Data technology and analytics to improve planning and procurement

Global IT services and consulting company DXC has more than 130,000 employees and generated US$17.7bn in revenue in 2021. It spends billions of dollars on IT equipment and services every year. Chris Drumgoole, COO of DXC who joined the business in 2021, understands transforming how this procured and paid for (the ‘procure-to-pay’ process) is a key priority.

“We had, and still do have, numerous purchasing tools and systems around the world, some of which are highly automated, and some highly manual,” explains Drumgoole. “This meant that we could not get a holistic view of how much we spend and who we spend it with. This leads to increased costs, because you spend more for the same commodity with one supplier than another, without knowing it. It was also difficult to get an idea of where delivery delays were affecting supply.”

To ensure a comprehensive view of their purchasing data, the business is centralising its procurement infrastructure, enabling its operations team to generate instantly analysable data that allows comparison of different prices, delivery times and geographies. Once DXC can lay out its insights in a graduated framework, purchasers will be able to make better-informed purchasing decisions, which may potentially lead to more favourable terms.

Importantly, local purchasers will still have autonomy around procurement. “We want global visibility, but local execution, because smaller local purchasers have greater agility and can get a better deal than purchasers in a big central team. That said, centralised systems also give us control to an extent, which enables us to comply more easily with legal directives such as sanctions.

Close collaboration between Operations and Finance is essential to the restructuring of the procure-to-pay system. “Operations teams focus on how we can drive efficiencies; how can we get key equipment more quickly; and how we get clarity on our supply chain,” explains Drumgoole.

Mondelez is a global food and beverage company that operates in 160 countries, generating more than US$25 billion in annual revenue across brands including Cadbury, Daim, Oreo, Toblerone and Ritz. Frank Cervi, the company’s Integrated Supply Chain SVP is interested in utilising new, advanced data technology to automate demand forecasting without compromising on data accuracy.

For Mondelēz, it is vital to predict accurately demand levels for its goods. These projections will determine the scale and timing of production for each product line and will be used to program high- and low-intensity periods for each factory. Cervi leads the efforts under way to transform the company's supply-chain planning system. Optimizing the accuracy of these forecasts is a top priority.

“Historically, these forecasts were determined by manually analyzing hundreds of spreadsheets and then going through lots of internal discussions,” he says. “Now, we're using a set of sophisticated, machine-generated algorithms to generate these estimates. It compares everything you forecast in the past to what actually transpired.”

Currently, the algorithms are set to analyze internal data, such as historical demand levels. They also determine the optimum conditions for the implementation of current promotions, using external data such as weather forecasts and the timing of major sporting or other entertainment events when there are spikes in demand for snacks and beverages.

As a result of all this data analysis, the company has managed to improve the accuracy of its two-month-ahead demand signals from 70% (which Cervi says was the industry benchmark) to around 75%. The benefits are even greater for longer-term forecasting, a notoriously difficult undertaking given the wider range of possible outcomes.

Cervi believes there is even greater scope to improve accuracy by systematically collating and feeding in greater supplies of data. “In the future, I imagine being able to scrape all point-of-sale data from supermarkets, as well as social-media metrics, to understand whether one of our brands is trending. This could help us to understand demand surges and generally fine-tune our forecasting.”

Obtaining the data and feeding it into the data analytics system is not the greatest hurdle for Mondelez. Cervi explains, the main challenges lie in budget allocation. “I know from speaking to peers in other CPG companies that securing budget for these types of digital investments is challenging. Finance will want to understand the return on investment and the payback period. But, with digital investments, you’re often not 100 percent clear on the benefits, yet you know you need to make it, or you will be left behind. We managed to overcome this because we have a visionary head of supply chain and a dynamic Finance team that approved the investment. They worked together to understand the benefits.”

In both of these examples, close collaboration between Operations and Finance proved vital, both in planning the refinements and understanding the motivations and intentions behind them.

However, this is something that many businesses struggle with today: just 43% of surveyed Operations leaders are 'very satisfied' with the extent to which they collaborate with their counterpart in Finance. And if those that lead these functions do not collaborate, neither will their teams: just 46% of Operations leaders are 'very satisfied' with the degree of collaboration between finance and operations team members.

Read our main report (PDF 587KB) to find out how Operations and Finance can work together for mutual advantage and to the benefit of the business as a whole.