information that you need. And every system is a little different, but this lets you configure the interface to fit your business suite and your processes. It has really helped us a lot with the implementation.