Signpost: Decision-making is
the secret sauce 

There’s nothing like a surprise tariff hike to ruin a perfectly good export strategy. One moment, your goods are halfway to Miami; the next, they’re too expensive to land.

This is the new normal for manufacturers, where planning is provisional, forecasting is guesswork, and the cost of hesitation is written in red ink. The rules don’t merely change; they ambush you, faster than an impromptu summit at the White House.

The only way to survive is to make decisions faster than the chaos around you.

That became clear during a visit to the Sunninghill offices of Syspro, a South African-founded enterprise resource planning (ERP) company with a global footprint. In an unassuming office park just off the Johannesburg Ring Road, a quiet overhaul is under way. ERP systems, once seen as slow-moving repositories of back-office data, are being recast as active engines of real-time operational intelligence.

“You need to be in front of those decisions,” said Jaco Maritz, recently appointed global CEO of Syspro, following its acquisition by private equity investor Advent International. “It’s not just about the system that runs your manufacturing. It’s about having the information you need to make the right decisions fast.”

Syspro is rethinking not only how its systems function, but how it functions as a company. Regional silos are giving way to global roles. Leadership is being structured around capability rather than geography. Product, customer, engineering and revenue are being managed as a single system, not parallel workstreams.

The business now spans a sharply defined global footprint. The Americas contribute 55% of revenue, followed by Africa – primarily South Africa – at 20%. The United Kingdom makes up 15%, and the Asia-Pacific region 10%.

Strategy is being redefined in terms of return, not tenure.

“Your business is no longer your forever home.” said Leanne Taylor, the company’s new chief revenue officer. “This is your investment home. If you want to extract value, you modernise it.”

Modernisation, in this context, means surfacing the value already embedded in existing deployments, and expanding it through tools that are often already present.

“Many of our customers still do their scheduling on a whiteboard,” said Maritz. “So when they ask about AI, the answer is: first, let’s get your scheduling into an automated system. Then let’s talk about using AI to optimise it.

“But when we started listing all the AI we’ve already deployed, we realised much of it was there already. We just hadn’t called it AI.”

The list includes machine learning models for predictive analytics and algorithms for anomaly detection in forecasting.

Said Taylor: “A lot of what’s called automation today was always AI. The difference now is agility. We need systems that don’t just process data but adapt in real time – especially when tariffs flip-flop hourly.”

That agility is not theoretical. Maritz gave the example of a Chicago-based client that halted a shipment to Canada within an hour of a tariff announcement, rerouting product back to its warehouse. The decision was not based on instinct. It was powered by information.

AI did not simply support the decision, but enabled it to happen at speed.

The operational environment driving this shift is not forgiving. Tariffs shift with politics. Supply chains remain unstable. Mid-sized manufacturers, who form Syspro’s core market, no longer have the cushion to absorb delays or mistakes.

“They’ve survived by doing things the same way for decades,” said Taylor. “But that’s no longer viable.”

She described the need to run in two lanes simultaneously. One is the fast lane: what customers need now. The other is the long lane: what the market will need in two, three or five years.

“If we don’t plan for both, we’ll get stuck responding instead of leading.”

A major challenge is that countries are building divergent compliance regimes and data standards. Without real-time visibility, and without AI agents to interpret shifting rules, global operations will slow to a crawl.

“You can’t make global decisions with local spreadsheets,” said Taylor.

The statement may be simple, but it frames a complex challenge.

Said Maritz: “We’re not trying to automate decision-making. We’re trying to make decision-makers faster and better at what they already do.”

* Arthur Goldstuck is CEO of World Wide Worx and author of ‘The Hitchhiker’s Guide to AI’. Follow him on Bluesky on @art2gee.bsky.social.