Welcome to our third quarter presentation. Martin is live on video in New York, and I'm sure he'd like to make a few opening comments before I start.

Well. We've got a lengthy presentation. Paul will cover the numbers for the third quarter and for the first 9 months. And then I'll come back and talk a little bit about how we see the industry and how we see the -- our strategic objectives and structural changes in relation to that. So Paul, if you will, just cover the numbers.

I want to just mention Martin, who we have with us. So we can introduce them now?

Well, we got in -- I think, in London, we have Brian Gleason, who heads our [m]PLATFORMs and Xaxis and Triad Operations; Mark Read, who heads Wunderman. And I think we have Jon Cook on the line, who couldn't be with you physically in London or physically here with me in New York, on the line, who heads VML. And they'll be able to talk a little bit more about 2 or 3 of the challenges that we see to the industry. Of course, I've seen in the analysts' comments this morning considerable debate as to whether these are the issues or not or which of them are the key issues. And no doubt, we can pick that up in the Q&A later. So Paul, over to you on the numbers.

Okay. Thank you, Martin. This is our third quarter trading statement for 2017. Okay. So on our third quarter reported revenues, they were up 1.1% GBP at 3.649 billion. Foreign exchange was positive 1.5% in the quarter, which means that on a constant-currency basis, revenues in the third quarter were down 0.4% and on a like-for-like basis were down 2%. Our preferred measure as you know is on a net sales basis, which takes account and removes the effect of pass-through costs and principal digital media buying revenues. So on a net sales basis in the third quarter, growth was 2.4%. Again, currency was positive 1.5%. So on a constant-currency basis, you had net sales growth of 0.9% and on a like-for-like basis, third quarter net sales was down 1.1%. On a 9-month reported revenues were up 8.9% at GBP 11,053,000,000. Foreign exchange was heavily positive to the 9 months at plus 7.8%, leading to constant currency reported -- constant currency revenues up 1.1% for 9 months and down 0.9% on a like-for-like basis to the 9 months. Again on our preferred measure, the year-to-date or 9 months net sales growth was 9.7% reportable. Foreign exchange was positive 8%. So on a constant-currency basis, the growth was 1.7% and on a like-for-like basis, the decline was 0.7%. The 9 months operating margin is up 0.1 margin points on a reported basis. It is flat with last year on constant-currency basis and up 0.1 point on a like-for-like basis. And on the full year basis, it's now targeted to be flat in constant currency for the year. The average constant currency net debt was up GBP 519 million for the first 9 months to just over GBP 5 billion, primarily reflecting continued significant acquisition activity and share buybacks over the last…

Thank you, Paul. I hope you can see me now. We've entitled the second section a changing industry question mark and tried to focus on 3 issues or areas that analysts, investors seem to be focused on. The first is to put it a little bit more crudely, are Google and Facebook eating our lunch, as the change in social media and search created structural change. The second whether consultants are doing, likewise, particularly in the digital area. And the third, whether low-cost, looser monetary policy since Lehman in 2008, whether that's causing significant changes to the industry as well. So just turning to the first slide, which is Slide 35. It's difficult to come to conclusions about what may or may not be happening. But if you examine our top 10 media owner destinations, that's -- of our current billings of around GBP 70 billion, of which media is about GBP 50 billion, where is the money going? Where is the investment going? We call it media and investment management. And if you look at the top 10, Google and Facebook certainly, last year and this year, probably, will be in positions 1 and 3. I say probably because it's possible that Facebook's growth will take it beyond 21st Century Fox, News Corp, Sky and Foxtel into second position. But you can see, there's been a fairly dramatic change, not so much in terms of the traditional media analyst in comparison to 2012, but in relation to Google and to Facebook. Google, in 2012, was #4. And Facebook didn't even make the top 10. In fact, on the footnote, you'll see that they were ranked 28th in terms of destination for media and investment. And I will just point out that both Google and Facebook do specialize in the…

If you could just let Martin know who you are, please. Yes, thank you.

It's Ian Whittaker from Liberum. Apologies, I'm actually going to ask 4 questions. First of all, just in terms of data investment management, I mean, that's been an area sort of where recovery's been talked about for a long time, but it's never come. I guess the question is, is it time now for maybe you to exit that business? It's not something that your peers have sort of in a substantial amount. The second thing is just in terms of digital. I think you said that the revenue growth in digital first 9 months was 2%. The slide, when you look to the top 5 online providers suggests billings were up 23%. That would suggest that essentially the long tail of digital is seeing a decline. Just wondered if you could talk about what the implications sort are of that, just in terms of the drag effect that you would have in terms of '18 revenues and beyond on the business. The third thing, just in terms of the margin guidance and the withdrawal, the sort of scaling back of that. So that had been talked about before that essentially, you thought you have enough protection for 2017 to keep that 30 bps of margin increase. Just wondering what exactly has changed here so that, is it that you're more concerned about ad hoc project work in Q4, which is something that Interpublic has mentioned. Is it that something has changed in terms of dynamics sort of in Xaxis [indiscernible] digital? Or is it something else which is happening? And then, finally, just in -- sort of from a more longer-term perspective. So if you went through the sort of perceived threats of sort of analyst [indiscernible] in terms of the threat to the agency business, which is a great explanation. I guess, the sort of question I would have, though, is the -- that's all fine, but what exactly can you do to actually get your growth going again? Because it sounded very much from those charts as though this is very much as though you'd have to wait for others to come back. If for example, you were to look at the broadcasting sector, they've taken steps to actually diversify their revenue streams away from TV ad revenues. Is there anything that you can sort of from your side to maybe change your business model to actually sort help your revenue grow?

