Q&A: Richard Jones, Proxima et al
With decades of experience right across the space and with a record of huge success behind him, Richard Jones is one of the most recognisable and respected figures in outsourcing. Currently chairman of procurement outsourcers Proxima and a board director of robotics radicals Blue Prism, he is involved at various levels with numerous other ventures and retains a unique perspective on outsourcing and the future of business. We caught up with Richard earlier this year for a series of interviews: here are some edited extracts of those conversations…
Outsource: Richard, a lot of the work you’re doing is with organisations which are at the cutting edge of automation technology in one way or another. Why is automation so important right now?
Richard Jones: It’s very important for outsourcing business models to start to push out the processing into technology – if you look at the difference in valuations between these body shop-type models and genuine big process models you’ll see much higher multiples. If you operate an outsourcing business in any market where you’ve got a very high staff turnover, it gives you immense problems as far as the cost is concerned. It’s very costly to run people in those first years of employment if they’re not 100 per cent productive. There’s also the threat of your IP walking out the door every time someone leaves – and in some markets they often go and join your competitor. So, it’s really important that outsourcing businesses that think they’ve got a long-term future start to push process and lower-value activity into technology engines. It gives you a much stronger position and probably a much higher quality of service. Arguably all outsourcing businesses, no matter what point of the market they’re in, have to tackle that. You’ve seen the way that outsourcing businesses can grow: it’s perfectly feasible to start in Poland or a local region somewhere and there’s such competition for talent that you exhaust the local labour pool. So you have to start dealing with pushing work into the technology, really quite early.
O: Do you think that’s something which the majors have got their heads round yet?
RJ: Well generally speaking I think that they’re still on – not the old approach but the prevailing approach, which is more and more automation of existing systems. At Proxima right now for example we’re looking at the layer which goes just above – what I call the hypervisor layer. We can use multiple tools and make it look to the user like it’s one process. Of course that gives you the ability to tailor by client, and if the clients want different underlying engines, on a P2P engine or something like that, you can still make it look the same to operate. So that’s very, very important to the model.
Most of those are about making your technology and your process go faster – but the Blue Prism robotic-based approach is to come in and straightforwardly identify opportunities to put virtual people where you’ve got physical people. That starts to provide you with the same kind of IP protection and pushing knowledge into technology and so on. It yields a prize in a different way. So many people have put in huge ERPs and then turned round and said “let’s shed the people now” but then they can’t find out how to shed the people. Whereas the Blue Prism approach – which I think is far more the next generation model for outsourcing – is very, very laser-focussed on providing another type of labour.
O: As providers move up the value chain, and as the capabilities that they have get more and more complex, presumably a similar thing’s going to happen with activity that currently is considered outside the scope of much automation? Is there going to be a point where you could have even the highest value work taken out of the human loop altogether?
RJ: I think the first part of the answer is, it’ll depend on which outsource tower you’re dealing with. In a Proxima case there’s a very high requirement for face-to-face, stakeholding negotiations. For the moment I can’t see that that’s ever going to be done in any other way. We might see more teleconferencing for instance, but it basically needs a human interaction.
If you take an outsource model out of Western Europe or North America it’s probably mostly processes, and they’re probably not operating brilliantly. And roughly speaking two thirds of the cost will be people and one third would be the infrastructure and systems. When you put the same work into a major offshoring location – India, Indonesia or whatever – basically, that ratio inverts, so I’d say two thirds of the cost of delivery becomes infrastructure and systems and one third becomes people. So the key thing here is that you’re paying X million dollars or whatever it is for your facility or your computer systems. Actually it becomes extremely attractive then, regardless of the number of people in the model. You have the opportunity to look at your building and ask how much revenue are you going to crash through there in a year. Because what you’re trying to do, if you could double the rate, then my proposition that two thirds of your cost is going to be infrastructure and buildings starts to radically reduce. So you know when you outsource work overseas it becomes an asset utilisation issue. When we’re in North America and Western Europe, it’s a people utilisation issue. A lot of companies don’t spot that, and you need a different sort of metric when you go and do it somewhere else. Hence you’ll go into an Indian or Thai centre and you’ll have people employed pressing lift buttons or making tea or whatever; but you’ll go and walk around the floor and to Western eyes you’ll see for example that the square footage per desk is much more constrained. That’s because the guys know that it’s about asset utilisation: you’ve got to get more people in the building and more work, more revenues through the building in a year.
