Component inventory reduction

PAPER PRESENTED AT SAPICS 37TH ANNUAL CONFERENCE AND EXHIBITION 31 MAY – 2 JUNE 2015, SUN CITY, SOUTH AFRICA
© SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 1
Component Inventory Reduction in a Mid-Sized Beauty Company - A Case
Study
JAMES FUSCO CPIM
I began working at Perricone MD Cosmeceuticals, a prestige brand of anti-aging cosmetic products, in
late-2008. At the time, the company had no systems in place for inventory control, but over the next
seven years, the company implemented major changes and updates that helped reduce inventory
levels while still maintaining a high customer service level.
In 2008, the company was just at the beginning of a massive period of growth that saw our yearly
revenues double over a span of five years. There are many challenges that go along with that kind of
growth, especially in a company that had only been around for less than ten years at that point.
The main mantra of the company was, and still is: don’t backorder anything! Of course, that poses
many challenges for the Operations department, but that’s the case for many companies in the
cosmetics industry. It’s also the case for companies owned by private equity- the goal being to make
the most amount of money in the least amount of time.
When you’re dealing with 100% (or nearly that) customer service levels, you need to maintain a large
inventory (safety stock) to sustain that. And, with the company’s revenue doubling, we needed a
larger inventory in terms of quantity just to ship out the larger orders we were getting on a daily basis.
Over this time, the inventory at Perricone MD did go up- we’re not perfect- but, the levels didn’t go
up as much as expected because of the many changes we implemented while using APICS best-
practices.
During this time of transition in the company, I arrived on the scene. I had a unique perspective
because I had a fresh start in the business. I didn’t have any preconceived notions (I hadn’t even heard
of APICS at the time) and I was ready to learn from my mentors. I didn’t bring stubbornness to the
table because I didn’t have prior related experience to go off of. I was ready to learn whatever they
could teach me. I had the opportunity to experience a fresh ERP system install. When I started, the
company had no ERP system at all. Everyone conducted his or her business on proprietary
spreadsheets. When the decision was made to install and integrate the ERP system, I was able to see
the process from many perspectives- Operations, Accounting, and Sales.
I was able to see the entire process from a fresh perspective and with a fresh set of eyes. And when
the ERP system was installed, because of my propensity to work with computers and pick-up on new
things quickly, I was given the task to be the ERP administrator right off the bat. I maintained the item
master and the connections that were made with bill of materials and work orders. It was important
for me to see how all of those pieces came together in an ERP system.
The cosmetic industry is a competitive and fast-paced environment: newness rules! If you don’t have
the latest and greatest product on the market, you’re “yesterday’s news”. Now, that creates a little
problem when you have the word “planning” in your job title. As I mentioned earlier, our customers
were used to a 100% service level, which essentially became a make-to-order environment. Why?
Because, if you wanted it, we’d make it for you- no matter the timeline. We’d still use make to stock

