NewsBite

Sponsored by Salesforce

When it comes to AI: don’t DIY, start with why

Subscribe to gift this article

Gift 5 articles to anyone you choose each month when you subscribe.

Subscribe now

Already a subscriber?

The majority of the 70+ Australian CEOs I’ve spoken to this year want to use AI to transform their business. They want AI to improve customer experiences and make their teams more productive.

What they’re struggling with is moving from experimentation to tangible impact.

Frank Fillmann, general manager, Australia & New Zealand, Salesforce. 

Enter AI agents. AI agents are an upgrade from chatbots and copilots. They take action instead of waiting for instructions. They unlock productivity gains and free employees to work on higher value tasks.

Agentic AI will change the way we get work done, how we engage and build relationships with employees, customers and partners. Even the way we structure our workforces.

Some leaders are turning to DIY efforts to integrate AI into their business, and are on a path fraught with traps, potential snafus, and wasted time. We think there’s a better way.

Advertisement

The CEO conundrum

The CEO’s office is often said to be the loneliest office. The board and investors want them to hurry up, adopt AI and deliver value.

Technical leaders are feeling the pressure to be AI experts and deliver returns on AI investments so they’re delegating up to the CEO.

Customers and employees want them to inject AI into their experiences. Customer expectations are growing. Their last great experience is their new expectation, and they don’t want to wait.

A whopping 41 per cent of employee time is spent on repetitive low-value tasks, leading to burnout and attrition. To make matters worse, employees are stuck with siloed data and AI that doesn’t live up to its promises. It’s no surprise they’re turning to consumer-grade AI, creating new security risks.

All of this pressure creates a perfect storm for CEOs. The urgency to make decisions and take action, with a lack of certainty and even some understandable fears.

Charting a path forward

Our research with YouGov found the key drivers for Australian C-suite execs to integrate AI were boosting productivity and efficiency, innovating customer and employee experiences, and remaining competitive.

As businesses have raced to get going on AI, they’ve put too much focus on the how, and not nearly enough on the why.

Every business needs an AI strategy, which must start with the value that’s trapped and the use case to un-trap it. The focus should be on areas of trapped value, the functions and tasks where AI can have the greatest impact on the business. It should start with the use cases and work backward, with a focus on speed to value.

Leaders should avoid getting stuck on the technology. It can be tempting to obsess over which UI and AI model to choose. We take an agnostic view on this, because what really matters is connecting a company’s trusted customer data safely, and that AI is used where it’s most needed.

Data is the foundation

Many businesses tell us their data is fragmented, trapped in disconnected systems, and not ready to ensure AI investments are truly effective.

A recent Deloitte survey of 2000 leaders found almost 80 per cent of leaders anticipated increasing their investments in data management as they see it as core for AI to be effective. It’s critical to unlock the rich organisational intelligence trapped in silos, and making it available to back office and front line workers alike.

Data Cloud helps solve this critical problem and has become our fastest growing product. It addresses data fragmentation, and brings trusted data to agents through a technical concept called retrieval augmented generation, which allows us to teach large language models about specific business needs without retraining them from scratch.

Why the biggest disruption will occur on the front line

McKinsey recently estimated 75 per cent of the value of generative AI will come from front line functions, including customer operations, marketing and sales.

For most businesses, demand far exceeds their capacity to serve customers. With AI agents, there’s no capacity constraint; you can build and deploy quickly, and they work 24/7 for your employees and customers. This allows businesses to scale quickly and frees employees to solve problems critical to the business.

We are seeing this play out already with our trailblazing customers. Fisher & Paykel uses agents to resolve 30 per cent of queries, freeing up customer service operators to focus on more complex cases, resulting in estimated savings of 3300 hours per month.

The future is now

AI is finally living up to its potential. It’s helping companies improve productivity and efficiency while also delivering a better experience for employees and customers.

The fastest, safest, and most sustainable path is to abandon DIY efforts that start with the technology, leverage the investments already in place, identify the most critical, high-impact use cases, and deploy AI agents quickly and easily.

We believe this is what AI was meant to be, do you?

Frank Fillmann is general manager, Australia & New Zealand, Salesforce.

Sponsored by Salesforce

Subscribe to gift this article

Gift 5 articles to anyone you choose each month when you subscribe.

Subscribe now

Already a subscriber?

Read More

Latest In Technology

Fetching latest articles

Most Viewed In Technology

    Original URL: https://www.afr.com/technology/when-it-comes-to-ai-don-t-diy-start-with-why-20241111-p5kpjl