O: How much is the drastic shift in living costs and relative incomes in various parts of the world impacting on that dynamic?
RJ: I think it’s definitely getting there in some towers of outsource in India now. It’s quite feasible that very high-end Indian-based ITO professionals cost the same as a North American might. It doesn’t take that many years in the Indian context of 12 per cent per annum rate inflation to start to make a nasty mess of your economic model. The outsourcing industry’s solution is a race to the bottom, just getting slicker and slicker at moving work between countries. Just find cheaper and cheaper countries. I don’t think that’s the answer.
This angle is very interesting: what’s happening in the currency markets? We’re talking about wage inflation being out of your control: way beyond that is what’s happening in the currencies. This can be very disruptive in your model and your planning and it would be great if outsourcing companies had a strong stream of revenue that they could renew every six months or something on short-term projects – which some of them do – but the bulk of what you’ve got is committed to supply for a long period of time. Five years, seven years, sometimes more. And in which case you’re always going to keep being beholden to currencies. You must have mitigation plans.
If you were speaking to a corporate with its own captive in India right now it would be massively topical. In the outsourcing sense you’ve got your invoices in one currency and your costs in another. It doesn’t take that much of a movement to make a very interesting dent in a relatively narrow margin industry. (Or a very interesting enhancement – it depends which way it goes.)
O: Well that further enhances the appeal of automation, of course…
RJ: Virtualising workloads: the more you can virtualise workloads, the closer you get to the point of saying “Mr Client: where do you want the technology?” The more you can do with that, the more of that revenue protection role you’ll get from your currency rate protection buffer. My argument would be that mostly the reality is you’ll never get 100 per cent protection. But you can move over time and it is quite nice if someone’s given you 500 FTEs of work or 800 FTEs of work. I think it’s much easier to do it frankly starting with “I’ve got 500 FTEs, what am I going to do about it?” than starting at the other end of the scale saying “I’ve got ten, I’d like to grow it to 500: what am I going to do?” That’s a lot harder.
I would say there is going to be one set of interesting lines of thinking for Western companies. The very interesting one is what’s being done in the Indian models. They’re trying to build up the basics faster but maybe what you’d see is them starting to diversify the cost base – and we have seen, actually, the big end, the top five going out and about trying to diversify the cost base. You’ll see Central and Eastern Europe being a focus. You’ll see some ventures in Costa Rica and so on, as they try to spread their risk by spreading the geography. But people have been doing it for a decade; they’re so good at moving work from place A to place B – but of course that’s one of the core values of the business perhaps.
O: Let’s look at it from another angle, at a Software-as-a-Service model. We’re talking about virtualisation but increasingly buyers are looking to have cloud-hosted stuff that they can just access through their own devices – so there’s another driver there for change amongst the major players, isn’t there? In that new entrants to the market might be able to offer cloud-based services (and crucially utility pricing) without being hampered by huge headcounts…
RJ: Well I’ve always felt with that vision of great big computers in the sky, the place that it would be fabulous is for the middle market-based economies – so the UK and Germany. Because really we have penetrated the corporate enterprise space for outsource services very happily. A great Proxima study a few months ago showed what’s been happening: corporate vitualisation at the top end of FTSE. Where it hasn’t moved in the UK is the great big middle-sized area, and to do that the problem has always been the standard platform of its type. I think the supply side needs to get really knockout solutions.