Okay. So a lot there. Let me deal with the last one first and then maybe Paul can talk a little bit about margin protection. I'll try and talk about digital and digital growth. And then, lastly, I'll try and talk a little bit about the data side of the business. What we tried, Ian, to point out is what we can do. And if we look at the 4 things that we focused on, certainly, integrating our offer, making it more seamless -- I mean, clients I don't think worry, as I tried to point out, as to where each participant in the team, whether it covered advertising or media or public relations or data, however they -- they don't care whether they come from. What they want is the best people working on their business. So that's one thing. You pointed to the comparison to broadcasting companies. Well, the analogy to us would be geographical growth, finding the pockets of geographical growth. And I think if you look at where, again, where the next billion consumers would come from or the next hundreds of millions of consumers are going to come, they're not going to come from the U.S. or Western Europe, whether U.K. is in Western Europe or not. So they're going to come from Asia. They're going to come from Latin America. They're going to come from Africa and Middle East and the Central and Eastern Europe. And it may well be also that the companies of the future come from those regions as well. So I think the geographical bucket is another bucket. The digital bucket, which again, we've tried to emphasize, is also an area where we do see significant opportunity. We've tried in those brief statistics on Wunderman, on VML, on GroupM, to show where we do collide in the digital area. And obviously, broadening our digital offer can help that to. On the data piece, which is related to your first question, we think data is an important determinant. I mean, I look at the public relations and public affairs business. That has been changed to some extent, some might say to a dramatic extent, by the rise of social, by the rise of search and last but not least, by the rise of data. So in every one of those areas, whether it is integrating the offer in those higher-growth areas of client development, whether it's the geographical opportunities, whether it's the digital opportunities or the data opportunities, I think we have an opportunity. Maybe as we've got Mark and, I hope, Jon Cook on the line as well, I could just ask Mark talk a little bit about how you see developing Wunderman's digital offer, particularly in relation to the consultancies that we've outlined you competed against.

I mean, I think, look, where [indiscernible] that's the most competitive with the consulting firms, along with VML. We could see this more in the U.S. and the U.K. than we do in the rest of the world, probably where we're the most advanced in that area. And we've been investing heavily the last 2 years, both in terms of sort of acquisition, and as Martin said, realignment, as Salmon's coming into Wunderman, the whole Wunderman commerce acquisitions that we made, like Pierry, which specializes in Salesforce; or Cognifide, which specializes in Adobe, of growing out the technology part of our business and then really integrating that technology with what we would traditionally do from a marketing perspective. We think Wunderman -- we grew up from a direct marketing background, integrating that technology with what we deliver to clients. As I think about it, we sort of compete with consultants, I guess, in 3 areas creatively. And if you look at the list, the areas on the creative space, that's the one where we lost the creative work to Heat, which is an ad agency with Deloitte; or one with sort of more technology part of the business that in partnership with [indiscernible] 6 on the media, digital analytics functions, sort of when read the wrong way. The second area will be sort of technology, sort of Martin-led technology programs, where I'd say on average, we win our fair share. And then it will be sort of technology-led transformation programs, where the marketing element is smaller, the client tends to be the CIO, and where we do win programs, Salmon or ones in commerce as we're platforming or 2 or 3 major retailers in the U.K. And they'd beat Accenture and Deloitte Digital against those assignments. I mean, to give…

Mark, can I add to that? This is Jon.

Yes, go ahead.

Yes, I would -- this is Jon speaking from VML. And I would second everything that Mark said. We're also seeing on the front lines of competing in this -- in the world of consultancies. And where we've had success, I believe it's similar to Wunderman even if at a bit different flavor. But it's the idea of a connected consumer experience. And the big difference we've seen that we've been victorious, which has been often, has been the context of things like creativity, social media, data and commerce. Whereas in those competitions sometimes from the consultancy side, you see that as an add-on or an adjunct part of the expression. The idea of placing creativity in the right place, data in the right place, commerce and social media in the right place in the connected consumer experience has been a big advantage point. And the second advantage point that we've seen from a VML perspective, I think Wunderman would say the same thing, is that the globalization of creativity. A lot of the acquisitions on the consultancy side have been creative organizations, like some that Mark mentioned, but they've been regional in nature. And we've seen a big advantage in the global scale that we've built with WPP for globalizing creativity and messaging. The integration of the connected consumer experience and the globalization of creativity have been 2 major ways that we've seen an advantage.

