What the 90s taught us about ERP: A legacy of failure

In the 90s, in the midst of the technological race driven by the internet boom, a series of important phenomena marked the market. On the positive side of this change, the digitisation of processes in companies has accelerated and has made room for emerging solutions that have brought more productivity and efficiency to the business.

On the other hand, this scenario also stimulated contexts such as the so-called internet bubble that resulted in significant losses for the economy: shortly after the turn of the millennium, for example, the Nasdaq lost more than $7.6 trillion in market value. 

Another joint effect of this scenario was the search for companies to create their own business management solutions (ERP), a move that soon proved to be a burden: heavy legacy systems that were expensive to maintain and update, hindering the innovation process of organisations.  

Remembering the lessons learned from that time is essential to avoid a new wave of decisions based on haste, lack of vision and planning. After all, within the new AI race, many companies are following the same erratic path of the 1990s, and this could be a recipe for disaster.

There is no doubt that the promise of generative AI, predictive algorithms and their already quantifiable benefits are vectors for the advancement of more innovation in the market, but it is not only possible to absorb these benefits through strategic alliances, but this path can be much more effective and reduce the costs of development and technological infrastructure for companies.

An example in this regard can be seen in Endries, a global supplier of industrial parts and components for assembly, which after adopting Infor WMS – which has customised AI resources for different sectors – was able to optimise its productivity by 40%, significantly improve warehouse traffic and validate in less than 90 days the value of the company's artificial intelligence in its logistics operation.

It is no coincidence that in the supply chain universe, we see that investments in AI could reach more than US$157 billion globally by 2033, according to a projection published by Global Trade magazine.

However, the truth is that haste, lack of vision and strategy can end up turning these investments into waste. Once again, we see that companies decide to invest in their own teams to develop their own solutions, believing that they will be able to create competitive differentials by internalising the entire process. It is the almost literal repetition of the illusion of the 90s: the belief that it is possible, alone, to overcome technological frontiers.

But the reality, as we have seen, goes against this view. Firstly, because the demand for talent in technology continues to grow, but the supply of labour does not keep pace. An additional 286,000 IT professionals are needed to meet the sector's needs in Australia alone, according to the Tech Council of Australia. Continuously, the need for professionals with AI skills is growing more than 20% per year, according to a global survey conducted by Bain & Co. And although the World Economic Forum projects that AI will open 78 million jobs by 2030, filling them will be the great challenge. The world is not ready yet.

Moreover, the most elementary prerequisite for any AI project to work, the organisation and quality of the data, is still far from most companies. Dell's Data Paradox study points out, for example, that only 21% of companies in the world know how to properly handle large databases. 

It is no coincidence that some of the most relevant private investments in the market are taking place in the data infrastructure of organisations. Salesforce recently bought Informatica for US$8 billion; DataBricks acquired Neon for US$1billion dollars; and Meta is in the process of making an investment of more than US$10 billion with ScaleAI, according to the TECH Drops newsletter. 

In other words, the real race is not just for AI, but above all, for the data that feeds it.

Given this panorama, companies such as Infor – which since the early 2000s has been dedicated to creating industry-specific ERP, covering up to 2,000 micro-verticals – are betting that, instead of reinventing the wheel to succeed in implementing AI, a more strategic path is to invest in solutions, concomitantly, adapted to different markets and, above all, that they have already been tested, validated and that continue in a process of continuous innovation. 

Such insight tends to allow for an assertive analysis of where AI can generate more positive impacts for the business, less time for implementation, control of operational risks, such as data compliance, and a clear commitment to delivery and results. The return on investment becomes faster and more predictable.

This is where the big decision of business leaders lies right now. It's not about ignoring digital transformation or outsourcing strategic business intelligence. It's about recognising that the adoption of technology, especially more advanced emerging ones like AI, requires pragmatism and long-term vision.

When companies try to develop algorithms on their own, train them with data that isn't ready, or integrate them with systems that don't communicate with each other, they're wasting time, money and competitive advantage. What's more, they're moving away from what matters, which is the ability to generate value in a continuous, reliable and scalable way.

Infor's global study, "How the Possible Happens," found that 80% of Australian entrepreneurs agree that business success will depend on the adoption of new advanced technologies. In fact, 79.3% of them say they will accelerate technology investments by 20% or more in the next 3 to 5 years to reach a new level of productivity.

But investing is only the first step. In the current context, it is essential not to repeat the mistakes of the past and to look for specialised partners, so that AI can be a bridge to greater efficiency, strategic gain and generation of value for the business.