Having stated his positions for enacting import tariffs and bringing manufacturing jobs back to America, our new President, Donald Trump, has his work cut out for him. I’ll refrain from making political commentary, but I can venture some predictions on what will come into focus for supply chain performance in 2017:
Prediction 1: Manufacturing won’t move on-shore
Information, on the other hand, must be accessible both centrally and in decentralized locations
Will tariffs and incentives to bring manufacturing jobs to America cause factories to pop up overnight, making the U.S. a center for industrial production?
In short: No.
The capital investments have been made, the labor or automation has been established, and the business terms have been set in various countries outside the U.S. borders. What will need to occur is central visibility for U.S.-based companies into the impact of new regulations on production, shipping and tariffs, to calculate profitability.
In Trump’s America, products or components sourced in China or Mexico will be subject to new tariffs. Therefore, companies will want to re-examine their supplier base countries of origin. As tariff fees come into play, they must be factored into landed costs of goods, balancing excise taxes with transportation costs and lead-time requirements, to determine the right mix. This information – supplier profiles, tariffs, shipping costs, materials costs – has historically been housed separately across strategic sourcing, procurement, operations planning and global trade systems. But it must all be viewed together to plan profitably and make decisions quickly when supplies fall short.
Business units will also want to identify and work with manufacturers located closer to where they sell goods to reduce exports. These considerations may further decentralize manufacturing, so that each regional business unit makes its own planning, purchasing and production decisions. For regional business units to operate with autonomy, they need reliable information they can drill into, as conditions change. However, the analysis must be performed within the centralized corporate procurement data, not a maverick set of agreements. Otherwise, each regional unit will operate as a silo, resulting in separate purchasing entities with no volume or scale.
Visibility into preferred suppliers and blanket purchase agreements will enable manufacturing sites to quickly shift buying arrangements to achieve optimal purchasing power. Bringing information from separate product lines, regional business units and acquisitions into a consolidated view will enable strategic sourcing to negotiate discounts and source materials without redundancy.
Prediction 2: Machine learning will become accessible for supply & demand adjustments
In the past, prioritization came down to the “Fast / Cheap / Good” three-point framework. The old adage was always, “pick two.” But in the supply chain, options have become more complicated than just isolating “Fast and Cheap,” versus “Fast and Good.” In 2017, as the pace of business continues to accelerate, we face sliding scales rather than binary choices. Due to constantly shifting constraints, the answer to this question has become, “What is right for right now?”
Silicon Valley has been talking about machine learning for years, mostly around a few deeply algorithmic problems such as biomarker identification, fraud detection, and social media monitoring. Applying machine learning capabilities to the enterprise supply chain, to identify trends and present recommendations, has thus far been out of reach. By automating machine learning to help select predictive models and leverage historical data to drive continuous tuning, data science at last becomes relevant to the supply chain.
If you’re working with 300 suppliers, 10,000 customers, and 20,000 components, the volume of data is too much for analysis, much less optimization. Add changing parameters from weather, shipping errors, geo-political events that impact supply, and then demand shifts, and calculation times become infinite. Automated machine learning speeds up the manual aspects of data science to deliver relevant business outcomes with “best-fit” models for forecasting and supply-demand match.
Lead time to get material to the factory can now be considered alongside cost of goods, shipping and tariff costs, as well as potential revenue losses for delays. In a volatile market, understanding the impact of decisions allows for rapid adjustment across the faster, cheaper, better spectrum, depending where priorities lie.
Prediction 3: Shared analysis connecting internal partners and external players will drive collaboration toward common goals
Exporting a data file or emailing a spreadsheet has defined how we share analysis thus far. In 2017, the ability to view common dashboards that reflect adjustments in real-time will make information consistent across all nodes of the supply chain. This capability will manifest itself in two primary ways – new customer offerings, and networking the ecosystem.
- The supply chain will drive after-market offerings: In 2017, as more markets converge, analytic offerings will be the key to differentiation and customer acquisition. Most companies already capture data about sell-through, equipment reliability and usage, but many do not re-package it meaningfully for their customers. By aggregating and visualizing information with KPIs and benchmarks, analytics become a product or service offering of its own. Packaged as a stand-alone offering, analytics yield higher margins in the form of product differentiation, customer retention, and up-sell service bundles with product sales. When customers are given transparency into the performance of what they’ve purchased, they develop greater levels of trust and loyalty, in even the most saturated market. The alternative is that sales strategies focus on price, which is a race to the bottom.
- A network of external suppliers and partners will put a focus on shared success: In 2017, companies will be empowered to share information between themselves via extranet capabilities that maintain secure, yet transparent, collaboration. From customers and suppliers to partners, businesses can share analytics outside the four walls of the company – exposing product usage and performance analytics, as well as benchmarking – via shared dashboards. As discussed above, when packaged as an offering, this information becomes a product or service. But shared within an ecosystem of suppliers and partners, it becomes a tool for improved performance across the value chain. In many cases, these separate entities have had access to analytics for years, but they operated individually and in disconnected fashion, emailing spreadsheets or exporting EDI files to share information.
On the other hand, when each node of the supply chain is connected and sharing insights, the linear nature of processes evolves into a synchronized network, serving common goals for the best product, at the lowest cost, with the fastest time-to-market.
Conclusion: As President Trump moves into office, many pundits are focused on policy impact to infrastructure and physical operations. The focus should instead be on the information. Company supply chains have been collecting data for years. Many are now looking critically to say, “Where’s the value?” The emphasis for 2017 should not be on amassing more data to create new reports. Top performers will instead be focused on how information can used both locally and globally to make decisions within the business, as well as across the ecosystem of partners, suppliers and customers.
To watch an on-demand demo of Birst’s supply chain analytics solution, click here.