Okay. Paul, do you want to just talk a little bit about margin and forecast for the year and margin protection? Ian raised the question about why, if we're flat in terms of the top line or down a little bit, we're so cautious on the margin.

Well, I think partly, it's because our businesses are feeling cautious, having seen what they've experienced in quarter 3 and quarter 2, to a degree. So if you sort of stand right back from it, our 2 higher-margin regions are the U.S.A. and Asia Pacific, and both have a certain degree of challenges. In the third quarter, although we did, I think, better than people generally expected, we didn't do as well as we had forecast ourselves in quarter 3. And the cost savings that we have achieved, that basically had to go against the revenue slippage as opposed to sort of come through in terms of incremental conversion. So incentives are taking some of the strain already. And I think it's sort of -- we're trying to give you a realistic view of how we feel things are going. What I would say is it's not, in my opinion, anything to do with the mix of our business between, let's say, the digital platforms or the media businesses, which actually have been very robust in terms of their profitability and performance. The research businesses, which have been very robust and sort of their profitability and performance, is more sort of our traditional businesses with the larger networks that have issues in certain markets, where they're finding it challenging to adjust the cost quickly enough. So I think they're being cautious about the expectation of flow-through revenues in the fourth quarter. I think it's our best view of where -- of how it's going to come out. Of all the categories of spend, staff costs are well controlled. Property's being well controlled. The old -- so are travel and commercial. The only area of business investment is in IT. And obviously, we also have had the malware attack where we had to make some additional sort of short-term investments to recover from that situation. But actually, investing behind the digital businesses is not something we are pulling back from. So that goes across the group. And yes, you mentioned -- as we mentioned, it's around 5 billion of revenues in this quarter. So Wunderman, for example, GroupM, Kantar do have significant investment programs in their various platforms to continue to provide the growth. So it's not about our digital media buying or our media businesses fading profitability. It's more about the mix of regions and the caution that we see having had 2 quarters where we haven't met our expectations, but being cautious about the fourth quarter.

Just Ian, coming to your point about digital. I mean, I -- the digital billings growth referred to in relation to '17, covers -- I think it's 5. It's not just Google and Facebook. It's Amazon. It's Twitter, and it's Snap as well. So I think that explains -- you've got to look at it, not just as being Google and Facebook, but those 5 as a whole. And I think what those figures demonstrate is the pressures that we've seen. I think Unilever, in its first half numbers, I think, said that it'd taken down agency fees by something like 17%. We have not seen that degradation in relation to our relationship. But we have seen significant cuts, not just in that case, but across the whole FMCG area. That's 30% of our business, and it has affected not just traditional, but digital as well. So I think it's wholly consistent with that. And I don't think you can read anything into it if, in terms of what's happening, for the long tail or the short tail. What I would also add is that given the pressures that we've seen, whether it be political brand safety, consumer brand safety, the Russian investigation or the Russian connections; the value, validation, viewability issues surrounding the 3Vs, surrounding Facebook and Google, what we have seen, I think, is increasing examination of spending, particularly by the short tail on Google and Facebook. I don't think that has dramatically influenced the growth rate. If I remember rightly, Google's 9-month revenue increase was about 21%. I don't think Facebook have reported their increase, or I can't remember what it was, for the first 3 quarters. But I think the growth rate is even stronger than we've seen at Google. I think, to some extent, what's…

Sure. To echo the point on horizontality and leveraging data, one of the points that Martin mentioned was leveraging our data asset. So mPLATFORM, which is our data and activation platform for GroupM, we brought that out to roughly 200 clients now. And one of the key differentiators in every conversation is our ability to link our asset to Kantar and something like Lightspeed. So as that platform progresses, I think you'll see true, true benefits that we're able to bring to the surface. In terms of digital in general, obviously, our relationship with Google, Facebook and others continues to strengthen. One of the things that we look to in the marketplace right now is, how do we build the best available partnerships with each one of these platforms? And how do we bring the greatest level of value that we possibly can? It's different for every client, the way that we engage with them. It's different in every market in the way that we engage with them. But we've been, in terms of Google, pleased with how that relationship continues to progress, whether it be with utilizing best practices across our search activity and partnering with them, whether it be with unique integrations with our platform to make sure that our audiences -- we match our audience into a Google Environment. And the same is true with Amazon, which I know wasn't listed. But that's certainly an area that we see a tremendous amount of interest in as we think about e-commerce and retail. Facebook, a little bit more distant in terms of where our relationship is right now from an integration standpoint. But we continue to move down that path and look for points of intersection where we think we can create value.

