Can your company grow in 2020 and beyond without scaling artificial intelligence (AI)?

I suppose it's possible, though I wouldn't recommend it.

To understand why AI at scale is crucial to growth today, consider where we were just a decade ago. In the pre-AI scaling era, I worked with companies that took three to five years to implement initiatives. At a time when business cycles spanned seven to 11 years, it was fast enough.

<<< Start >>>

75% of business leaders feel that they will be out of business in five years if they can't figure out how to scale AI.

<<< End >>>

But few of the companies I work with today have the luxury of that much time. The modern business cycle can be as short as two to three years. Relying on traditional business processes and protocols to implement change means that by the time you roll out something new, another business cycle has begun, leaving your company behind the curve.

Many are already painfully aware of this: 75% of business leaders feel that they will be out of business in five years if they can't figure out how to scale AI, yet 76% also admit that they don't know how to do it.

The good news is, they can learn. I've worked with many companies to make sure they're scaling AI the right way, right out of the gate.

Let's take a look at three smart AI scaling strategies and how companies I’ve worked with embraced them:

1. Create critical mass with intentional AI

Imagine that the C-suite at your company, not just the IT department, is focused on using AI to drive important business and strategic objectives. Executives are intimately involved in championing the adoption of AI and are explaining how it aligns with business goals so that everyone at the company understands AI's importance — so much so that it becomes part of company culture. That’s what intentional AI means.

In practice: How AI helps a life sciences company speed drug development

I've worked with such leaders recently at a life sciences company, led by a CEO who made it clear that a digital transformation was afoot. He stressed that they were going to use data and analytics very purposefully to accelerate their journey. Among their goals? Speeding up the onboarding of partners in developing medications.

It used to take up to nine months to set up pipelines for partners to share clinical data and other information. We made a plan to change that. The company would establish an intelligent data foundation and create an application program interface (API) gateway. They pre-programmed common algorithms for onboarding and offered that as a product through their API. They were able to move quickly to scale the AI initiative, largely thanks to the leadership's advocacy.  It was a thoughtful, intentional application of AI towards a core element of their business strategy... and it's working.

After two years, onboarding times shrunk significantly in proof of concept cases, while sharing data happened more effectively. This spurred the development of new drug therapies and revenue streams. The company's digital transformation is far from complete, but the early results are very promising.

2. Tune out data "noise" to capture the right information

A powerful benefit of AI-based tools is how they help your business derive insights from large volumes of data. But not all data is created equal. If you're overloaded with data, it's hard to isolate the information that matters. You may be caught in a state of paralysis, unsure how to move forward.

In the real world: How an energy firm drilled down on production forecast data

My team recently began working with an oil and gas firm whose challenge came from the ground up, literally: they wanted to forecast production declines in aging oil wells. That would allow them to plan the reinvestment required to return production to original levels — a key advantage at a time when oil prices are low and energy companies aren't exactly flush with cash.

But when it came to finding data to base forecasts on, the company didn't know where to begin. They had hundreds of options — including seismic, geological, and flow data — from multiple sources, such as equipment sensors and third-party providers.

<<< Start >>>

If you're overloaded with data, it's hard to isolate the information that matters.

<<< End >>>

So we started small: We picked a section of one oil basin and examined some 200 data fields. From there, we narrowed it down to roughly 50 critical data elements and designed rules and algorithms to manage that information and test data quality. We also put tools in place to trace the data so we knew where it was coming from and who may have transformed it. This resulted in high-quality data sets that will guide the company in deploying capital efficiently, helping it withstand economic pressures for years to come.

3. Make AI a team sport

Earlier I mentioned how important it is for your company's C-suite to champion AI efforts. But successfully designing and implementing an AI-driven initiative will likely require a multi-disciplinary team.

Developing the technology takes machine learning engineers, visualization experts and design thinkers. Then, deploying the technology requires engaging an even greater set of stakeholders, including end business users who will work with the new AI tools daily. Bringing employees throughout the business on board encourages widespread adoption of the tools and ultimately helps embed AI into company culture (which, as I noted earlier, is an important facet of intentional AI!)

In the real world: How a convenience store chain forged broad support for big changes

I worked with this global company on an AI solution that would help them determine, on an ongoing basis, which merchandise and price points would entice more customers to enter stores and fill their baskets on any given day. But ensuring stores had the right inventory in stock entailed working with the company's supply chain and warehousing operations.

Meanwhile, transforming pricing and store shelf-space allocation meant convincing the chain's franchise owners and regional managers of the merits of the AI-powered model. It was a matter of getting folks into the tent from day one, or it wasn't going to succeed.

It did succeed. Though it took significant effort to ensure support across the organization, the results were worth the work: pricing that yielded higher margins in critical categories and a large increase in foot traffic and customer loyalty.

Preparing your company for the future

Success stories like these are happening every day, and yours can be one of them. With commitment and advocacy from your company's leadership, a team-oriented approach that draws support and active collaboration throughout your organization, and tools and strategies that help you gain critical insights, you can scale AI effectively.

In doing so, you'll build a foundation for your company to thrive (and become a truly intelligent enterprise), no matter what challenges the next business cycle may bring.

Joe Depa

Global Lead, Data-led Transformation

Subscribe to Accenture's Intelligent Functions Blog Subscribe to Accenture's Intelligent Functions Blog