Emmanuel Frenehard, chief digital officer (CDO) at biopharmaceutical giant Sanofi, recognises that the CDO role often means different things in different companies.
At Sanofi, it was decided that the role would be an all-encompassing position, overseeing business applications, infrastructure, cyber security, data, artificial intelligence (AI) and digital services. There are also professionals in Frenehard’s team who cover research and development, manufacturing and commercial activities.
“But there’s a common thread – and that’s our intention to save time for patients,” he says.
“We’ve set ourselves this ambition – which is not mathematically tracked – to say, can we halve the time of drug development? Because it takes 10 to 12 years to create a new drug and take it to patients, and 90% of drugs fail. What if, through the power of data, digital, AI and software development, you could reduce that time to five to six years?”
Frenehard describes his role as humbling. Drug development failure rates can be as high as 90%, which has important commercial ramifications, but also, more importantly, health implications for patients. He says it’s a privilege to lead the team that works to find potentially life-changing answers to challenging scientific questions.
“It’s a job where you feel like you have a lot of responsibilities, because you certainly don’t want to fail, and there are difficulties because we work in a slow-moving industry,” he says.
“If you’re going to do an experiment on finding new biomarkers for patients, you’re going to need a certain amount of patient data that might include biopsies and blood samples, and it might take a couple of years to validate the hypothesis, so the cycles are very lengthy.”
Embracing the opportunity
Frenehard has been Sanofi CDO since September 2023, having originally joined the organisation in 2020. He previously worked at media and entertainment specialists iFlix and Walt Disney.
“I thought media was great and I believe in the power of entertainment,” he says. “We all live difficult lives, and I felt like there’s an escapism that you have when you watch a movie or sports. There’s a moment where all your worries are gone, and so I always thought delivering that content was a noble mission.”
“We’ve set ourselves this ambition [to] halve the time of drug development. Because it takes 10 to 12 years to create a new drug and take it to patients, and 90% of drugs fail. What if, through the power of data, digital, AI and software development, you could reduce that time to five to six years?”
Emmanuel Frenehard, Sanofi
However, when Frenehard was contacted about the role at Sanofi, he recognised that the chance to improve patient life opportunities was an even nobler mission.
“It was something new. I wanted to see whether I could be effective in a fresh industry. I think it’s good to test yourself,” he says.
“So, that was the pull, and you quickly realise that the sense of urgency is very strong. And that’s where all the tools and all the things that I was able to learn when I was in media, such as prototyping, failing fast, doing minimal viable products, and agile ways of working, became pertinent in this industry.”
Three-and-a-half years into the CDO role, Frenehard points to achievements in two areas – people and technology. Regarding the human component, he says bringing a team together comprising professionals from multiple disciplines is a complex but necessary task.
“We came with this idea of saying that everybody can be stronger together when they work across one digital organisation,” he says.
“Creating that sense of teamwork has been a big accomplishment for me, because we didn’t have that before. It’s all about the notion of one team rowing in the same direction. This team might have different convictions and points of view, but they pull together for the benefit of moving in one common direction.”
Leading digital transformation
Sanofi was founded in 1973, although the formation came from the combination of a diverse group of companies dating back to the 19th century. In the contemporary era, Frenehard says the company has continued to grow, with the firm today employing more than 83,000 people across 70 countries.
“I like to say Sanofi is a company with a heritage of 53 years of acquisitions that were never integrated successfully,” he says, referring to his achievements in technology.
“Now, as an organisation, we’re running on a single data lake – but when I started this job, we had what I call data dumps. Everybody had one, and you started thinking, ‘How can I even get insights from this data?’”
We wanted to be able to take advantage of AI, whether it be predictive, prescriptive, and now generative, because we knew that if we took the linear path that our peers have taken, it would take us a long time. We had to find ways to leapfrog our competitors Emmanuel Frenehard, Sanofi
Frenehard says data must be treated as a continuum in the biopharmaceutical industry, through which unstructured or unorganised data is turned into meaningful insights.
“That’s tricky when the molecules you discover are treated in different data dumps with different naming conventions,” he says.
“Getting that process right was a big element for us, and we wanted to be able to take advantage of AI, whether it be predictive, prescriptive, and now generative, because we knew that if we took the linear path that our peers have taken, it would take us a long time. We had to find ways to leapfrog our competitors.”
