Ed Rennemann, CIO Crate and Barrel Jiri Nechleba, CEO 4R Systems January 17, 2006

1
Ed Rennemann, CIO Crate and BarrelJiri
Nechleba, CEO 4R SystemsJanuary 17, 2006
Managing In-Store Inventory Levels with
Analytics PROFIT OPTIMIZING INVENTORY TO BETTER
FULFILL CUSTOMER DEMAND
2
Agenda

• Retail Supply Chain efficiency
• Crate and Barrel Profile Challenge
• Building the case
• Pilot and Rollout
• Where are we going?

Context Supply ChainEvaluation Results
3
The Retail Investment CycleCash ? Product ? Cash
ROI Speed x Efficiency Turns x Earns
Context
4
Three major decisions drive profitability of the
cycle

• Matching buy with aggregate demand
• Buy too much ? excess inventory
• Buy too little ? lose sales
• Balance excess inventory costs versus lost sales
  costs
• Putting product in the right place
• Too little available at POS loses sales
• Too much raises inventory costs
• Long lead times mean managing buffer inventory
• Balance cost of inventory versus costs of lost
  sales
• Getting the most out of excess goods
• Too small a price reduction, goods unsold -gt low
  revenue recovery
• Too large a price reduction, low unit price -gt
  low revenue recovery
• What price generates the most revenue from a
  fixed inventory?
• Keeping markdown activity from becoming excessive

Context
5
Understanding uncertaintyis the path to more
profit

• If we knew the answers, it would be easy
• Wed buy exactly the right amount of product
• Wed put it exactly at the right place at the
  right time
• Wed have no excess inventory to liquidate

But uncertainty creates risk for retailers
Natural causes (randomness of events)Self
inflicted (we analyze data incorrectly)
Uncertainty drives inefficiencies that erode
profits Too much product in some stores, while
others stock out Some products sell out quickly,
while others dont sell Some seasons overbought
while others under
Context
6
Optimization links uncertainty to economic
outcomes

For most retailers, optimizing these decisions
can reduce out-of stocks, increase customer
service and improve profits by 2 or more of sales
Context
7
Seasonal vs. Replenishment GoodsDifferent
drivers to maximizing profits

• Seasonal Goods All three areas are important
• Margin performance driven by the size of buy
  versus full-priced demand
• Efficient, dynamic in-season allocation key to
  minimizing POS supply/demand mismatches
• Markdown and other tactics key in changing demand
  to meet supply
• Replenishment goods Key is proper in-store
  inventories
• Too much leads to unproductive inventory and low
  turn
• Too little boosts turns but at the expense of
  service and sales
• Right levels
• Maximize customer service
• But with an eye towards profit management

Context
8
Context
9
How was in-store inventory managed?

• Strong corporate systems provided managers and
  associates with relevant data and history on past
  sales
• Push vs. Pull
• Decisions made by managers responsible for the
  consequences and close to their decisions
• Crate and Barrel owned outlets used to sell off
  excess goods

Context
10
Why change?

• Hypothesis that inventory could be better managed
• Symptoms
• Judgment
• Culture
• Current model risks
• Inventory rebalancing generated work cost
• Desired better visibility on relationship between
  inventory decisions, customer service and profit
• Crate and Barrel has experienced, well-trained
  people. Sustaining profitable growth drove need
  to provide better information to staff

Context
11
Requirements for change

• Demonstrably improve profitability and customer
  service
• Consonant with company culture and values
• Rational risks, costs and rewards

Context
12
How were the requirementsfor change met?

• Looked at key decisions in the cash ? product ?
  cash cycle
• We worked with 4R to benchmark opportunity in
  each area
• Opportunities found in each area

Evaluation
13
Potential found in all areas with different
implementation challenges

• Initial buy optimization
• Store replenishment allocation optimization
• Markdown optimization

Crate and Barrel decided to proceed with a store
inventory pilot
Evaluation
14
Store demand is highly randomHow much inventory
should we have?
Evaluation
15
How much inventory should we have?Benchmark
shows relationship between lost sales, inventory
levelsand profitability
Lost Sales/Margin
For Illustration Not actual results
Weeks of Supply
Evaluation
16
Replenishment benchmark SKU level analyses
revealed suggested changesthat made sense to
Crate and Barrel
For Illustration Not actual results
Evaluation
17
Crate and Barrel started witha store
replenishment pilot

• Pilot started at end of October 2004 running for
  5 months
• We took 4Rs optimized inventory levels and
  merged theminto our existing process.
• Store personnel training

Results
18
How does it work?
19
Pilot was under a literal microscope

• Each store was given recommendations
  specificallygenerated for it and store
  management would reviewthe recommendations in
  detail at least once a week
• Beyond generating good results, the pilot needed
  to gain confidence from the organization

Results
20
When we introduce optimized store inventory
levels, what happens?

• What was the measurable business impact?
• What did store staff experience?
• What was the feedback from the stores?

Results
21
Lost margin dropped quickly and significantly
at start of pilot
For Illustration Not actual results
Results
22
while store inventory costsrose somewhat
For Illustration Not actual results
Results
23
resulting in a Lower Total Cost vs. Rest of Chain
For Illustration Not actual results
Results
24
Net Economic Impact

• Lost sales and associated margin dropped
  significantly
• Inventory was raised in pilot stores
• Combination of margin recapture and extra
  inventory costs leads to increased profits and
  better customer service

Results
25
Impact on in-store inventory store ops

• Generated higher in-stocks
• Store staff smarter on how to manage inventory
• Overlaying economic data drove inventories to
  capturemore revenue profitably
• More precisely setting inventory
• Reduced time for ordering and allowed store
  personnelto devote more time to other activities

Results
26
What did pilot store personnel see?

• Reduction in priority orders for replenishment
  SKUs.
• Backorders for replenishment product virtually
  eliminated.
• They saw significant increases in sales from
  deeper inventoryon some items and stocking other
  for the first time.
• Design teams spent less time remerchandising
  displaysand more time on new displays.
• Staff selling confidence increased where raised
  inventory ledto higher customer service levels.

Results
27
Positive pilot results ledto chain-wide rollout

• Improvements were made through the pilot period.
• Rolled out to entire chain Spring/Summer of 2005.
  
• Chain-wide results continue to show improvements.

Results
28
Where are we going?

• Extending reach of inventory optimization
  initiatives
• DC Inventory Vendor order optimization
  (multi-echelon)
• Optimized season management
• Optimizing allocation of seasonal items
• Coupled with optimized markdown management
• Other initiatives identified and planned
• Excitement throughout the organization
• Inventory recommendations driving real
  improvement
• Higher service levels
• Lower lost sales
• Stronger sales force selling
• Process seen as building on company culture
• Enthusiasm throughout the company for future
  initiatives

Results