Palo Alto Networks didn’t see Covid-19 coming but was better prepared than most when the global supply chain abruptly froze. The cybersecurity systems business had been digitizing its supply chain for several years and had modelled for various sourcing extremes.
“I wouldn’t say we were more prepared, but we had things in place,” explained Jonathan Morgan, senior director of sales, operations and planning (SO&P) for Palo Alto Networks. “We carried a large buffer of inventory for our new products, and we already had demand signals out to our suppliers so we didn’t need to expand our forecasts.” By running through both upside and downside supply scenarios the company identified the key components it had to have on-hand.
The $4.3 billion business builds firewalls and appliances that run its security software and was on a mission to improve its supply chain operations. Palo Alto Networks partnered with Anaplan, a business planning software firm, to reduce its reliance on spreadsheets, messaging and line-item orders to fulfill its bills of material (BOM).
“We had 800 parts we were using, manually tracked by spreadsheets. It was a ‘bear,’” said Morgan, a self-identified evangelist for efficiency.
In its first few years with Anaplan, Palo Alto Networks moved its SO&P process onto the tool and initiated demand, capacity and logistics planning. Instead of building to forecast, the company was building to demand. Enabling this was component sourcing in square sets, or kits, which contained all the parts necessary for a finished product.
“We were able to set our supply targets based on finished goods; accounted for fluctuations in demand; spread the demand out and then rolled it forward,” Morgan explained. “We always had a horizon of supply sent out to our suppliers to keep our purchase orders open. Square sets helped. If you have 99 percent of the parts you need it doesn’t do you any good. Square sets helped solve that problem.”
Item-by-item planning also requires managing lead times for individual parts. Eliminating that process enabled huge gains in efficiency.
Barriers to digitization
One of things holding the digital supply chain back is the cost and complexity of solutions. ERP and MRP systems are expensive and time-consuming to implement. The return on investment (ROI) of such efforts are difficult to measure: efficiency saves costs but doesn’t directly drive sales. But with cybersecurity threats spiking, Palo Alto Networks needed to keep its production lines moving. The investment wasn’t a hard sell, and the tool was up and running in 10 weeks.
“We were able to put tools in place to keep things from becoming a problem,” said Morgan.
“As people got better at adopting Anaplan – the interface was easy — we wanted to get more functionality out of the tool,” he added. It’s now used for SO&P; demand planning; MRO inventory, capacity planning and outgoing freight needs. Like many companies, Palo Alto Networks ships a lot of products at the of a quarter.
The electronics supply chain is designed to keep inventory lean – a dilemma for businesses that support their products for a decade or more. Chip makers want to move on to new technologies, and holding inventory ties up manufacturers’ capital; loses value as it ages; and may end up written off as excess.
Palo Alto Network’s products have a lifecycle of 9 or 10 years. “Ten years is a long way to plan out a forecast,” Morgan acknowledged. The enterprise runs models comparing the cost of holding inventory to the risk of running short. “We’ve been able to de-risk some of these things where we know we going to have demand [for spares] to come.”
The company now:
- Plans all of its buffers through finished goods and compares those buffers to baseline plans in real-time
- Identifies what’s short to buffer vs. what’s short to revenue needs for faster course correction
- Connects demand plans to supply and financial plans to see how different forecasts/scenarios will impact revenues
- Easily loads different dollar views to compare what costs a supplier is driving vs. Palo Alto’s own cost of goods sold (COGS) and average selling prices (ASP), and communicate that with suppliers, manufacturers, and finance teams
- Scales to add new products and suppliers as Palo Alto expands its product portfolio
- Makes intentional decisions about the impact of carrying inventory to meet future demand
Morgan measures success in several ways — there are fewer spreadsheets (which was somewhat of a cause célèbre.) “We have developed a new production introduction (NPI) model using this system,” he added. “We look at actual and historical consumption and our suppliers on one source map; we not only capture actual demand, we can look at customer segments to see where demand is increasing. It takes some time to get to the next project.”
Palo Alto Networks is also mapping its supply chain to identify pain points such as shipping delays due to inspections or backlogged ports, and how inventory needs change during a product’s lifecycle. “We have a data-driven view of what our expectations look like, so we can build a baseline and share information with other departments such as marketing or finance,” said Morgan.
“We focused on our needs rather than chasing buffered supply,” he added. “We look at customer requirements for installation so we can make [those distinctions] and focus on having materials for the right build.”
Like all tech companies, Palo Alto Networks has felt the effects of the global chip shortage. “It’s been significant — its had an impact on boards as well as power supplies,” Morgan said. “But we’ve been able to build the right things at the right time.”