3 Ways to Build a More Resilient Supply Chain

Modern supply chains are incredibly complex, and supply chain resilience is often the largest source of risk for enterprises. It’s hard to imagine a more brutally efficient supply chain disaster than the current coronavirus pandemic, which has had a significant impact on both the supply and demand sides of the equation. According to a recent survey, 81% of companies rely on Chinese suppliers, where the outbreak started. All face-to-face transactions are disrupted. 

The impact is global and felt in ways both direct and indirect, and a resilient supply chain will be the deciding factor for whether many businesses are still operating a year from now.

So where to start? Assessing the realistic demand, estimating available inventory throughout the value chain, and managing cash and working net capital are a few of the actions companies should take now, according to McKinsey*.

All of those actions require supply chain analytics data, much of which is locked away in arcane technical systems. We know we need to take action, but how can you leverage the data that’s out there when you’re not even sure what questions to ask yet? Since connecting data to decision makers is what we do at ThoughtSpot, this is definitely a conversation we’ve been having a lot.

Your supply chain is probably not as resilient as you’d like. Let’s take a look at some of the ways you can start charting that course.

Bring your data together

During times of normal business, most executives are focused on keeping their companies moving forward. Systems are built with this in mind, usually containing siloed data needed for each specific function. Finance data is in the finance system. Customer data is in the CRM system.

When things are less certain, we don’t always know where the insight we’re looking for will come from. A supply chain that looks robust in good times may be much more brittle than you expect.

It’s important to have a view of the data that spans your business, not just your function. Some very mature companies have comprehensive enterprise data warehouses, but even those are missing some key sources of data. Data warehouses also take a significant effort to build, time that’s in short supply in today’s world.

Start with the questions you want to ask. Which products are dependent on a single vendor? How many geographies does our supply chain encompass, or is our “global” supply chain highly dependent on one or two countries? Which products are likely to see increased demand over the next few months, and how can we prioritize and strengthen the supply chain for those products? Can we estimate readily available inventory? How has customer demand changed by vertical, geography, and contribution to revenue? Where can we find more working capital?

Next, identify which sources of data you think you’ll need to answer those questions and make them available through a platform, like ThoughtSpot, that’s capable of asking questions across multiple sources. Think in terms of the data, not systems. The goal here isn’t to build a data warehouse, but to quickly answer questions that span multiple datasets, discover an insight, and actually take action.

Improve your reaction time

Getting the data together to form a 360-degree view of your business is a good start, but it’s not enough to actually get value. While we’re fond of talking about modern data-driven businesses, in most companies it’s actually technical analysts who have access to data. The rest of us are limited to answers published as reports or dashboards that only answer questions we asked a week or more ago.

Supply chain resiliency isn’t just about asking the right questions and planning, it’s also about the ability to take quick action. If every new question you ask needs to be interpreted and answered by an analyst, it’s not only challenging to make decisions quickly. It’s impossible to have a dialogue with the data and rapidly iterate to find insights.

In fact, fewer than half of the respondents to a recent McKinsey survey said that they were able to make decisions in a timely manner, and 61% said that at least half the time spent making decisions is ineffective. 

It’s important to create transparency on multi-tier supply chains. An obvious first question is, “how are our vendors distributed by country?” When you see that 98% of your supply comes from China, you might ask a series of questions, each informing the next:

  1. “How are our vendors distributed by country, by volume instead of count?” (Vietnam is 40% by volume)

  2. What products do Vietnamese vendors supply?

  3. Which vendors outside of Vietnam supply the same products?

In a traditional analytic environment, each question needs to be answered before the next can be asked, potentially taking weeks. A modern business intelligence platform built with the type of technology used by Amazon and Google can answer those questions almost as quickly as you can type them.

Identify capacity and working capital

This question-first approach works well when you know or can discover what questions to ask, but finding the unexpected question can be just as important. The questions you’re asking are informed by experience, but the playing field has changed. What if you’re not asking the right question? 

In many cases, you might not know the right question to ask. For example, McKinsey recommends identifying additional logistics capacity and working capital. While generating a hypothesis and iterating rapidly can work for this too, it may be very challenging to find the specific answer you’re looking for when it’s buried in a large amount of data.

Modern artificial intelligence and machine learning can take the effort out of finding the right question. Instead of asking an analyst to run 20,000 permutations of analysis to find one opportunity, a modern analytic platform like ThoughtSpot can run all those combinations in seconds then automatically filter them based on the user’s past behavior so that the most relevant information is highlighted.

One of ThoughtSpot’s customers, a large electronics distributor, calculated that their accounts receivable was costing them over 5 million dollars per day. Identifying patterns in the data like customers who are habitually delinquent or who received unusual terms allows them to take right action to find working capital, with a potential impact of millions of dollars per day. Work that would normally take analyst weeks or months to do can now be done in seconds by anyone with the right access.

These are challenging times for most companies, and by connecting data directly to the business leaders who need it, the best companies will come out stronger. Will you be one of them?

*ThoughtSpot is a technology partner at the Digital Capability Center Chicago, part of McKinsey & Company’s global Capability Center Network.