As published in IT Wire.com
While AI can be the catalyst for significant gains, organisations that don’t implement it smartly will struggle to realise them.
Have you recently introduced AI to your operations and been underwhelmed by the results it’s generated or, more to the point, the lack thereof?
While faster growth, increased productivity and an enhanced customer experience are the usual goals for businesses putting AI into production mode, they’re proving remarkably unattainable for many businesses.
So much so that an extraordinary 80 per cent of AI projects to date have been deemed failures, according to University of Queensland research. Indeed, by the end of 2025 Gartner reported that at least 50 per cent of Gen AI projects has been abandoned after proof of concept.
This is a missed opportunity and is avoidable. And, in most cases, it’s not the AI technology itself that’s at fault; it’s how organisations have gone about the business of deploying it.
Here are some common issues causing AI projects to come a cropper and how you can avoid them.
Solutions in search of problems
An AI solution shouldn’t be an end in itself. It’s not something to implement just because you can, or so senior leaders can be assured your organisation is doing something – anything! – to keep up with the competition. But, in many businesses, that’s exactly what’s happened and is continuing to happen. AI is deployed with insufficient analysis of the problems it’s intended to solve, including the question of whether they’re worthwhile solving at all. If you’re smart, you’ll take the opposite tack: first defining the business outcomes you’re seeking to achieve and then making a study of how AI can help.
Promising pilots that fail to scale
It can be easy enough to make something work in a contained environment and on a small scale. But does that mean it’s worthwhile investing time and money putting it into production across your operations? It’s a $64 million question too many businesses don’t stop to consider – and it’s a big part of the reason many projects fail to deliver the anticipated outcomes. Last year, Forrester found that only 10-15 per cent of AI pilots successfully scale into long-term production. Validate the value of yours before scaling and it’s less likely to form part of this regrettable statistic.
Workforces that aren’t on board
While the prospect of an AI implementation may generate optimism in the senior ranks, sentiments can be markedly different further down the line. Fear, uncertainty and doubt are typically in ready supply, as employees leap to the not unreasonable conclusion they may be automated out of a job. Not acknowledging and addressing their concerns is a serious mistake and one that may lead to significant resistance to your AI initiative, both overt and covert. Be clear and transparent about what you’re doing with AI and design your Human + AI workforce carefully and you’ll stand a better chance of obtaining staff buy-in and support.
Governance and risk coming late to the party
AI projects are not without risks. Fail to identify, quantify and mitigate those risks in the planning process – a common omission for project teams racing to get AI initiatives up and running – and you’ll pay the price later. That could be when your governance and risk team calls a halt to the deployment at the eleventh hour, or after your AI solution has gone live and caused damage to the organisation’s revenue, reputation and customer experience. Embed governance in your AI adoption strategy from day one and you won’t find yourself battling last minute roadblocks.
The power of professional support
Standing up an AI project successfully is far from straightforward, even if the technology itself appears deceptively simple.
That’s why it can pay to enlist expert assistance, from consultants with a wealth of experience deploying AI across a gamut of settings.
Ideally, you’ll elect to work with a partner that can demonstrate they’ve delivered genuine benefits, in the form of productivity gains and customer experience improvements, for businesses experiencing similar challenges and pain points. With their support, your organisation will stand a better chance of achieving a measurable and compelling return on its investment in AI.
Setting your business up for a stronger future with AI
AI is a gamechanger for businesses that recognise it for the tool that it is and deploy it in ways that enable them to achieve genuine gains. The challenge is no longer whether AI works. The challenge is whether organisations can embed AI into business processes, data, governance, better customer experiences and ways of working to realise measurable value.
Understanding and avoiding the most common pitfalls will help your organisation enjoy great results and a competitive edge. If that’s your desired end, it’s worth investing time and resources in the means that will help you achieve it.
Andrew Carlton Principal Consultant & CX Practice Director, Customer Science