TRANSFORMING BIG DATA INTO SUPPLY CHAIN ANALYTICS
principles, though, like having safety stock for all of our products. This caused our inventory to balloon
even more.
There were four main issues we faced in our operations department:
Issue number one was our lack of an ERP system. Previous component buyers placed blanket orders
for mass quantities of shared componentry. And that’s fine, but there’s a more efficient way of doing
that. The previous person would not take future forecasts into account. They only used past weighted
sales averages which are normally better for durable goods that have a longer shelf life and product
life. But for the cosmetics industry, I’ve seen products go from product development, to testing, to
launching, to market, and then “out to pasture” (discontinued) after only a year. And if you’re only
looking for past sales averages when ordering components, you’re not looking to future retailers’
plans. So, it really isn’t an effective way to plan for componentry. Previously, all planning was done on
proprietary spreadsheets. We didn’t have an MRP system at the time (and still don’t!). There was no
consistency to the data on these proprietary spreadsheets and there was no opportunity to use this
data between departments. For instance, finished goods planners had their own spreadsheets that
they had been working on for five years. They would tweak the information to show exactly what they
needed…and only what they needed. This information couldn’t be easily shared with the other
departments or even people within the operations department.
The cosmetics industry is completely marketing-driven. Since I joined the company, we’ve had three
brand revisions to our product line. The biggest change was when we switched over the look of all our
packaging, including containers, labels, and cartons. We originally had screen-printed packaging and
then went back to glass packaging, which the company was originally known for. And, we had an
expansion of component variety. Originally during the rebrand, we were able to whittle-down the
number of primary jars and bottles we used to just one colour and then would put labels on to
distinguish the individual products. After a while though, the marketing team decided to tweak the
product packaging and created different colours and sizes of jars and bottles. Once this happened,
the product line started to grow and the component inventory began to swell.
Issue number three was SKU proliferation. As I mentioned earlier, our yearly revenue doubled in less
than five years. And the growth on components outpaced the component lead-time. The products
were selling so well that we couldn’t keep up with the component lead-times. If the forecast came in
at 20,000 pieces per month, we’d end up over-selling at 30,000 pieces per month! And because the
glass lead-time was around four months at the time, we would always be a month behind all of our
jars. The product line was expanding rapidly as well. Over time, we ended up with five times the
amount of components part numbers due to the rapidly swelling number of products and variations
among those products. On a daily basis, I would seemingly order a new component that we had never
ordered before. When the company started in 1998, there were only four finished goods SKUs. There
were only four original products that Dr. Perricone came up with. When the ERP system was installed
in 2009 that number jumped-up 149 SKUs. And now, six years later, there are 752 finish good SKUs!
So you see the exponential growth of this product line and the company since the ERP system was
installed. And, there are no signs of this slowing down in the future.
When you take a look at the component SKUs, the same trend appears. When we installed the ERP
system, they were under 400 component part numbers. And now we’ve just surpassed 2,500
component part numbers. So, it’s that much more to plan for and that much more to keep inventory,
when you take safety stock into account.

That brings me to issue number four: high costs. When the company started, it was envisioned as a
prestige brand. That means there are very high-end products in the $200-$300 price range. And some
people would love to pay for that, because the products really work! But at those price points, we
really weren’t doing high-volume. We weren’t buying components in 100,000 or 200,000-piece
increments. So we don’t have the opportunity to go to the suppliers and negotiate pricing because
we were only order 10-20,000 jars/caps/bottles at a time. And because we had these very specific
items that our marketing team wanted, we’d have to order them from many different suppliers to
match marketing’s needs. Because we only ordered one or two products from each supplier, we didn’t
have the leverage to request better pricing. We didn’t do any benchmark testing on our component
pricing at the time because we didn’t have any other options to compare them to. When working
with a prestige brand that charges $300 for a product, the margins are really, really good. So a jar
costs a dollar (a high price for such a small item), no one’s going to make a big deal out of it. There
wasn’t pressure to reduce component costs because we simply didn’t have the volume to negotiate
better pricing. Plus, the margins were good and the company was making money- reducing
component costs wasn’t a priority for a “prestige” brand. And, if the company stayed in that “prestige”
lane, we probably wouldn’t have changed anything. Of course, the growth of the company and the
movement towards a more accessible (mass-marketed) brand lead to higher quantities and more
opportunities for savings, which are discussed later.
At the time, we also had two warehouses on the east coast. We had one in Connecticut and one in
Florida. We started with the one in Connecticut, but as the business grew we had to move to a larger
facility. Instead of getting rid of the Connecticut warehouse, we decided to run the Florida warehouse
in parallel. But the problem was, we would have products made and had to ship partials of our
production runs to each warehouse to fulfil different customers’ needs. This would force us to
constantly transfer products from one warehouse to another when one warehouse couldn’t support
the needs of the customer.
Now, for our solutions, which are still works in progress (kaizen!).
Number one, and probably the biggest change, was the implementation of formal forecasting. We
began looking to the future for production based on budgets. We asked the question, “How can you
hit budget if your forecast doesn’t support the product equalling that dollar amount?” So, if your
budget was $10 million and your forecast totalled products equally only for $9 million in sales, how
are you going to hit that budget? We know that the channel heads have a way of hitting their budgets
(and, during this period of growth, everyone hit their budgets), so we now asked them to dictate what
products they expected to sell in order to hit their budgets.
Second, we established a formal sales and operations planning meeting (S&OP). For this, the heads of
each channel and department would meet and discuss if the forecast aligned with the planned budget
numbers. If not, changes we need to be made to the forecast. Items that sold either much higher or
much lower than the forecast which trigger a review. So, if the product sold 20% lower over the last
three months than expected, management would ask if the overall future budget for that item should
be revised. And, if the channel had was still planning on hitting their overall budget, they would have
to explain what products they would plan on selling to make-up the difference. So, to prevent some
of the expediting we’d do, we would try to better forecast for the items that would help make-up the
budget shortfall. This meeting would also let us know about future product discontinuations, so that