Yes, I just want to add to Ian's question, the one other thing in relation to data. I think one other things that our clients are increasingly concerned about is their ability to control the data, particularly in relation to what Brian just mentioned in relation to Amazon and Alibaba. And they're seeking to find ways of influencing or controlling much more effectively. I think, given the pressures that Google and Facebook are facing in brand safety, the political area, et cetera, they are becoming more responsive to the need to share data. Whether that is true of Amazon or, indeed, Alibaba in the longer run, is a question that is very much top of mind for our clients. And so coming back to the data and, again, using the word that Ian used, which was exit, it makes little sense to exit if that is going to become more and more important. So I think data and control of data and influence will become more, not less important. Want to move on, Paul, from Ian's 4 questions?

It's Patrick Wellington, Morgan Stanley. Just 2 questions. The one for Paul. Paul, the debt has gone up by about GBP 1 billion between the end of June and the end of September. From my count, which is not normally that accurate, I think GBP 270 million of that is acquisitions and share buyback. Can you say what the rest is and what the effect is of the India sales tax and the cyber-attack? And how much of that you'll get back while you're doing that? And for Martin, we've dropped the margin target. What about the long-term target, which are 10% to 15% earnings growth. I don't think you're going to do that in 2017. It's very possible you won't do it in 2018. So what point does one switch these long-term targets? Or do you just keep repeating them irrespective of the ability to do it in the short term?

Okay. So why don't I start off with the working capital. So the average net debt is around GBP 500 million worse than it was a year ago. That's really the effect of the spend being greater than the cash generation over the last 2 years. So I would expect that our point-to-point difference is very similar to our average difference, which it was in June, but is not in September. And there are a number of factors that have sort of gone against us in the quarter, I think. One, the malware attack and the Goods and Services Tax in India have been quite significant one. India was actually hard hit by the malware attack. We had all our systems down in India for the best part of probably 2 to 3 weeks. As I mentioned, then you had the Goods and Services Tax. It delayed our billing by really up to close to 2 months. It is around $125 million per month. So we're definitely, I think, $200 million behind in our India market in terms of collections. So that's one factor. We did an estimation of what it was in Europe. Around the end of June, recall, it's 100 million. Most of that should have been collected by now, to be quite frank with you. The other factor is whilst we're using a balance sheet, so our cash position is about 100 million, 150 million better than you would look in terms of the cash flow sort of point to point. The first factor, which is very difficult to explain, is when I look at the working capital. I know where I am in terms of trade working capital. And we had a very good significant improvement a year ago of around GBP 300 million to GBP…

[indiscernible]

The top end, yes.

Shall I move to the margin question? As far as the margins are concerned this year, Patrick, as we've indicated, because of the pressure on the top line, and Paul has responded on the margin question, we see margin improvement being either flat. I mean, you've seen what's happened in the first 9 months. It's up 10 basis points against targets of 30 basis points. If you look at reported or like-for-like on constant currency, it's the same as last year in the first 9 months. I'd just point out that the last quarter tends to be about 40% of our profit for the year. So we'll see how things shake out in October, November, December from a margin point of view as we go through the quarter. As far as next year is concerned and beyond, we're in the midst of doing our 3-year plans. We're in the midst of coming to our budgets. So we'll start looking at our budgets toward the end of November and into December looking at next year. It's early to come to a view as to what those budgets will show. But again, if we look at what we can do and what we will have to do from an internal point of view to deal with the compression we see in the top line, whether that's -- that lasts into 2018 and beyond or not, we're going to -- just like our clients are doing, we're going to simplify our offer and simplify, more importantly, our back offer -- we have to, we have no choice, in a more effect way. So when you come to looking at the margins, what I think we have to do is to look at our SG&A in -- or effectively, how we deploy that in…

Julien Roche with Barclays. The first question is on Q4. To be flat, if you use 0 for flat, you need to be up 2.1 in Q4, which is a 3.2 improvement versus Q3. Now the comps are 0.7 easier. So you kind of have a 2.5 gap. Where is that coming from? I know there's some benefit of account losses cycling, if you could quantify that. But if you could give us more color on why Q4 is going to improve? That's my first question. My second question is a follow-up on Patrick's on margin. Basically, your answer was if the top line stays there, we're going to have to work harder at integrating, simplifying, so lots of work for Paul and other people. You gave us one number in the last couple of years, which was 100 basis points of 2, 3 years from IT. Your preliminary assessment of the kind of simplifying cost-cutting [indiscernible]. Are we talking about another 100, another 200, another 300 to have a sense of what the margin can do if there's no growth for the next 2 years? And then the last question is on the lack of growth. So I accept all your explanation, not digital disintermediation, not consultant. It's pressure from clients. And your answer is at one point that I'll have to reinvest. The issue is that you have working media and nonworking media and you can invest in working media and continue to cut in nonworking media for a while. So why will that not happen for the foreseeable future? What's going to make them invest more in nonworking? Those are the 3 questions.