Frenehard says cultural change was the key factor to making this leap successfully. His digital team adopted agile working at scale and built accelerators that allowed the organisation to recruit great digital talent, even from high-profile technology firms.
“These professionals had the opportunity to work much like they were in a startup,” he says.
Today, these accelerators employ 150 people who build digital products for the company. Frenehard says he took the lessons from his earlier leadership experiences and focused the firm’s innovations on consumers.
“We became very patient-centric. We have apps for patients that we’ve built to support them in their treatment,” he says.
“We’re not trying to sell them anything – we’re trying to help them in their healing journey. Many of our drugs are for chronic treatments, which means, as a patient, that you’re going to medicate potentially for your entire life. And so, how can we ensure that you do that in a way that you don’t find intrusive? Can a digital companion app help you with that process? As a team, we adjust little things to improve the quality of life of patients.”
Scaling emerging technology
Frenehard accompanied his cultural change programme with a focus on improving underlying technology systems. He says 53 years of marginal rationalisation and many acquisitions meant the company had to deal with a heavy weight of IT systems.
“We’ve moved that legacy,” he says. “We’re now entering the era of AI coding, where we can modernise that legacy at a pace we never expected. So, I’m thankful for the moment that we’re in, because I’ve got a lot of IT baggage to deal with.”
Frenehard says the crucial component supporting this shift from legacy to modern IT services is the firm’s underlying Snowflake data platform. Sanofi has been a Snowflake user for more than five years. Through this technology implementation, the digital team aimed to integrate a fragmented environment and support secure cross-business data sharing.
“We looked at the footprint of data and realised that it was incredibly fragmented,” he says. “We had multiple data technologies. We needed to have insights that cut across the company, and we didn’t have that capability.”
The first stage of the transformation process was building a series of mini data lakes on the Snowflake platform. This stage meant professionals could discover insights more easily. However, Frenehard wanted further progress.
By working with Elementum, one of Snowflake’s technology partners, Sanofi professionals can now develop richer insights by building data-driven workflows that operate directly on the Snowflake AI Data Cloud.
With this connected approach, the firm’s professionals use AI-powered agents to query information held in the mini data lakes. Sanofi is scaling AI across 80 countries, using Snowflake as the data foundation for governance and collaboration. Crucially, the organisation has built an AI-powered Concierge service to help employees exploit data assets.
“Staff no longer interact with a system of record – they’re interacting with an agent,” he says, outlining how 65,000 of the firm’s 80,000-plus employees already use Concierge monthly across a range of business functions and activities.
“All these things are happening, and we’re doing this work for the benefit of our company.”
Building custom workflows
Frenehard says effective use of AI is all about reinvention, not automation. To be successful in the age of agentic AI, companies will need to redesign their work processes. Two years from now, he hopes that the foundations he’s putting in place, particularly via the Concierge service, will have helped professionals in the business to reinvent the way they work.
“By then, I’d love to tell you that we’re not as constrained by the friction of systems of record,” he says, referring to the desire to use AI to query data directly, not via traditional enterprise applications.
“We’ve got about 2,500 systems at Sanofi. That means there are 2,500 integrations – there are 2,500 things that move up and down. As a professional, you move from one system to the other.”
Frenehard gives the example of a marketer who has to work across many interfaces, such as those for content generation and translation.
“Why does that process have to be so complicated?” he asks, before suggesting where Concierge can bring reinvention. “I think the future is an engagement layer – and that’s where Concierge comes in. It’s an abstraction layer to all the complexity that happens behind it.”
This engagement layer means the digital team will provide employees with the right workflows, depending on their roles. If someone changes their job, the company will know their roles and access rights. The same rule will apply if someone changes geography. The key to success, says Frenehard, is that this AI-enabled approach will allow his department to build custom workflows that give people secure access to the data they need.
“It’s not quite the path of custom software, because custom software is hard to write,” he says. “We’re talking about custom workflows, which are much easier to build.”
The Fog ransomware crew has been observed ramping up attack volumes and targeting new, more lucrative verticals in the never-ending search for a pay-out, and…
Mayor of London Sadiq Khan has unveiled a London taskforce on artificial intelligence (AI) and the future of work, during his annual Mansion House speech.…
Table of Contents The rise of inefficiency in coding Tackling code bloat Inefficiencies in testing Greening AI The demands of modern software development mean programmers…
A Chinese-state advanced persistent threat (APT) actor tracked as Storm-0558 hacked into email accounts at multiple government agencies, and was able to lay low for…
