The WiseTech Question:
When AI Becomes a Headcount Strategy

On 26 February 2026, Australian logistics software giant WiseTech Global announced it would eliminate approximately 2,000 jobs — nearly 30% of its global workforce — over the next two years. The company's CEO, Zubin Appoo, was admirably direct about the reason. "The era of manually writing code as the core act of engineering is over," he said. Projects that previously took six or seven months can now be completed in a single day. A global customs rollout that once required up to two years now takes a matter of weeks. The market, which had spent months worrying that AI would make WiseTech obsolete, responded by pushing the share price up eleven percent in a single session.

So: good news for shareholders, mixed news for engineers, and a genuinely important question for the rest of Australian business. Because what WiseTech has done is not particularly unusual, or even particularly surprising. It is simply unusually visible. And it represents a choice — one that other companies are making right now, quietly, in businesses of all sizes across the country.

The choice is this: when AI makes your people more productive, do you keep your people and grow, or do you keep your margins and cut?

Source: WiseTech Global ASX announcement, February 2026

The Efficiency Argument, Stated Fairly

WiseTech's position deserves to be understood on its own terms before it is critiqued. The company operates in a highly competitive global market. Its share price had fallen 68% from its November 2024 peak — partly due to governance concerns surrounding its founder, but partly because investors were genuinely worried that AI would undercut its core product. The announcement of deep AI integration, combined with a cost base reduction of this magnitude, was a direct response to that existential pressure.

And the productivity numbers are real. When a software task that took six months can now be completed in a day, you do not need the same number of people to do it. This is not a corporate fiction. It is arithmetic. The question is not whether the arithmetic is correct — it is — but what you choose to do with the answer.

— Industry commentator, quoted in news.com.au coverage of the WiseTech announcement

Interestingly, WiseTech itself acknowledged that not all roles are replaceable. The company noted that positions requiring direct customer interaction — sales staff, in particular — would remain. Which is, if you think about it, a fairly significant admission. The company is cutting the people who build the product while retaining the people who sell it and build relationships around it. This suggests WiseTech's own leadership understands, at some level, that there are things AI cannot do.

The Other Way to Do the Maths

Here is an alternative set of calculations that, as far as we are aware, did not appear in any of the analyst reports published after WiseTech's announcement.

If AI genuinely allows your engineering team to complete in one day what previously took six months, you have not merely reduced your cost per feature. You have also dramatically increased your capacity to build features. A team that could previously ship two or three major capabilities per year could, in theory, now ship dozens. You could expand your product's capabilities at a rate that no competitor matching the old model could keep pace with. You could move into adjacent markets. You could tackle technical debt that has been accumulating for years. You could build the things your customers have been asking for since before anyone had heard of a large language model.

In other words, you could use the productivity gain to grow rather than to shrink. You could keep your people and do more with them, rather than doing the same with fewer of them.

Both approaches are arithmetically coherent. Only one of them treats the people who built the company as an asset rather than a liability.

What This Means for Small Business

WiseTech is a publicly listed company with 7,000 employees across 40 countries, reporting to institutional shareholders who have watched the stock price fall for seven months and would very much like it to go in the other direction. The pressures it faces are real, and the decisions available to it are constrained in ways that are specific to its situation.

The small and medium business owner reading this article is almost certainly not in that situation. They are not fielding calls from fund managers. They are not managing a share price. They are running a business — probably one built on relationships, reputation, and the specific capabilities of a small number of people who know their customers, their industry, and their craft.

For that business, the question of how to deploy AI is entirely different. And the answer, we would suggest, is almost always the same: use it to make your people better at what they already do well.

The Empowerment Model

Consider what AI actually does well when deployed thoughtfully inside a small business. It handles the work that consumes time without requiring judgment. The first draft of the proposal. The follow-up email that should have gone out three days ago. The summary of last quarter's sales data. The social media post that needs to go up before 9am. The job ad that needs to be written by Thursday. The research that should have been done before the client meeting but wasn't because there were six other things happening simultaneously.

None of these tasks require a human to do them poorly and slowly. All of them benefit from human oversight and finishing. And when a person is no longer spending two hours a day on first drafts, follow-ups, and data summaries, they are available to do the things that genuinely cannot be automated: building the client relationship, solving the unusual problem, making the judgment call, being present in the conversation.

A bookkeeper who uses AI to handle initial data entry and basic reconciliation is not being replaced by AI. They are becoming a financial advisor who happens to also handle bookkeeping. A marketing coordinator who uses AI to generate first drafts and schedule content is not being automated out of existence. They are becoming a strategist who happens to also execute. The job title may stay the same. The actual value delivered changes considerably.

The Trust Economy

There is also a practical commercial argument that tends to get overlooked in coverage of AI workforce stories, and it was captured neatly in one of the responses to the WiseTech announcement: customers will still want empathy, reassurance, and human judgment.

This is not a soft observation. It is a competitive reality. In a world where an increasing number of businesses are cutting customer-facing staff and replacing human interaction with automated systems, the business that still picks up the phone, sends the personal note, and has an actual person available when something goes wrong has a genuine differentiator. Customers notice. They remember. They tell people.

For small and medium businesses, this is not an abstract point about brand values. It is a tangible commercial advantage that requires exactly zero additional investment — because the people who provide it are already there. AI, deployed properly, means those people spend less time on the tedious and more time on the valuable. The outcome is a business that is both more efficient and more human, which is a combination that tends to be rather difficult to compete with.

A Word About WiseTech's Customers

One detail from the WiseTech announcement that deserves more attention than it received: at the time of the announcement, eleven of WiseTech's largest freight forwarder customers had less than 20% of their expected users live on the CargoWise platform — meaning major implementations were mid-rollout when the company announced it was cutting its engineering and support teams by up to half.

Whatever the long-term logic of the AI transformation, these are real businesses with real operational dependencies on a product that is about to lose a significant portion of the experienced humans who understand it. The next six to eighteen months — as seasoned engineers depart, AI systems mature, and support processes are re-wired — will be the period of highest exposure. For those customers, the comfort of having a human being who actually knows the product and can pick up the phone is about to become considerably scarcer.

This is what happens when AI is used as a headcount strategy rather than a capability strategy. The efficiency gains are real. But so are the gaps left behind.

The Reasonable Conclusion

WiseTech made the choice that made sense for WiseTech's situation. Two thousand people are now looking for new roles, and the market has judged the decision favourably in the short term. Whether it proves wise in the longer term — whether a leaner, AI-led WiseTech outcompetes, or whether it discovers that the institutional knowledge it shed was harder to replace than the headcount saving suggested — remains to be seen.

For most Australian businesses, however, the WiseTech model is the wrong template. Not because AI is not genuinely transformative — it is — but because the value AI creates does not have to be extracted by reducing the number of people in the room. It can just as readily be extracted by making the people already in the room dramatically more capable, more confident, and more valuable to the customers who depend on them.

That, in the end, is the choice. AI as a mechanism for removing people, or AI as a mechanism for elevating them. Both are available. One of them tends to build something worth having.