A checklist for the first 60 days of your AI journey

Pavan Bedadala at Commvault cuts through the hype about artificial intelligence and explains how organisations can embark on their AI journey with confidence

 

AI has dominated the headlines, trends and opinion articles for the past 18 months and every business leader wants to know how they can use it to automate processes, dazzle customers and drive better margins. But how do they kick start this journey of transformation and put the plans in place to utilise AI in the most effective way possible? 

 

Simply put, AI implementation is a marathon not a sprint. While there is something to be said for ‘first mover advantage’, being the first to cross the start line doesn’t guarantee that you’ll win the race. AI is also a technology that’s still in its infancy and investing in the wrong areas without first conducting the necessary due diligence can risk wasting significant amounts of money, or even worse, create chaos that ends up doing more harm than good. 

 

For these reasons, when it comes to launching an effective, long-term AI strategy that can support the business for the next decade or more, the first 60 days are often the most important of all.

 

This is the period when senior leaders from across the business must pinpoint the exact challenges or issues that they are looking to AI to solve. It’s also when plans need to be put in place to properly democratise data access, so employees can make best use of any new technology without compromising IT security in any way. Finally, it is the time when the foundational processes of any new AI implementation must be specified and tested, which will enable the technology to be effectively scaled over time. 

 

Like so many things in life, taking the time to prepare properly at the start of the journey will pay major dividends further down the line. Setting the right goals, creating the proper data strategy and putting the right oversight structure in place from the very beginning lets companies set out on their AI journey with confidence, instead of apprehension. 

 

Set tangible goals from the outset

Unfortunately, in the current AI goldrush, pressure to embrace the technology as quickly as possible (whether from executive teams, shareholders, or elsewhere) is causing some businesses to completely bypass the goal-setting phase in favour of chasing quick wins. 

 

However, goal setting and strategic planning plays a critical role in overall success. This is because a successful AI implementation is about finding solutions to existing problems. Deploying applications that don’t directly address current issues is often just a waste of time and money. 

 

A great place to start is identifying tangible bottlenecks in current operations or pinpointing lower-value tasks that are dominating employee time and considering how AI can eliminate or streamline these processes. That way, when the board demand an update on their investment, leaders can easily demonstrate how the technology is solving real-world challenges, rather than being a costly outlay still looking for a problem to solve.  

 

Set the right data management strategy

Most modern companies have been gathering data for decades, in ever-growing volumes. A lot of the time, top executives that aren’t involved with IT day-to-day think these raw data sets can immediately be used by AI models to start deriving high quality data-driven insights about the future of the business. Unfortunately, given the data disparity and siloed systems of many enterprises, that’s rarely the case. 

 

To make sure AI models are running properly and generating accurate insights, companies must first take time to set the right data management strategies for their operations. To do this, they need to conduct a comprehensive audit of all their existing data sources and decide how best to bring them all together. Next, they must ensure that all the data being fed into the AI model is of high quality, meaning it is clean, accurate, and properly formatted. Finally, they must ensure that employees only have access to the data they have permission for.

 

Once these steps are completed, businesses can begin training their AI. While they are unlikely to work perfectly right out of the gate, with time and feedback, they can become highly accurate in a surprisingly short amount of time. 

 

Continually assess and adjust as necessary

AI is still in its infancy as a technology, so businesses need to ensure they have a process of continual assessment in place to make sure things are working as intended. This way, if any issues arise with data quality or inaccurate results then they can be quickly addressed before they lead to bigger issues. 

 

As AI learns over time, the accuracy of results should increase significantly, meaning businesses can have real confidence in the data provided. However, until that point is reached, it’s prudent to maintain a watchful eye and adjust as necessary. 

 

We’re only just scratching the surface when it comes to the potential of AI, but in the rush for instant results, too many businesses are failing to lay the correct foundations for success. Taking the time to create the right strategy at the outset and preparing the data and personnel accordingly will let businesses reap the rewards for many years to come. Conversely, skipping this crucial step can quickly have the opposite effect. Ignore it at your peril!

 

---

 

Pavan Bedadala is Senior Director, Product Management at Commvault

 

Main image courtesy of iStockPhoto.com and Midnight Studio