Published on the 16/07/2025 | Written by Heather Wright

Neha Ralhan wants local organisations to do more premortems when it comes to ERP systems and other enterprise applications.

Ralhan, Gartner senior principal analyst for application strategy, told iStart the current challenging environment means many ERP and enterprise application leaders don’t have a mandate to do widescale change within their organisation, but still need to be making tangible changes to move the organisation forward towards the organisation’s overarching strategy, while also dipping their toes in the innovation around generative AI. (While Ralhan won’t say it outright, those ‘tangible changes’ are also likely to include dealing with vendor deadlines, including end of mainstream maintenance for one large ERP vendor in particular.)

“This isn’t just about keeping the lights on, it is also about configuring your organisation to be in a better position.”

At the same time there is less tolerance, within both commercial and public sector organisations for ‘failure’ or not hitting key metrics asap.

“The research is saying what can you do, what is your minimal viable product when it comes to enterprise applications, but also, whether an MVP or a larger transformation, how can we absolutely ensure there is no failure, there’s no excess and we’re set up for success from the get-go?

“The tolerance that was previously a bit more built in, has evaporated in the last year or so.”

Finding that balance, however, requires a changed mindset.

She says while application leaders typically focus on how to position their application portfolios for success, a less risky approach is to focus on how to avoid failure – something that can be done via a premortem.

The management technique, devised by Gary Klein in 2007, takes a deep dive into projects before they kick off. Unlike a critiquing session which looks at what might go wrong, the premortem, Klein says, effectively assumes the ‘patient’ has died and so asks what did go wrong, with team members tasked with generating plausible reasons for the project’s failure. Participants envisage failure at a set date in the future and work backwards to identify root causes, brainstorm solutions and then formalise those solutions into an action plan.

“A premortem for enterprise application portfolios doesn’t shy away from failure; instead it actively confronts and engages with it, revealing both potential risks and solutions that might otherwise be overlooked,” Gartner’s Failure *Is* an Option: Prevent Application Disaster with a Premortem report says.

It’s a process that requires a mindset change, moving from pursuing success to identifying potential failures to reveal hidden risks, Ralhan says, and one that also requires involvement from a diverse range of stakeholders, with failures in enterprise applications often stemming from a combination of cascading factors, some of which originate outside the typical domain of the IT team.

“This is not just the enterprise application leader’s remit, this is more collaborative,” she says.

“Sometimes when we talk about application failure it is quite black and white – meeting go live, coming in under budget, meeting whatever the metrics might be. But if you’re an end-user at a manufacturing plant, what you require from your application is quite different from what someone in finance at head office is focusing on.”

She’s keen for Australian and New Zealand businesses to expand their idea of what success looks like when it comes to enterprise apps, adding additional metrics such as around user experience, particularly for ERP, or how well individuals are buying into the vision and, crucially, whether applications actually match the organisation’s strategic goals.

Gartner research last year showed almost 75 percent of respondents said their ERP strategy did not strongly align to the organisation’s business strategy.

“With this latest research, we are focusing on a more considered approach and it honestly matches up with the sentiment in the market at the moment,” Ralhan says.

“Organisations and enterprise application leaders who do this are going to be better prepared for all the change that is happening with AI agents and generative AI. This isn’t just about keeping the lights on, it is also about configuring your organisation to be in a better position.”

Speaking of AI, Gartner has also released its ERP Hype Cycle for 2025, which includes a strong AI focus.

Three key trends emerged this year: Future of ERP which is some of the transformation things such as adaptive experience and headless ERP, AI tools and capabilities, and ERP fundamentals, including sovereign cloud, sustainability in ERP and cloud ERP for the public sector.

“There is a clear theme of AI tools and capabilities, highlighting the importance of these tools for ERP moving forward.”

Local organisations are not, however, actively seeking AI in ERP at this stage, with Ralhan saying at the moment the focus is on adding on to the ERP with AI tools and capabilities, rather than an organisational wide AI strategy linked to enterprise applications.

She notes that the AI tools and capabilities are all at different points in the hype cycle.

While headless ERP and adaptive experience are at the ‘innovation trigger’ (aka very early) stage, sovereign cloud and embedded AI in ERP have reached the peak of expectations and hybrid integration is moving up the slope of enlightenment.

Generative AI too, has reached the peak of inflated expectations. Ralhan says it’s the first of several peaks expected for genAI.

“It highlights that you have to be playing in the pool. AI is not going away, it is really cementing and the uptick in AI tools and capabilities is fundamentally changing ERP at its core and foundation.”

Utilising a headless ERP model, together with decision intelligence and hybrid integration for ERP will enhance the ability of the ERP to align with greater business value creation. Generative AI in ERP, agentic AI for ERP and embedded AI in ERP, meanwhile, are central to an adaptive experience in ERP. And IT leaders can develop robust methodologies and frameworks to govern AI practices via digital twins.

“The challenges are different to what organisations had previously, but there are lots of opportunities.”