Well, I think, well, let's just deal with the third one first. If you're not getting the volume growth, what do you do? And if the number of users of your product is declining or not growing sufficiently, what do you do? I mean, I posed that question. That's number one. Number two, if -- you might be able to make onetime cuts. Whether you can -- you continuously cut costs by the same amount every year, I don't think you can do that. There is a limit to what you can do. You can't -- one line in our release is you can't cut your way to prosperity consistently. At some point in time, you're going to have to invest. I mean, the irony of what you just said is that in the U.K., where there are high levels of uncertainty, we're seeing about the strongest growth not everywhere, but one of the highest growth markets. Why is that? Well, that's perversely because there's so much uncertainty. I don't think clients are investing in plant and equipment. They're investing in variable spend on the top line to gin up, to tick up the top line growth. So I mean, it's a reverse of what we're seeing elsewhere. And the irony is, it's in a world of increased uncertainty where people, in order to get some growth, are investing in variable marketing spend. So I just -- I think at a point in time, the chickens are going to come home to roost. I mean, there are several companies that have pursued cost-based strategies, where analysts who follow those companies are raising the question about how you get top line growth or otherwise, it just becomes a downward spiral in terms of volumes, getting lower and lower. Now you mentioned…

No, I think we'd actually like to go faster because quite frankly we now have a plan of action that is now taking in stages the 6 major markets, including countries like China and India, and then moving on to the big decision on North America that has to do with how we've been very efficient in outsourcing. So we've kind of got 2 models in plan or in place. One is the centralization of our back offices [indiscernible] and getting efficiencies for centralization. The second, which actually Kantar have adopted the most aggressively is outsourcing directly to a third party, such as Genpact with guaranteed savings. What I would say is, yes, there is more we'd like to and we'd like to go faster and now we're proving it to be successful, both on the IT. We've got a shortened time line to -- in very simple terms, move all our data centers into the 4 global data center support of IBM, and we have a time line for the next 24 months, which we have to achieve, partly through the security initiatives that we've got and partly because actually, it's how we get our efficiencies the quickest. On the finance side, as I said, we are working through it. We'd like to go faster. Now we've got a proven model. But I also think it's -- there is some of the front-office side that helps as well. So if you take the combination of the Health & Wellness businesses, we're basically going to have once a year for what was previously 3 businesses. And that has actually already taken place. We're going to have -- we're not going to need 3 CFOs of the same skill that we have before when they were running independent businesses. We're going…

Yes, now just to add to that, the new business record in the first 9 months has been reasonably good. And that would -- should start to kick in as we go into the fourth quarter. And then the other point is that we start to cycle, I mean, the 2 principal losses that we had last year were AT&T, which started, I think it was in November. The loss was triggered, I think, in November of last year. VW, it really impacted from January 1 of this year. So we start to cycle out of the comparatives in a way get easier as we go into Q4.

It's Lisa Yang from Goldman. The first question is on the client spend. I mean, a few of your, actually clients in this [IT] space have been talking about increasing their spend in the second half. Is it something that you have seen already or you expect that to happen in Q4? Or are they reinvesting, but not going through the agencies? That's the first question. Second one is on your balance sheet. It now comes towards the upper end of your target range of the financial leverage. So just wondering, given it will be quite a good use of cash to buy back shares or do M&A, whether you would consider to be selling some of your stake to give yourself more headroom for that? And lastly is on Zaxus. Is it possible to have a growth number for Q3 in terms of bidding? I think in the first half, it was up 10. And given all the transparency concerns in U.S. I mean, is there any kind of strategic repositioning to be done to get around all those transparency concerns?

Okay. On client spend, I'll ask Brian to comment on Zaxus. I think for the year, we've indicated that globally, Zaxus would do about 10% in terms of increase of billings on a global basis. But Brian can talk to what's happening on a U.S. basis. On client spend, I think there has been talk, Lisa, that there would be increased spend. If you look at our Q3 being a little bit, but I wouldn't say it's been pronounced. And we'll have to see what happens as we go into Q4 as well and start to look at next year as well. In terms of selling stakes, I mean, we would only sell stakes if we thought it was the right time to do so or rather than where we had to do so to reduce leverage. We are at the, as you say, at the top end of our range of 1.5x to 2x EBITDA. And -- but the cash flow of the business is strong and will improve given the comments that Paul made in relation to the net debt position as of September 30 and the impact of the cyber-attack and the GST in India. So I think in answer to selling stakes, we would sell if we thought it was the right time to do so rather than be a forced seller. Do you want to add any more on Zaxus, Brian, and particularly, the situation in the U.S. with respect to transparency?

Sure. So Zaxus, we continue to pace high single digits to, as Martin mentioned, 10% growth for the year. And in the U.S, particularly, we've had last month, the month of September globally was the best month we've had for the year and continues to pace well. In terms of diversification of the portfolio, just to give you an idea where we're going, in the U.S., the transparency in the ANA, which I think was almost 15 months ago, it was less than 5 clients who pulled out of Zaxus in the U.S. Where the U.S. where we've seen globally as well, the business continue to evolve is around outcome-based meeting. So we're running AI or artificial intelligence on nearly 40% of our campaign. And the ability to drive top line results is the singular focus for Zaxus. At the same time, we diversified the portfolio in the U.S. about where we've introduced a new product or a new media product called [indiscernible], which came out of EMEA, which is our native offering. That is -- has significant growth in the U.S. and will continue to evolve our product offering as well. But overall, based off platform is where we continue to pace [indiscernible].