we wouldn’t over-order components. Sometimes, from an operations perspective, a product with
seem to be selling well. But, based on conversations retail channel heads would have with the
retailers, they’d learn that a product’s popularity would be waning. During the sales and operations
planning meeting, this information would be disseminated to the operations department and allow
us to adjust our ordering patterns for items that might not be doing as well in the future as they had
been in the past.
We also instituted production plan meetings for the operations department. These were monthly
meetings where we would compare the forecast submitted by the sales team against our current
inventory of finished goods. We based it on the weeks of supply that we had on-hand and then would
plot the future production runs accordingly. One of the things we were tasked with was going from
holding 13 weeks of safety stock down to only eight weeks of safety stock. Now, with the reduction
of safety stock, it was more difficult to maintain the 100% customer service level that the company
was so used to having. Because, without extra safety stock on hand, there was a greater risk of going
out of stock on any given item. As you’ll see in the charts at the end, though, we were able to maintain
the high service level despite the reduction in safety stock due to careful planning and better
forecasting.
Then we installed the ERP system. This took the whole company online, which was a big deal for us.
Every department was able to talk to each other using the same language and the same set of data.
We saw some instant changes, like being able to have checks and balances. No longer would the
accounting department have to sort through mountains of paper to match up a purchase order with
an invoice from a vendor. The amount of time it took the company to process orders was drastically
reduced because all the information was contained in the system. We were also able to more easily
compare our inventory at our contract manufacturers’ locations because we had a solid record of
every in and out transaction in the system. This reduced the amount of inventory discrepancies.
Having a centralized item master and bills of materials helped me, as the component planner, to easily
combine orders for shared components. I was able to aggregate all of the demand into one place,
rather than doing it manually. We still don’t have an MRP system, but I am able to take information
from the ERP system and create my own “mini– MRP” in Excel. Before the ERP system, I had to go
through each of the production work orders and manually match up the shared components. It was a
manual process. Now, with the centralized information contained in the system, everyone in the
operations department was able to see the same data at the same time.
We also had new strategies for procuring componentry. One of those ways was ordering using “make-
and-hold”. We used “make-and-hold” agreements with our vendors for shared components. It
allowed us to order in higher volumes at one time, but not pay for all of the components at once. It
also helped us with lead times because we always had excess to draw-from at our vendors’ warehouse
locations. We were able to establish these “make-and-hold” agreements with the vendors because of
our long history with them. The vendors knew the strong reputation our company had, so they were
more willing to store extra components and allow us to pay for the items only we needed them. This
gave us an economy of scale that we never had before. When the business started to grow and we
started to need greater quantities for our production runs, we were able to order larger quantities.
Now, there are some issues when ordering with the “make-and-hold” process. For instance, the
cosmetics industry is always changing. So, when ordering things like cartons, we’d have to watch-out
for regulatory changes, marketing updates to the packaging, and ingredient changes. After

