Find Research
Our World View
Industry Perspectives
Research Program
Parallax View Magazine
> stay tuned
View our collections of research around key subject areas:
Momentum 2012 and McKesson's Total-Cost-to-Serve Project

Highlights from Manhattan Associates' annual conference, with a description, lessons learned to-date, and expected benefits from McKesson's implementation of Manhattan's Total-Cost-to-Serve product.

Full Article Below -
Untitled Document

One thing I can say about Manhattan: they know how to stick to a theme. Three years ago their annual conference theme was Platform Thinking;” then last year “Platform Activation.” The theme this year? “Platform Payoff.” It also speaks to the steady nature of their strategy and marketing. Maybe not the most buzz-inducing, but there is something to be said for consistency and stick-to-itiveness. Below are a few tidbits from the keynote addresses at the conference. Then I spend most of this article on what I learned at McKesson’s session on their implementation of Manhattan’s Total-Cost-to-Serve product. Being able to calculate total cost in a meaningful way (i.e. so that you can actually make improvements to performance) is not easy. It is a capability that I think is important and that too few companies have figured out how to do.

Keynote Addresses

CEO Pete Sinisgalli said that Manhattan wants to be the world’s leading supply chain platform and plans to get there by offering a complete logistics/supply chain suite with a common platform, easy integration to other systems, domain knowledge across supply chain management, a superior ROI, and competitive TCO.

COO Eddie Capel elaborated on the conference theme by describing four types of payoffs that arise from having an integrated supply chain platform: Operational Payoff—such as warehouse labor productivity, inventory accuracy, throughput, and lower freight rates; Technology Payoff—fewer systems, more current technology, streamlined integration; Competitive Payoff—The ability to innovate and competitively differentiate; and Financial Payoff—like reducing COGS, increasing turns, deferring capex. The premise here, which I think is sound, is that you can get a certain level of ROI from implementing WMS, TMS, and optimization systems. But when they are integrated together on a common platform, then you gain the end-to-end logistics visibility and the ability to do global, cross-functional tradeoffs and optimization required to get to the next level. Manhattan contends that their customers are now realizing those types of benefits and payoff from their platform investments.

Macy’s Omni-channel Initiatives

Also at the keynote session, Brian Leinbach, Macy’s SVP of Systems Development and Field Services, described what they are doing to serve their multi-channels in an integrated way. Their theme is, “If we own it, she should be able to buy it.” This set of initiatives involves things like: connecting shoppers with social networking while in store; arming sales associates with the tools and information to sell expanded product offerings with confidence (while optimizing inventory and aiming for zero disappointments); innovating with several different flavors of kiosks (designed very differently for each department or purpose); the mobile enabled version of their website with price check features; offers via mobile phone; electronic receipts; digital content delivered in-store; ‘Beauty Spot’ Touchscreen for cosmetics; Google in-store maps; and navigation for mobile phones, mobile payments, and mobile POS. A lot going on there!

Macy’s is also going through some big renovation and re-invention of their stores. And item level RFID. Brian showed a video of a store associate taking inventory with a handheld reader. We heard maybe a dozen or more beeps per second, reading an average sized rack full of clothing in a few seconds. He said that they are seeing 20X improvement in counting speed and 95% inventory accuracy. It’s a big improvement for Macy’s. Instead of doing an inventory count once or twice a year, they can do it once or twice a month.

McKesson’s TCS Implementation

Last year, I reported on Manhattan’s introduction of Total Cost to Serve (TCS), which I think is one of their most differentiated products. It is one of the capabilities that really leverages the integrated platform they tout. This year, I heard about McKesson’s implementation of TCS from Brian Doyle and Brett Harrop, Project Leads for TCS and SCI. For those not familiar with McKesson, it is a healthcare services and information technology company, ranked 14 on the FORTUNE 500. Founded in 1833, McKesson partners with payers, hospitals, physician offices, pharmacies, pharmaceutical companies and others across the spectrum of care to help improve the business of healthcare and patient care. McKesson provides systems for medical supply management, clinical workflow, practice management, pharmacy automation and care management. McKesson is also the largest pharmaceutical distributor in North America, distributing one-third of the medicines used every day.

The Journey to an Integrated Platform

The rollout of TCS should be seen in the context of McKesson’s larger technology roadmap. They made a decision to switch off of their legacy WMS1 system (customized over the years to work just the way they wanted) and move onto Manhattan’s WMS. Importantly for TCS, part of this effort was driving to a consistent set of metrics and engineered standards across all DCs (Distribution Centers). Without that consistency, it would have been much harder to measure total cost across the DCs in a consistent ‘apples-to-apples’ manner.

Ultimately, McKesson will have WMS, TMS, Labor, Slotting, and Inventory Optimization modules from Manhattan running, along with SCI (Supply Chain Intelligence) and TCS. McKesson is using Manhattan’s TM (Transportation Management) for managing inbound and outbound transportation from their DCs, resulting in fewer surprises at the receiving dock door and being more proactive and prepared. WM (Warehouse Management), LM (Labor Management), and Slotting work together to provide the functionality to run their DCs. SCI (Supply Chain Intelligence) is Manhattan’s Business Intelligence (BI) product, based on the IBM Cognos BI platform. SCI gives McKesson full visibility from the time they first receive an order from the customer to the time that order is delivered to the dock door of their customer.

