Member Info for 1GW_

Sorry - those stats are (of course) US daily people, not monthly.

I don't know. I'm still looking for something to explain the "collapse" in R1 Premium reach from round about 10th Feb. Giant Realm, Stadium, MomIQ (and indeed "Other") were all hit: Giant Realm: 1.5m US monthly people on 10th Feb, down to a low of 47k on 9th March (now 139k) Stadium: 1.0m on 10th Feb, 33k on 9th March (now 118k) MomIQ: 101k 9th Feb, 31k now Other: 1.2m 10th Feb, 150k 5th March (now 265k) At first I thought, if it wasn't just a temporary blip, maybe they were pushing supply away from these segments towards AdKarma given the way the AdKarma numbers were growing (possibly a tribute to AdKarma sales skills). But it would make sense also to push visitors towards the owned AllMedia sites if they thought they could retain the majority of the reach.

Shroder - yes, now you mention it, I remember. But the message suggests that they expect either advertisers or punters to arrive at those pages somehow - and so are trying to push them towards their owned supply. And presumably "Giant Realm is no more" refers to the channel itself - so are they abandoning that segmentation of supply (the offer of 4 or so "major" vertical channels to advertisers) in favour of a more granular segmentation backed up by Quantcast analytics?

Update found by "GiveaToss" on advfn would help to explain the dive in R1 Premium stats. The owned and operated sites should be higher margin than the third party sites so if they can channel their audience into these sites then that ought to be positive for the financials as long as they don't lose too much reach in the process. "Game over - Giant Realm is no more, but maybe one of our other sites will interest you: AllMusic, AllMedia, Sidereel" http://www.giantrealm.com/

Tractorhead - I think "collapse" is overstating things. On the US people metric, the monthly total has fallen from a peak of about 214m people (30 days to 25th Feb) down to the current level of 191m - i.e. a fall of just over 10% in US reach. Mechanically, what seems to have driven this is a "collapse" (and here I think "collapse" is a reasonable word) in the stats for the RhythmOne Premium segment. This has fallen from an initial value of about 63m monthly US people on 19th Feb (when there was 30 days data in the system) down to the current value of just 21m (a fall of 2/3). RMax (RhythmXchange) US monthly people stats have stayed relatively steady during this period and AdKarma stats have increased. The question is why has R1 Premium fallen off so sharply and not recovered? Has blinkx moved inventory from this segment to others? Or have blinkx and/or the publishers required inventory to be taken offline in order to upgrade e.g. to optimise for mobile? Or does it really reflect a dramatic drop off in reach for the R1 Premium segment - which would be worrying if the name "premium" reflects the relative advertising rates available. If you look at the sub-segmentation of R1 Premium, it looks like the old Burst Media channels - and Burst Media network by itself had a reach of 200m US monthly people at one point (September 2014). Not sure anyone but blinkx management will know the answer, so I think we just have to wait for further interpretation via the blog and/or a trading update/results.

I think £10m-£11m of (pre-exceptional) finance costs would account for most of the gap wouldn't it? Finance costs last year were £9.9m (excluding exceptional finance costs) and in 1H this year finance costs were up a bit on the previous year.

There can be various motivations for borrowing stock. Pace recently was subject to a takeover and therefore a lot of arbitrage with arb players going long Pace and short the company taking Pace over. At the same time a huge stock on loan position (over 20% from memory) in Pace was built up and the most plausible explanation I could see is that US-based arb players were engaging in swaps (total return or otherwise) to acquire their Pace long positions. i.e. the derivative instrument (the swap) they used to acquire their long position involved someone else loaning them (or their counter-party) stock. So here the stock on loan position wasn't really bullish or bearish, it was just a by-product of companies taking out derivative long positions in the company that would be matched by short positions in the acquiring company and would net to zero if the acquisition went ahead. However, I think the usual assumption is that an increasing stock on loan position is likely to correlate with an increasing short position. So this has a negative impact on the share price while the short position is being built and you can regard it as a negative going forward if you think the shorters are likely to continue building their positions, or you can regard it as a positive going forward if you think the shorters are wrong (or finished building their positions) and they are going to want to buy back in the relatively near-term to close.

just appeared...

A couple of extracts from the annual report for FY15, which I interpret as meaning no particular weighting to "high end" London flats in FY15 at least. Don't know what other "high-end" projects may be on the go other than nine elms. "The number of legal completions in the London boroughs rose by 19.8% to 1,481 (2014-1236) as a result of high demand and significant land investment in this part of the country in recent years." "Within the London boroughs, the average selling price of private homes was £306,949, which is still affordable in the context of the London market"

According to link below, over 500 homes due for completion in 2017. On the Bellway site, it seems this is 510 homes, of which 434 are open market (76 affordable), with launch in October this year. http://nineelmslondon.com/news/bellway-starts-building-in-nine-elms To put this in context, in FY15 Bellway sold 7752 homes (6178 private, 1574 social) and in 1HFY16 its recent trading update said it completed 4188 homes. So on this basis, and given the group's strategy of volume growth, maybe Nine Elms could account for a little over 5% of FY17 completions, even if all of them fall into FY17. Doesn't sound like it should be disastrous even if Nine Elms prices do come off significantly around the time of the launch.