experimenting with the “make-and-hold” ordering process for cartons, we decided that there were
too many changes to allow us to order cartons in a large enough scale to make the making whole
process worthwhile. It’s a shame that we can’t take advantage of “make-and-hold” for cartons, as
ordering larger quantities of printed materials is a very advantageous. Pricing on printed materials
improves considerably as the quantity goes up. But because our marketing department still reserves
the right to change the artwork on our cartons at any time, it just wasn’t worth taking the chance on
having extra inventory. The “make-and-hold” process can also put a strain on vendor relations. If a
product’s popularity starts slowing down, we may not use-up the specific component that goes along
with that product as fast as we originally told the vendor. The vendor may be stuck with components
for a lot longer than they originally expected. So, while we are utilizing “make-and-hold” ordering, we
try to limit it to only our most popular items and most consistent sellers.
We also started utilizing outsourcing and ordering our components from China. We had these new
make and all policies that allowed us to order in greater quantities, so moving component production
overseas finally made sense. When ordering components overseas, the volumes must be greater then
what you’re able to order domestically because custom moulds for glass need to be made. Of course,
there were some start-up issues in moving our componentry to China. We had custom moulds made,
but because of the distance and language gap, some of marketing’s preferences on these components
weren’t translated properly into the finished product. This led to delays and some extra-added costs
for new moulds. And, because our products are such prestigious items, there were certain
expectations of how each component should look. Because of the natural variations in glass during
production, our marketing team wasn’t happy when the amber colour differed from jar to jar. This
slowed-down development of our componentry in China even further. Lead-times in America and
China for our glass componentry are very similar at about 4 months. But, in America, that lead-time
was all production lead-time. At the end of the four months, we would be able to ship to our fillers in
just a couple of days. In China, the production lead-time was only three months and that fourth month
was used for shipping the items by sea freight. So, the standard lead-time was the same. But, this
offered us a potential advantage in our efforts to meet a 100% customer service level. Because the
production lead-time in China was only three months, we had the option of airing-in our glass
componentry and potentially saving a month of lead-time.
We then used our ERP system to provide metrics. It was now much easier to see if we were over or
under-stocked on any item in real-time. I was able to come out of our production plan meeting with
the clear view of what I needed to order based on the numbers I ran from our system. I was able to
establish, because of this information, better run rates.
We also had improvements in vendor relations. I was able to convey our run-rate information with
our vendors so that they could potentially order raw materials early and be ready for us when we
eventually placed our normal monthly order. For instance, if I was able to give the vendors information
on how our latest QVC did, they would know whether or not they’d need to ramp-up production in a
couple of months. This would allow us to reserve time on the production line, and potentially save us
overall lead-time. I was also able to establish component open order reports that showed each of the
vendors what was on order and what had already shipped. This would reduce a lot of confusion, solve
problems with receiving paperwork, and let me know when all of the items would be coming in. The
vendors would go through these reports line by line and double-check that our current due dates
matched their idea of when the products would actually ship.