Integrated Platform as an Enabler of TCS

McKesson has been working on rolling out TCS for the past nine months and expects to go live in the next few months. The TCS rollout was dependent on getting some of these other pieces in place. Their TCS system is fed by TM and LM solutions and other systems, such as getting overhead costs from their ERP system. All analytics and dashboards are done by SCI. Having a platform that integrates the systems actually running the operations (warehouse, transportation, labor, etc.) was central to being able to do consistent, enterprise-wide total cost calculations.

Identifying Cost Drivers and Attributes, Calculations for Allocating Costs

A key part of the project was to do a detailed review of supply chain cost drivers, asking the question: “What attributes of a product drive cost in our organization?” For example, cost drivers for allocating transportation costs include attributes like the weight and cubic volume of the product, since both of those attributes impact the cost of transporting. In contrast, when allocating a portion of the rent for the facility, McKesson will use just the cubic volume of the product, but not the weight, since weight does not play a factor in amount of ‘DC resource’ used, but volume does.

For DC rent allocation, they don’t yet use average dwell time in the DC per item, even though that is arguably a material factor in the amount of ‘DC resource’ used. McKesson wanted to keep it simple initially. They might use that attribute (average dwell time) in the future to provide a more accurate representation of the allocation of rent and other facility-based overhead costs.

This is just one example of numerous judgment calls that are made in a project like this about how to calculate and allocate costs—which attributes best encapsulate the cost drivers. A big part of the tradeoff is complexity vs. accuracy/fairness. Part of the art and science is identifying the attributes that are most significant in driving each type of cost.

Decisions on how to apportion the landscaping bill at the DC, for example, are not as scientific or as directly measurable as say the amount and cost of the labor required to put away an item. McKesson will eventually bring in capital expenditures as well, such as the cost of automated handling equipment, introducing more decisions about how those costs should be allocated.

Sources of Data

Other costs that McKesson is trying to measure with TCS include the cost-per-item to receive, putaway, replenish, fulfill order, and transport the products. For these they get feeds from their Labor Management and Transportation Management systems, which in turn are integrated with WM, Slotting, and Inventory Optimization systems. Other costs, such as the monthly overhead costs for the DC, are pulled in from ERP or other systems.


Initially, McKesson is focusing its TCS efforts on measuring costs in customer service and operations, but not in sales. This reflected a recurring theme for them: keep it simple and manage scope and complexity—a good idea for any project, but especially for introducing brand new concepts and technology into the organization. With TCS in particular, it is probably wise to resist the temptation to be over-ambitious in granularity, complexity, and scope, at least for the initial rollout.

Uncovering Opportunities for Improvements

McKesson said that using TCS should help them identify and size various opportunities for improvement. They ask tons of questions and are constantly comparing vendors, DCs, customers, and products to find out why one cost more than others. For example, they can ask what are the right supply chain channels to move various products through? Why would a particular SKU have a different cost to handle in different DCs? How do costs differ by type of customer (e.g. hospital pharmacy vs. small independent retail pharmacy vs. large mass retail pharmacy) and why? Why does one location have different cost-to-serve numbers than another?

There are often valid reasons for differences. For example, all else being equal, the per item transportation costs will be higher in a DC that services a less densely populated area, where the average travel distance is greater. By combining TCS with the analytics tools in SCI (Supply Chain Intelligence), McKesson expects to be able to explore the reasons for differences in costs, find where underlying practices are resulting in efficiencies that they can replicate elsewhere, and optimize per-item, per-shipment supply chain efficiencies based on the actual costs.

Cross-Functional Team

McKesson said that it is critical to get the right cross-functional team on the process. They had large kickoff meetings, led by Manhattan, with key people; the ones who make the 3-5 year strategy decisions. They involved people from operations, the process improvement group, project management, financing, pricing, inventory, transportation, and others. The cross-functional kickoff was a good introduction that got everyone thinking about how their piece fit into the bigger picture.

Having a cross-functional team was necessary for McKesson to get a handle on the cost drivers across the various domains within the company, as well as to better understand the interactions and tradeoffs when making logistical and operations decisions, such as how much inventory to hold where, target service levels, modes of transportation, etc.

Change Management

The cross-functional team also helped with buy-in across the organization. McKesson said that implementing TCS will not be disruptive because, in contrast to a tool like SCI, TCS will be used by a much smaller (but strategic) group of people. It will only be run by the few people who are doing multi-quarter and multiyear analytics across multiple DCs. So far, the TCS team hasn’t run into much resistance. In fact, the response has been more along the lines of “Let me know when it’s here. If you can really pull this off, I will use it every day!”

Expected Results

McKesson expects that the TCS will standardize the numerous analytical groups and their reporting systems onto a single solution to identify and analyze supply chain costs. They expect this to enable them to do consistent, standardized benchmarking of cost and performance across the network, giving them information to identify best practices, make strategic decisions, and enhance their processes and operations.


Recommended Reading:


1 Warehouse Management System -- Return to article text above

To view other articles from this issue of the brief, click here.

MarketViz powered.