On the supply side, the challenge we’ll suffer soon is around the licencing model. Where we’ve seen big breakthroughs in business processing in the past, it’s tended to be a bit more licence-based. Which enables you generally to put more than twenty per cent of your revenue into building new product parts besides anything else. Just as you want more and more sophistication, actually there’s less and less money coming in through the revenue model; it’s very tail end-loaded relative to what it used to be. So there’s going to be a sweet spot there. Right now I’m very excited about some of the things in that space I’ve been seeing in the last twelve months – certainly in HR services. Suddenly I’m seeing what I had thought was a dream! We’re starting to see numerous highly customisable cloud-based solutions starting to arrive now. In F&A we’re also starting to see that.
O: You do get the sense that there is a huge untapped long tail in the mid-market space – but most big suppliers don’t seem to care because they’re doing too well in the enterprise layer. Do you think this will change?
RJ: The intriguing thing is that most breakthrough software in the world was written for a big enterprise client at the outset. They funded development because it’s a breakthrough story, they wanted to get there first. That story is a standard story in the software industry. So if we’re going to see highly customisable cloud-based platforms that enable penetration of the middle market because they’re so customisable, it is quite likely that it will come downwards from the enterprise sector.
There are plenty of very large corporates who have large geographic tails which cause them to want to have the systems bought for them. Some multinationals are operating in well over 100 countries and yet they are looking for software and BPM-type solutions that are consistent through the whole lot. So of course they’re innovating product. I think that is a reasonable area for incubation for these kind of requirements to come through.
O: Do you see more acquisitions by majors of start-ups which have developed technology suited for the mid-market?
RJ: Yes, I would, partly because the majors have huge cash piles and they’ve got to apply the cash – and in a lot of cases very nice stuff happens. I was looking at one in the procurement space the other day. A tiny amount of revenue – five million or something – and a multiple valuation on AIM of about 80. Something is obviously running really hot and people are skulking around looking for ready-built engines that can be plugged in. I wouldn’t be the expert on those but one of my projects is a delightful 25-person company on seven or eight million revenue. Blowing all the doors off. They think the end time for them is going to be somebody big comes along and buys the business; it might just be a straightforward asset sale and off we go. In this market there are a lot of people innovating – not that they will all get through, but the incubation ground if you like is very good, is very strong. The UK’s got a very strong reputation in the incubation of software.
O: Let’s shift focus a little: we’ve discussed the issue around currencies but what other issues are there that could, if you like, derail the kinds of trains that you’re talking about?
RJ: Well, you start with all the things you can’t control…. I think it’s very likely that what’s going to happen in North America is legislative changes. We’ve had that ripple of fuss about these: that won’t be the first step. If the economy’s in distress, most of the European economies and North America will start to find ways to make changes. And if I’m sitting in Indonesia with my own company I won’t be able to do anything about it.
The very smart guys have been working on it at least a decade. Starting to experiment with models which use, for example, a local region. And learning how to build the process models so they financially perform and give you what you want and to be a blend of, for example, Northern Ireland and somewhere in India for a lot of their clients. So I think we’ll see a bit more of that, just as a protectionist move against protectionist legislation,; some things weren’t possible before but maybe the technology wasn’t there to enable you to control your processes and drop process layers in different countries. And then there’s macroeconomics. I’ve always had the view of that outsourcing is counter-cyclical – so the harder life gets in the economy the more corporate virtualisation, the more outsourcing normally goes on. I accept completely that the outsourcer has still got to make money out of it. But I think the market demand to lay off non-core work increases while your country or your marketplace is under duress.
O: Has that been born out by what’s happened over the last few years?
RJ: That’s a very good question. I’ve probably had some smart-arse answer like “if you can view it over decades you’d see it”. It’s a very interesting question as to whether or not you see it right now. Look at what happened over, say, the last four years with regards to Indian outsourcers. I think the decision-making slowed down massively. And the uncertainties built that up. Now what we see at the moment, I think in the last three quarters, is a huge acceleration again and they’re growing faster than the economies that we’re talking about. By miles. We’re back to seeing double-digit, sometimes twentysomething growth announcements. So my feeling is they’re away again.
This article originally appeared in Outsource magazine Issue #34 Winter 2013.