We were also able to reduce our component SKUs to a certain extent. For instance, we have three
different sizes of treatment and cleanser bottles. We ordered them in 4-ounce, 6-ounce, and 8-ounce
sizes. But, we were able to standardize the neck size of these bottles so that our pumps would be able
to fit on all three bottles. The only thing we’d need to change is the dip-tube length and then be able
to put the same pump on all three bottles. This allowed us a lot more flexibility with our pumps. If we
ran out of 6-ounce pumps, for instance, we’d be able to cut the dip-tubes on our 8-ounce pumps to
get us out of a jam.
We were also able to find some efficiencies with our different colours of bottles. Now, it’s possible to
get glass in many different colours including green, black, and blue. But, each one of those colours
requires its own special custom mould and a minimum order of 250,000 pieces. And because we only
had one or two of our products that used this special coloured glass, we would never be able to hit
those minimums. So, we were able to work with our vendor in China to create a clear mould for our
glass. The glass is produced clear and then we were able to spray paint the correct colours onto the
glass. This gave us very consistent colour and allowed us much smaller production minimums of about
only 10,000 pieces. We were also able to work with the vendor using our “make-and-hold”
agreements to keep a stock of blank clear glass so that we could paint them whatever colour we
wanted when the need arose. This gave us more flexibility in working with the whims of our marketing
team.
Other cost savings included benchmarking, which I mentioned earlier. Now that we had this
opportunity to create custom moulds, we had a greater number of companies willing to talk to us
because we now had the volume to support getting better pricing. Through this process, we actually
changed some of our component vendors. We were able to move to vendors that not only gave better
pricing, but offered greater consistency and better flexibility. We were also able to reduce our vendors
because we were able to take our various one-off components and duplicate them in China. Normally,
this can be an expensive proposition, but since we were doing all of our business with the factories in
China, we had leverage due to the overall buying power we now possessed.
We also made suggestions to use up excess inventory. This wasn’t being done previously. In the past
we would have excess finished goods for kits that didn’t sell well. And then someone would come
along and take one of the minis that was in the kit and use it for a sampling program. But then, we’d
be left with an incomplete kit. Those would just sit around and eventually expire. There was also no
visibility to these items in our inventory because there was no central document distributed to the
channel heads letting them know that these items existed. Now, we distribute an excess inventory list
to the channel heads and encourage them to come up with new bundles and offers using these excess
items. We have seen a marked decrease in the amount of items we’ve had to dispose of because of
this. It’s also helped sales because the channel heads would be able to create new value propositions
for our customers without overselling their regular forecasted items.
We also reduced power number of warehouses from 2 to 1. The decision was made to only go with
the larger warehouse in Florida because it could support the growing company’s needs. Unfortunately,
the location of the Florida warehouse is not really ideal because it’s tucked-away in one corner of the
nation. Plus, our main filler is located in the centre of the country. We’re forced to truck all of our
completed items down to the warehouse, which takes an additional three days and adds cost.
Hopefully this is something the company will look to change in the future.

Now, for the results of these efforts so far. The charts below show how the company has managed to
maintain a high customer service rate while slowing the growth of inventory during a time of massive
growth in company sales.
As you can see, our order fill rates, line item fill rates, and unit fill rates are all Close to 100% all the
time.

On the SKU proliferation side, we’ve seen the number of finish goods SKUs we sell increase by 67%
over the last 3 years.
And our components SKUs have increased by 61% over the past three years. The reason this number
isn’t even higher is because some of our components are shared over multiple products.

Our finished goods inventory has gone up over the past three years. But when you look at our
average finish good inventory level, it has only increased 20% over the past three years. Compared
to having 88% more finish good SKUs over the same time-period, an increase of only 20% means
we’re running a leaner operation.
For components, our average component inventory value increased by 37% over the past three
years. Compare this with the fact that we have 61% more component SKUs and we’re also running
leaner on the component side.

SPEAKER PROFILE
Jim has a Bachelor’s Degree in Communication from Southern Connecticut
State University. He began working at NV Perricone Cosmeceuticals (a
prestige brand of anti-aging skincare products) in late 2008 as an
International order coordinator. From there, he was promoted to ERP
Administrator, and then Component Planner & Buyer. Jim was then given the
responsibility of managing the company’s emerging Infomercial account, one
of the company’s biggest sales channels. Jim earned his CPIM designation in
2013 and is the Vice President of Membership for the Hartford, CT USA APICS
chapter. He and his wife are expecting their first child in August.
Contact details
Email address jimfusco@me.com
Website http://www.perriconemd.com/
Telephone 2035003879

Component inventory reduction

More Related Content

Similar to Component inventory reduction (20)

More from https://logisticscompanies.co.za (20)

Recently uploaded (20)

Component inventory reduction