I mean, I'd just point out one point, maybe you can amplify this, Brian. One of the things we've noticed, coming back to the focus on cost, is that some clients took a view on the so-called nondisclosed model. Just to be quite clear, we -- Zaxus operates on an opt-in model where the buy price is not disclosed. But everybody is clear that we're acting as principal and that, that buy price is not disclosed because although the client, of course, knows the price which he pays, he or she pays for the inventory. So one of the interesting things, I think, is happening is this, is that whereas, let's say, 12 months ago, clients might have taken the view that, that model was not right for them, their business units and their business divisions are under such pressure for performance that if you can demonstrate that Zaxus or indeed, any other disclosed alternative, opt-in alternative is more effective, then the client's business units would suggest to the center that, that might be an extremely effective way of improving efficiency and performance and reducing costs. I mean, do you want to say anything, Brian, about -- without being on specific clients as to what -- whether you see that as a trend?

Absolutely. I mean, I think you see the number we probably in the past is on average Zaxus is driving efficiency of roughly 25%, if we compare to other models in the marketplace. So we've seen clients come back to that model as they've gone or attempted to use other things. Obviously, that 25% efficiency shows up very, very quickly in our other media spend. The same is true across all of our portfolio. As I mentioned, Mark mentioned Triad before. So Triad driving eCommerce or retail revenue in-store is a direct correlation to media spend. So we're able to show return on ad spend, $1 equals X. And that is the entire focus of our meeting product. So we've seen, I think, to Mark's point, 2 or 3 clients actually move back into Zaxus that previously hadn't been clients that have been long-standing by WPP relationship.

Charles Bedouelle from Exane BNP Paribas. I've got actually 2 questions. The first one is on the digital growth because we've heard Zaxus kind of coming back. We've heard the [indiscernible] doing well. We see that you did 23% with the top 5 guide on the gross billing. But the digital is up 2% to 3%. So can you maybe explain where is the digital suffering or repositioning? That's the first question. And the second question, just kind of housekeeping. Paul, what do you think will be the associate contribution or increased contribution on a full year basis so that we can have many of the underlying, kind of margin target for this year at this stage?

Do you want to address the housekeeping question, Paul?

No. Not quite yet.

Right. I mean, on digital, I think, Charles, the answer to your question is sort of a similar to Ian's question is that we're seeing a pressure on cost across the board. And I don't think you can discriminate at the end of the day between what you see happening traditionally or what you see happening in the digital part of the business. If the digital part of our business is 40% of our revenues, if it's 30% of the market, which, what appears to be the market, the cost pressure is applied to both. And the concerns around brand safety and viewability and measurability are similar. I mean, I'll ask Mark and maybe John to comment on what they see on pricing. Do they see clients willing to commit to digital programs willy-nilly without looking at cost at all? Do they see clients being indiscriminate on the cost side on digital whereas on the traditional side, they see them being more -- more efficiency driven? I mean, Mark, do you want to comment on that on pricing?

Yes. I mean, I think the answer is no. I mean, whilst we're probably in the bit of the business the clients want the most of, there's still pressure on pricing. There's still pressure on cost. They're still going through your detail if you still spend 2 months negotiating the scope of work on the contract and the pricing. So procurement department heavily involved -- as heavily involved as they are in other parts of the business. So I think that directionally, we face some of the same headwinds from a cost perspective, although probably we've seen tailwinds from a kind of overall demand perspective. But there's no difference. John?

John, do you want to add?

I concur. I think where there's no difference in -- especially when it comes to content, it's about content creation, about the efficiency of creating that content and that storytelling and getting that word out. And there's not a discrimination between whether that content lives in traditional media or digital media. It's similar pressure, and I agree with Mark.

I mean, the only area, to be fair, I think Charles that you see less discriminatory pricing is probably the highest growth areas. And just going back to the S&P 500 analysis, if you focus just purely on the tech side, you probably would see the growth rates are so strong in that particular example, 4x what you see in packaged goods, the likelihood is that the tech clients are going to be more concerned about implementation than they are about pricing, that the growth rates are so strong that they're more concerned about getting things done than arguing about the cost of which they get done. Are you ready for your housekeeping, Paul?

So last year, just to remind people, we did GBP 46 million income from associates. At the half year, sorry, that's -- this is the half year number. We did GBP 46 million in the half year. We're looking at around GBP 90 million on a full year basis. That is considered to be stronger than last year's number, GBP 65 million. And part of it, if you remember, we won a big business with Haworth at the beginning of this year that is a big contributor to the associate income change and we have done considerably better in some of the Kantar associates. So net-net, we'll be around GBP 90 million this year versus GBP 65 million last year, double the half year run rate.

