Decision-Making Processes

The research uncovered that one reason for much of the lack of analytical maturity was that current reporting and analytical capabilities were designed around either legacy or sub-optimal decision-making processes. Supporting this was a pervasive finding that CPG companies have historically been investing in analytical tools in advance of broadly understanding how those tools would support and optimize their cross-functional decision-making processes. Without the proper business context, it is difficult for analytically lagging CPG companies to get the right information, at the right time, to the right place, even with the correct tools in place. 

CPG companies with demonstrated analytical maturity have taken a business-led approach to understand cross-functional decision making prior to, or in lockstep with, investments in tools and techniques. This approach treats decision making more like a science than an art. The most advanced CPG companies profiled in this research took this approach one step further by looking at decisions that involved entities external to their organization focusing on collaborative decision making and joint business planning.





For an example of how a Big Data use-case was jointly designed and executed between a CPG company and one of their largest retail partners, see below for a case study written by management from PepsiCo and Safeway.

There will be no shortage of options and opportunities to leverage Big Data to enhance cross-functional decision making. However, companies need to be cognizant that there is a fixed or limited amount of internal resources capable of developing the required supporting infrastructure. This constraint will make understanding business context even more critical. CPG companies will need to have the capability to quickly assess which decisions should be given a higher priority (e.g., impact to top- and bottom-line growth). Once priority has been assigned, the organization needs to be able to efficiently target and establish the sources of existing data, and emerging Big Data, that can be converted to insights that best support these decisions.  (See Key Recommendations and Conclusion #4 for further details).

Below is a case study focused on the integrated decision making process between PepsiCo and Safeway. This case study illustrates the decisions both organizations were trying to make, the questions they were trying to answer, and the business use cases that required collaboration between their two organizations. The example with PepsiCo and Safeway (below) started with a business use case from where it was determined that data visualization was an optimal tool to improve information and insights. Below Kirsten Curtis, Director, Demand Planning and Business Intelligence, Safeway describes their experience.

Deloitte Hosts PepsiCo and Safeway at the HIVE (Highly Immersive Visual Environment)


Clay Broussard - Director, Customer Supply Chain and Logistics, PepsiCo

Monica McCombs - Vice President, Supply Chain Operations, Safeway

Kirsten Curtis - Director, Demand Planning and Business Intelligence, Safeway


His eyebrows raised in disbelief.  “Are you sure?”  Taking his question as either a mild attempt at insubordination or just pure laziness, I assured him that I did indeed want the past month’s point of sales data for PepsiCo’s 19 mega brands from all 1600 Safeway stores.  In an effort to improve supply chain visibility, Safeway had launched a data sharing program with key vendors, designed to increase sales and reduce costs along the entire supply chain.  I wanted to see for myself what data we were sharing.


Four days later 5 flash-drives were delivered to me.  “Here’s the data… just like you asked.”  Big data doesn’t even begin to quantify the terabyte plus of data I had sitting on my desk.  I sheepishly installed the first drive to find a spreadsheet (and there were several on the drive) with over one million rows of data.  Now I understood his uncertainty around the request.  How could patterns and trends be identified for optimal decision making from data presented like this?


It was a blessing to receive Clay’s call.  Clay Broussard, Director of Customer Supply Chain & Logistics at PepsiCo, had been engaged with the Safeway Team on data-sharing since the project’s inception.  “Safeway’s Data Visibility program is very forward thinking.  Paired with PepsiCo’s 360* Retail execution program, our teams are equipped to improve an already lean supply chain.  But if we really want to take it to the next level, we need a different way to view the data.  Deloitte Consulting has offered to partner with us to provide an effective way to interpret massive amounts of data at its state-of-the-art visualization center called the HIVE… are you interested?”  Was I interested???  It was evident a data visualization strategy was paramount to our success.


The HIVE (Highly Interactive Visualization Environment) lived up to its name.  HD touch-screens mounted on bright green walls, it was a cross between a war room and an atelier—a place where serious technology and business met creativity and innovation.  Deloitte Consulting, PepsiCo, and Safeway collaborated to understand how days of supply could be reduced from the supply chain while at the same time maintaining service levels, a project that would save both PepsiCo and Safeway millions of dollars. 


Visual maps were mocked up to see where inventory (both PepsiCo’s and Safeway’s) was above and below targets, and guidelines around prioritization were established.  Info-thread-mapping enabled the teams to dive into the first layer of root cause analysis.  “Big Data” is often talked about, but analytical visualization tools like geo-spacial maps and info-threads quickly isolated conditions and root causes within enormous data sets, targeting areas of focus for collaboration and supply chain optimization.


But our work at the HIVE was just the beginning.  In order to react to what the data was telling us, it became clear that business processes and joint targets would have to be created and communicated between Safeway and PepsiCo.  Yet data visualization enabled us to have a common language and tool for the foundation of that collaboration.  As such, topics like days of supply, on-shelf opportunities and service level are quickly becoming sexy and strategic.  Add to that the monetary potential from even minor improvements in these areas, and it’s easy to see why these metrics are garnering attention from vendors and retailers alike. So the next time someone asks if I’m sure I want big data, my reply will be, “How many flash drives do you need?”

Authored By Kirsten Curtis


Previous Section:

Information and Insights

Recommendation and Conclusion #1

Next Section:

Talent and Organization