Video

Three Important Questions on Supply Chain Network Design

Video Transcription

Dan:
In our top story, I am here with Toby Brzoznowski, the Executive Vice President of Llamasoft. I think, he’s going to say the leading company, in the field of Supply Chain Network Design. We’re here at their summer con, user conference, in New Orleans, Louisiana. Very pleased to have you with me here today, Toby. Thanks for joining me.

Toby:
Thanks for having me.

Dan:
I’ve got a few questions, very interesting I think, around Supply Chain Network Design I’d like to ask you. Starting off with, obviously there’s been so much with OMNI Channel, over the last few years, and retailers, and consumer venture companies trying to figure out where to put fulfillment DC’s, and “Should I fulfill from the store, and what stores?”. Beyond that, is what we know as the high level: What kind of insights have you gained into this process from working with retailers, especially maybe some consumer goods companies, over OMNI Channel Design, over the last few years?

Toby:
Sure.

Toby:
Okay, all right. Well, I think there’s a couple things. Number one, is really the idea of what stores should fulfill orders, and is it right for all of them to have that ability. I think that one of the things that we’ve identified, is there’s really a service level component to this, and a service coverage, that a lot of people don’t understand. When you have stores, you can use them, when you need to, but they’re certainly not equipped to deal with that very well, and so, there’s an additional cost, there’s an additional training that goes along with that.

What we found, is there’s a sweet spot, with many organizations, as to how many of those stores should be allowed to, or should be enabled to fulfill the demand from the eCommerce network. There’s an interesting para-dime, where there’s the executives that want to come down, and say “Let’s get this in the hands of everybody that’s going to help enable same day, or near same day delivery.”, but in the analysis that we’ve seen from our customers, it’s really, in many cases, less than half, sometimes it’s only ten to twenty of those stores, that really needed to be enabled, in order to still achieve that same service level target, that they have. That’s one of the main areas.

I think a second thing, is really around some of the pricing strategies that are either used by the different last mile delivery providers, that they have, primarily parcel. You can certainly partner with a FedEx, you can partner UPS, or the US Postal Service. If you really take a look at that last model, what we found is, that just the way in which many of these companies have engineered their pricing strategies, they’ve engineered the pricing, in many cases, to force demand into some of their desired channels. You’ll find a lot of counter intuitive pricing, that unless you can overlay all three of those different channels, for every different order, you might be surprised that, maybe there’s a mix that you don’t consider, that is by far the best option to get you your lowest cost of serve, in certain markets.

Dan:
Very good. Okay, excellent insight there. Changing gears a little bit, obviously you have some of the largest, most well-known companies in the world are users of LLamasoft. Is there a play for Supply Chain Network Design, in the mid-market?

Toby:
Yeah. That has been a question for us, for many years. I think, part of it was, we had always said, until you get to a billion dollars or more in revenue, you probably don’t have the capacity to have analytics, or an analysis team dedicated to this, even if your supply chain is sophisticated enough. I think what I’ve identified, recently, as a trend, generally is, two things have happened. Number one, with volatility and change in the marketplace, even if you don’t have an overwhelmingly complex supply chain, you have to adapt faster. The need for design is there.

Secondly, as the ease of use of this technology, and the accessibility through SASS based models, as opposed to the more hardcore modeling tools, the become accessible, they become easier to use. I think that you can get a good combination of some analysts on your team, that are doing this when needed, and you can still react to the changing market conditions.

Dan:
Very good. Today, even a billion dollar company isn’t that big anymore, by last year, so that one to five billion dollar, still the sweet spot, very much?

Toby:
One and above. Honestly, you get into the fortune fifty, the fortune one hundred. Pretty much everybody has to be doing Supply Chain Design all the time, because whether it’s by a region, or by product line, or business unit, there’s just so much volatility going on, there’s so many new products, new costumers, new suppliers that are coming on board all the time. You have to evolve.

Dan:
Okay, very good. Last question. Obviously, so much of this from a mid to long term perspective, is based on forecasting the business, forecasting demand, forecasting that kind of stuff. Almost by definition, forecasts are wrong. Actually, came up in a presentation I saw this morning from Starbucks. That’s just a fact of life, you have to have something to base your design on, but are there any practices, or techniques that companies can use to somewhat mitigate or deal with this issue of forecasting accuracy?

Toby:
Well, there are. I think the biggest thing is, what you have to do, is sensitivity around that demand, and figure out is there a tipping point where my strategy, or the policies I’m using to govern how I operate my supply chain, when does that tipping point happen? What we’ll see a lot of our clients do, is they’ll come up with a forecast that comes in from the forecasting, or demand planning systems. They’ll build that into their model, they’ll project out what the plan is, or what the policies are that they need to do, in an optimal situation, but then, you can do sensitivity around that demand, and say, “Well, what if it increases by five, ten, fifteen, twenty, fifty percent, in either direction?”.

We built a scenario planner layer, on top of that where you can quickly add in all those different increments, run all of those scenarios in the cloud, very fast, get the results back, and you can see very quickly, at what point does the strategy change. Many times, even if you’re off by fifty percent, in either direction, the overall strategy, or the overall supply chain policies that you’re using, are robust enough that it’s not going to make an effect. If that’s the case, you don’t need to be more accurate with your forecast. If you find that a five percent change, in either direction, is going to drastically change your policy, then you really need to monitor that a lot closer, and probably drill into another level of detail.

Dan:
Toby, great stuff, thanks for joining me today.

Toby:
No problem, thanks.

Ready to learn more? Click to request a live demo with the LLamasoft team.

Request A Demo