It's Richard Eary from UBS. Just 3 questions. Just obviously to get back to the margin question again, unfortunately. Just if we go into '18 and if we do get a repeat of '17, where we get a flat growth environment, I mean, can we get a view that given the cost-saving initiatives that Paul, you talked about and both Martin and you talked about, can we get a situation where we can deliver a flat margin target in '18? So that's the first question. The second question just comes to, if you look at the advertising and media business, and you look at the direct costs, i.e., between the gross revenues and net revenues in the third quarter, it seems as though that direct costs actually fell year-on-year after being up quite strongly in Q1 and Q2. I'm just trying to get an understanding of what happened within that and whether there's any mix issues that we should be aware of. And then the last question just on client wins. I think you actually put in the deck, obviously, the announcements of the losses in October, the ABI and Lionsgate that take sort of in net billings about GBP 5 billion year-to-date. How do we think about like contribution in terms of revenue growth into '18?

I mean, on the client wins, I think we're early on in the process to give you a view as to where we'll end up on revenues or revenue budgets for next year. So I mean, I think you look at it as being net-net a tailwind. Obviously, when you refer to specific accounts, the specific terms on each of those accounts are particularly important. And in fact, if they're losses, the terms on which you operated on the historically is important, too. So I would say on the client wins or the net win position for this year going into next year, it's early to estimate what that impact would be. On mix, I mean, do you want to comment on mix, Paul?

Yes, it is really difficult to analyze for you, actually, because the direct cost of combination are 2 things. And actually, I did pull the numbers out. So the half -- the quarter is GBP 460 million combined between the media businesses, which is principally Zaxus and platforms acting as principal, not solely, but principally that. And the Kantar direct cost for field research. And actually, on the 9 months, it's GBP 1.5 billion. So we're kind of running at around GBP 500 million for the quarter, and that hasn't really changed. That is the interrelationship between the growth of Zaxus, which you've heard from Brian, is doing well, versus the spend in the field force numbers, which tend to say correct or say, the projects tend to be completed for the half year. So what you tend to find, what we've noted, is quarter 1 and quarter 3, there's a differentiation that widens between revenue and net sales that closes quarter 2 and quarter 4 at the same time as many of the continuous tracking project get completed and delivered to client. That's probably the best explanation I can give you. So it's running around GBP 0.5 million per quarter. That's been a consistent rate. It then depends on really the speed at which Zaxus is implemented in other markets. For example, now it's doing very well in Latin America. That will probably distort fractionally the relationship between revenue and net sales in Latin America. That's probably the newest market. It's a small base, but it will grow strongly. So those interrelationships that are really difficult to predict, but they tend to self-correct in quarter 2 and quarter 4. They tend to be more out of alignment in quarter 1 and quarter 3, partly due to the [indiscernible].

Just going back to the first question, which I think if I heard it right was, if your revenues are flat or our revenues are flat next year or net sales are flat next year, do you think we can deliver flat margins? Was that the question?

Yes, that's correct, Martin.

Okay. Well, I think the answer is we've always felt that we had 2 years flexibility in the event of revenue slowdown. So if we came -- let me put it like this. If we having done the budgets, felt -- came to the conclusion that flat revenues were again going to be the situation for next year, we would be focused very much on flat margins. In fact, we would be calling for our business units to be trying to improve their margins in that context. But overall, to answer your question head-on, we will be looking for at least flat margins for next year.

Tom Singlehurst from Citigroup. 2 questions. One is [indiscernible], just trying to work out whether no deal is better than a bad deal in terms of the proposed transaction there. And then, secondly, on the change in remuneration. You talk about half of remuneration being linked to overall good performance. I was trying to work out whether that's a good thing or not. I can see it obviously helps cross-selling. But at the same time, it does mean your highly talented staff in fast-growing areas get dragged down by it not perhaps doing as well.

Yes, I mean, on that last question, I mean, we have to get used to the fact that -- and by the way, that 1/2 of the group incentives being tied to overall performance is not new. And when we scaled it up, it was, I think, 1/6, then 1/3 and then 1/2 over 3 years. I think it's been operating for at least a couple of years. I mean, we have to get everybody attuned to the fact that we're trying to get the company to think as one instead of a myriad of different atoms. And that realization to -- again, to answer your question head-on, I think it's critically important. There may be better ways of achieving it. You said, Tom, to achieve cross-selling. It's not cross-selling. It's to achieve a situation where, I mean, for example, one client and this is one of a significant number, have said to us they don't want to deal with multiple P&Ls.

Why don't we take another question and then we'll come back to Martin. I'm sure they'll be a second here to get reconnected. If there's another question, let me try and deal with that. Have we lost the mic as well?