IAB Programmatic Marketplace event is on in New York on Monday, with RhythmOne as a supporting sponsor and 7 RhythmOne particpants (including Dan Slivjanowski, Dwight Ringdahl, Sudhi Herle and Ken Napolitano). If you go to the "sponsors" tab and click on RhythmOne you can read the para pasted below, which highlights R1's "top 10" ranking in Quantcast as evidence of having one of the largest supply footprints in the industry. On the public stats of course it has now dropped just out of the top 10, but I see this as evidence that R1 do believe Quantcast now serves as a shopwindow and they are confident that R1 will regain its top 10 position in due course. Together with Dan's recent reported comments on figuring in future pixalate rankings, this seems to me evidence of a creditable initiative to build the public profile of R1. "RhythmOne connects audiences and brands across devices at scale. We maximize ROI by providing an efficient marketplace for digital advertising. With one of the largest supply footprints in the industry (top 10 according to Quantcast), our cross-screen programmatic platform, RhythmMax, provides a quality environment for advertisers to reach audiences at scale, across video, rich media, display, social and native ad formats. Furthermore, we take a very strong stance against fraud. RhythmGuard, our proprietary pre- and post-bid filtering technology, eliminates suspicious traffic before it reaches the marketplace – delivering the highest quality inventory to advertisers. Find out more at www.rhythmmax.io." http://www.iab.com/events/iab-programmatic-marketplace/#sponsors

Excellent. Thanks for that Bmac. I look forward to next month's index.

I don't disagree, Shroder, but that's the same for all exchanges isn't it? If the pixalate index has credibility (and the message I got from Ed was that it did) and if blinkx think this pre-filter analysis is prejudicial to R1 because of R1's own filters (i.e. R1 would rank higher if the analysis was done post exchanges' own filters) then R1 ought to be trying to get that message out there - maybe a further blog piece on pixalate to follow on from the Quantcast and (hopefully soon) ComScore ones.

Well as I read the press release, they say they're using their filter to determine the "true" reach of the networks, which is reported in the global seller trust index (GSTI). "As part of the new GSTI, the company introduced TruReach™, a new revolutionary reach metric that filters out fraudulent devices and accurately measures the real human users a seller or exchange has access to." hTtp://blog.pixalate.com/pixalate-launches-its-next-generation-global-seller-trust-index/

Some familiar names in the lists but no sign of RhythmOne on either the Global or Mobile seller trust index as far as I can see, which I find a bit disappointing. I asked Ed Reginelli at the AGM about the relevance of Pixalate and he said they used Pixalate, think the index is reasonable and in time we should see RhythmOne appear in the Pixalate lists. http://www.pixalate.com/sellertrustindex/mobile/#!global

That last month (bought 0.75m shares) should be Jan'16 of course.

Morningstar is pretty good on those funds/institutions that supply regular data, and River & Mercantile is one of these. Morningstar updates monthly for R&M and I have cross-checked in the past against the 1/2-yearly published accounts (via Trustnet) and got the same number (The Morningstar number was the sum of the R&M published numbers for the UK Equity High Alpha, UK Equity Long Term Recovery and World Recovery funds). R&M could hold other blinkx shares that are not reported in the fund disclosures, but I believe Morningstar is accurate for those funds. The R&M position can also be cross-checked against the FT markets tear sheet, which currently reports 9.69m shares as of 30th September - which is the same number Morningstar was reporting for end-September. R&M was both active in the market and a heavy seller through 1H 2015, but started buying again in November according to the Morningstar data. Given that almost every other institution reported by Morningstar has sold out, I am prepared to celebrate one that seems to be in acquire mode. Jan'15 Sold 0.49m shares Feb'15 Sold 1.44m shares Mar'15 Sold 1.61m shares Apr'15 Sold 1.01m shares May'15 Sold 0.50m shares Jun'15 Sold 1.53m shares Jul'15 Unchanged Aug'15 Sold 0.14m shares Sep'15 Sold 0.35m shares Oct'15 Sold 0.40m shares Nov'15 Bought 1.7m shares Dec'`15 Sold 0.08m shares Jan'15 Bought 0.75m shares

increased their holding by 750k shares in January according to Morningstar, which puts their current holding at 2.8%. http://investors.morningstar.com/ownership/shareholders-overview.html?t=BLNX

I believe what forced the write-offs was the reorganisation. They wanted to move to RhythmOne as soon as possible and were in discussions with the auditors about the reporting implications. Having gone down this path, I don't think they would have had much choice but to recognise the write-offs in 1H. I discussed this with Ed Reginelli at the AGM (late August), and made the following note at the time: "CGU growth rates from the annual report. With the move to the RhythmOne organisation, the forecast growth rates for individual CGUs (Rhythm, Burst etc) will no longer appear in the annual report. They will move the goodwill across to RhythmOne subject to any necessary impairment (exercise underway at the moment apparently)." And in terms of BM's options having such a "high" trigger point, this would have been reflected in the quantum of options. So running various options possibilities through Black-Scholes (or whatever model) to ascertain fair value, obviously the higher the trigger point, the more options have to be awarded for equivalent fair value. And I doubt that RG and Toscafund would have been supportive of a much lower trigger point, which to many of us I think seemed very low at the time given the share price history.

Well purely in terms of the Quantcast rank, we're tightly bunched with the two above us and we seem to have had a bit of a weekend dip, so it wouldn't take much to turn that round. 11th, 210.6m Rhythm One (US network rank, US monthly people) 9th, 214.7m sovrn And number 2 in the list (Pulsepoint Programmatic), which got a day at number 1 a few days ago I think, is "only" at 240m US monthly people, so having 7/8 of their total really isn't bad in terms of US reach. It's now perhaps more about what we do with that reach (i.e. monetising it) that counts. The apparent rise in the monthly stats over the last month has been mainly if not entirely due to adding data in to Quantcast, I believe, so it would be unreasonable to expect continued growth at the sort of rate we were seeing in early February. However, having said all that, I agree that the trend in daily data from the peak of 29m US people / 30m US uniques on 10th Feb down to the current levels is disappointing.