Dr Khan says PepsiCo uses modelling similar to that seen in aviation and energy sectors

Dr Khan says PepsiCo uses modelling similar to that seen in aviation and energy sectors

In the second part of a two-part interview, Dr Mehmood Khan, chief scientific officer and head of global research and development at PepsiCo, gives Olly Wehring, editor of just-food sister site just-drinks, a tour of the inner workings of the food and beverage group's R&D approach and defines success and failure for his department

The importance of R&D to PepsiCo continues to grow, according to Dr Mehmood Khan, chief scientific officer and head of global research and development at the US food and drinks giant.

"In fiscal 2013," notes Dr Khan, "9% of group revenue came from innovation. In 2012, it was 8%." Prior to his arrival in 2007, however, the department looked very different to its current guise.

"When I joined the company," he says, "we had a lot of talent on both the food and the beverage side. However, things were not cross-pollinating; you developed your career in one part of the company or the other. That really limits the leveraging of your portfolio when it comes to cross-fertilisation of ideas."

Another major difference between then and now is the genetic make-up of the division. "At the time," Khan continues, "100% of my executive leadership team were American. They were also all men, and they all had the same background, in food engineering and food technology. Fast forward seven years, and half of my direct reports are women, half are non-American and their backgrounds are completely diverse – from beauty care and life sciences as well as from traditional food.

"Now, we have a much broader richness of background, culture and geography that pays off in buckets. This talent has truly transformed R&D in PepsiCo."

The role of technology has also revolutionised PepsiCo's R&D in recent years, with computational modelling speeding up the process. "Instead of working on a traditional research bench, today we can use computers to simulate research for us: That's how powerful they have become," says Khan.

"Instead of building and then testing new equipment that makes our products, which takes time and is expensive, we can use computational power to model our processes. That's how it's done in the aviation and energy sectors, for example. We've started to bring a lot of that know-how into our sectors. Traditionally, we haven't had that kind of expertise within the company."

Recent PepsiCo headlines have explored the possibility of splitting its food and beverage operations. Much as CEO Indra Nooyi has tried to calm this speculation, it doesn't seem to be going away. I put it to Dr Khan that a stronger focus on one particular market segment would benefit his team's efforts.

"The challenge is how to create the matrix for integration," he concurs. "However, that's relatively minor to the upside of the opportunity. We use the term 'power of one'. The business opportunities that we can unlock by leveraging across our brands is phenomenal. Our flavour technology that was developed on the CSD side was first used on a Quaker Foods product, for example. The return on investment when leveraging across multiple brands is much better and, because we're pooling resources we can make more longer-term investments. If we're just focusing on part of the portfolio, we're either not going to have the resources or the depth in expertise or the investment capability."

The largest hurdle the soft drinks industry is struggling to clear right now is one made out of sugar. Around the world, health organisations are calling for food and drinks companies to do more to combat rising obesity among consumers. With PepsiCo on the front line in this battle, Dr Khan's team also has a part to play. "We've put a lot of know-how into being able to create low-sugar launches," he says. "It's very much part of the R&D mission and we'll continue that."

We've not got much longer left, but I'm curious to know what 'success' and 'failure' look like to Khan. In R&D, presumably, is the laboratory littered with failures? "I'd have a voluminous CV if I filled it with the R&D projects that have failed," he says. "But, you can look at this in one of two ways; either they're failed projects or they're learning opportunities.

"If projects aren't failing, then we're not doing our research. We're in the business of research. I look for planned failure – it's not that we have predicted that something will fail,  but we plan that some of our innovations will fail. It's only a failure if we don't learn from it."

What of a definition for success, then? "The overall success rate is drawn from the top and bottom line as well as from our operational efficiency," he explains. "On the research side, the failure rate is higher early on because it's unknown territory. We've had multiple projects in new flavour systems and sweetener technology, for example. You can't put all your eggs in one basket. We've had plenty of failures along the way and plenty of successes. Of the failures, we've learnt from them and incorporated them in the final delivery. Early in the funnel, there's a much higher failure rate, while at the tip it should be down to almost zero."

My final question is, I confess, one I have already worked out the answer to during our time together.

Do you enjoy your job? "About 1.3bn people a day consume a PepsiCo product," Dr Khan says. "I can't think of a job in the world where you can make a difference to that many people with the work that your team does. It's an amazing responsibility and a daunting task, but it's what keeps me going.

“At the end of the day, I'm a scientist, a physician and a businessman – I always have a degree of impatience. I want things to happen now, but it's critical to the sustained success of the company that we nudge people along. It's a balance between getting things done optimally and bringing everybody along. You cannot do that overnight – it's a journey.

"Part of being a leader is that you always want more and you always want faster."

That's a long 'yes', then.