So it's Matthew Walker from Credit Suisse. 2 questions please. The first one is on -- going back to transparency. It sounded like [indiscernible] not that worried -- you're not that worried about transparency. But U.S. was down really heavily in Q3. If you're being like totally honest about it, how much of Paul's group's profit, not that you aren't always, obviously -- how much of the group's profit are at risk from people doing things like tightening up media clauses and contracts to eliminate unearned income, all sorts of further contract, all that kind of stuff. What percentage of the group's revenue -- profit is at risk from people tightening up on those things? The second question is, you raised a really interesting point when you're talking about Facebook and Google in the statement, it also raised an interesting point about the platforms, the big platforms, the tech companies in China. If you look at what's happening in China, advertising revenue has grown reasonably strongly, but the group's revenue has not kept pace at all with that [indiscernible]. You're also saying that some of the platforms in China, maybe clients are going much more direct. And given your investment in China and how important it is for you for the future, what can you say about the prospects for China? Why are people perhaps potentially going direct to Tencent, et cetera, in a way that they don't do for the U.S. thing?

So let me just start on the transparency answer and maybe Brian can help in a second. So whilst the ANA report came out, I think what we said at the time and certainly it's old news if you go right back in time, one, we have a higher degree of multinational clients than any of our competition. The issue really surfaced in 2000, went into public, came out after Eliot Spitzer made an issue on speaker commissions in the insurance industry. Then the public came out with a hit of around GBP 250 million for not honoring the full terms of the client contract. So ever since 2000 and then certainly, with the advancement of soft and other legislation in the U.K. and U.S.A., there has been full transparency with our clients on all forms of media, rebates, incomes, et cetera, earned. That gets audited repeatedly at -- by third-party auditors around the group. So whilst there has been a lot of noise around transparency and if you do read the ANA report, it is a lot of hearsay about what used to happen in industry and executives that are no longer in the industry. It has not had an effect in terms of our returns nor our, I'd say, transparency with clients about the information they have received in the past from us. So it has not had a change in our relationship with them in terms of fees earned and total return versus what media owners pay us. So that's my view of the situation. And whilst there was a lot of noise, actually of all the markets that doesn't have any rebates at all, is the U.S.A. So what was clearly was an issue is a lot of the industry started [indiscernible] client side had difficult…

Are we back on yet?

You are now back on.

Okay. Fine. Okay. So just coming back to China for a minute. I think the issues that we faced in China, particularly in the last couple of years, have been twofold. Firstly, around media broking and our unwillingness to use media broking. And we've certainly made begin roads, particularly in the last 2 or 3 quarters, for example, in the Chinese auto business in pricing. And ironically in relation to the first part or the first question, transparency in the Chinese market. So that will be one thing. And the second thing is our data business in China has been under significant pressure, also primarily in the auto category. But I think that has been one of the reasons. But I think we've seen a little bit of a lift in that area as well. But I think, principally, the Chinese GDP growth rate of 6%, 6.5% has not translated into packaged goods or consumer goods or multinationals in China. If you look at the results of the major multinational companies in China, they've been under pressure too for exactly the same period that we've seen, the last 2 or 3 years. So my view is it's not so much of -- we do acknowledge in the statement that the rise of Alibaba and Tencent have probably had more of an impact in the sort of Google and Facebook category question than in the cases of Google and Facebook. Coming back to the U.S. transparency issue. I mean, Paul, I think answered it very well. We have not seen that impact. What the impact has been to cast suspicion and innuendo, not fact. In the reports that the ANA issued, we have to say yet again, there is no specific case given whatsoever, none. And I think our competitors have…

Ruchi Malaiya from Bank of America. Martin, in the background, we can see your accolades from Cannes and the [indiscernible]. And perhaps in the context of addressing your cost base, how do you weigh up the cost benefits of entering both product [indiscernible]?

Well, obviously, we expressed some concern last year. One of our competitors, because they were concerned about the cost, withdrew. And I would say that many in our organization felt that, that was a wise decision by the competitor. We placed a proposal or a series of proposals in front of [ Essential ]. They've responded to several of them. They haven't responded to all of them. We found their response to date not meeting our objectives. We haven't come to a final decision as yet. But we're seeing whether there is a meeting of the minds principally around cost. So what we are saying is if you were starting from scratch from a zero-base, where would you have a creative competition or festival of the sort that Cannes represents? Is Cannes in the middle of the summer necessarily the best place to do it or the most cost-effective way to do it? Or has it become too big and too expensive? That's at the heart of what we've said to [Essential] and where we're still waiting for an adequate response. As to the important of awards, we think they are important. Whether they're EFFIE awards or whether they're creative awards, the only question is whether Cannes or [ Essential ] has executed in the best way. And I would just give you one other observation. The Eurobest Awards, which is another European-oriented award within the [ Essential ] stable is often used as a condition to further involvement in Cannes. That's something we find unacceptable. For example, one of our agencies in Europe has been told that they will only be considered for a position as a judge at Cannes if the enter the Eurobest Awards. We think that's an unacceptable practice. So net-net, we've made some proposals. We've had some response from [Essential]. But as yet, we haven't had sufficient response.

Thank you. I think we have drawn to a conclusion, if anything. Thank you, Martin.

Okay. Thank you. Sorry about the break.