At the Appnation show in San Francisco yesterday Nielsen presented data from a survey of 4,000 US mobile users conducted in August. The focus of the survey was about apps, popularity and associated user behavior. But there were a number of advertising-related questions included. I've included the four ad-related data slides below.
Here are some high-level takeaways:
Clicked on an ad within an app
Reaction to mobile ads by age group
Preference for ads within apps or outside of apps
Actions taken when viewing mobile ads
What's interesting about the data immediately above is that it shows mobile helping to generate awareness (like TV or other traditional media) and then driving additional research (online, search). There's plenty of direct response activity in there too. But the largest response, "used a search engine to find out more information," is consistent with what users do online after being "stimulated" by an ad (newspaper, TV, onlne) or product/service mention elsewhere.
The data also indicate that there's a potentially important connection between search and display in mobile. However it's not clear whether "search engine" here refers to any search engine (PC or mobile) or just on the PC. If it's PC-based then marketers will need to coordinate mobile display campaigns with online paid search.
Last week Nielsen released data based on a recent survey of 4,000 mobile users (geography uncertain) who had downloaded an app within the past 30 days. The survey found that iPhone owners had "an average of 40 on their phones . . . while Android owners report having 25 apps on their phones . . . and BlackBerry owners report having 14."
Nielsen also reported that "Facebook is the most popular individual app on all of the major operating systems." In terms of categories, games leads followed by weather (strangely) and maps/search.
Here I assume that "search" refers to map-based search or local search given the way the Nielsen has grouped them together -- although it could just refer to general search.
The question of user attitudes toward mobile advertising is widely discussed. There's lots of survey data showing that most mobile users don't like or want ads. Alternatively there's some data (including ours) that show when there's a clear exchange of value (saving money) that receptiveness goes up.
I've long discussed the differences between user survey responses and user behaviors in context. If mobile ads were so widely reviled why would they perform better (per Dynamic Logic and Insight Express) than PC ads? Let's look at some data.
Parks Associates released an in-apps ad forecast. In the context of their press release the firm showed some consumer survey results about mobile ad acceptance:
What the chart above shows is that 90% or more of adults (sample size undisclosed) either don't like in-app mobile ads or are indifferent to them. Across the board 50% say they don't like them.
A survey we conducted last year shows a higher percentage of users say they don't want any ads on their mobile devices:
Source: Internet2Go (n=707), 2009
If you segment for smartphone users you find somewhat less hostility to mobile ads; "only" 54% say they don't want ads in our survey. But these survey data don't tell the whole story.
Ad network inMobi recently presented survey data (n=4,399) that showed when consumers are clear about the value exchange (app is free, ad is personalized or more relevant) they respond favorably and their level of acceptance goes up. And when inMobi tested four mobile display campaigns with these respondents it found that 66% (n=2,893) would click on one or more of the ads presented.
Another survey conducted (n=1,000) found that up to 43% of respondents were interested in ads that would help them save money.
What this all shows is that when consumers are asked things in the abstract ("do you want ads in-apps/mobile?") they often respond "no." But when shown ads concretely and exposed to their benefits their behavior will often be quite different and even their formerly hostile attitudes will change.
Mindshare is consolidating around Android and many pundits are predicting that the Google mobile OS will soon become the number two smartphone platform globally. The only question is when.
Financial firm Piper Jaffray says it will be 2012, IDC says 2013 and IT consulting firm Gartner now says it will be this year as well.
All of these predictions are based on current sales momentum enjoyed by Android, perceived weakness from Nokia, RIM and Microsoft. They also assume that Apple sales will be steady but flat compared to Android. Some firms also believe that RIM and maybe Nokia will be compelled to put out Android handsets.
Here's the Gartner forecast, which predicts share losses by everyone but Android during the forecast period:
Clearly Android has momentum but it's possible that Winodows (Mobile) 7 and a reinvigorated Nokia (MeeGo) could moderate or diminish some of Android's potential gains. In the US a more widely available iPhone would almost certainly do the same. The Gartner forecast also probably underestimates iPad and iPod Touch sales.
Indeed, forecasts inevitably turn out to be incorrect though they may be "directionally" accurate. Whether or not Gartner's numbers above come close or miss the mark by a mile, it's clear that Android is a remarkable success story for Google.
Related: Smartphones selling faster than anticipated
As though it were a big surprise dozens of articles appeared in the past 72 hours based on Quantcast data that showed Android now represents 25% of overall US mobile Web access compared to the Apple's 56%. This data doesn't capture activity on apps I believe. And while the chart below says "iOS" it's not totally clear that this encompansses iPhone, iPod Touch and iPad.
As more Android devices are sold more people use them to access the mobile Internet. It's a very simple equation. There are now well over a dozen Android devices in the US market, from all the major carriers, with more on the way. Contrast that with the iPhone -- a single device from a single carrier. Apple has throttled its own growth through allegiance to AT&T.
The more significant information in the Quantcast data is not that Android is growing and gaining share but that other platforms are not -- RIM and Windows Mobile in particular. Windows awaits its total refresh and consumer reaction but RIM is in trouble if the Quantcast data are correct.
Contrast the Quantcast chart with on from StatCounter below, which shows a similar trend (iPhone vs. Android) but a different picture for RIM in particular. However the StatCounter data is simply OS market share, not necessarily mobile Web access.
Update: Here's an update on the Quantcast data, released today. The company cut the data slightly differently in this analysis by device OEM. And it offered the following commentary:
Apple's iPhone has seen a solid reversal of its downward trend since the introduction of the iPhone 4, however, this is offset for iOS share by the decline of the iPod Touch (which we include in the mobile web though it is not a smartphone).
As we can see below, the iPhone put in a solid performance in August 2010.
Pew is out this morning with more data on US mobile user behavior and attitudes, this time concerning calling and texting. Here are several of the interesting findings from the survey of 1,917 US mobile phone users:
Pew also speculates that in some cases people may be using text/SMS as an "Internet replacement." However this conclusion is inferred and not based on any direct question(s). The idea is that Internet users have other communication tools and channels available to them and so they rely on SMS less heavily:
Adults with cell phones who text, but who use the internet infrequently, are more likely to use text messaging to have long conversations. Adults who use the internet less than once a week are more likely than those who use the internet more frequently to say they have long message exchanges on important personal topics several times a day – 22% of infrequent internet users report this, while 8% of daily internet users and 6% of those who use the internet several times a week report having frequent, long text exchanges. These infrequent internet users are also more likely than weekly users to use text messaging to coordinate meeting up with someone (21% vs. 8%) and to use text messaging to communicate silently with someone (19% vs. 7%). Taken together, these findings suggest that some cell phone users may be using their phone’s texting capabilities as a substitute for internet access on a computer.
I'm an Android EVO user and I rarely click on ads, but that's apparently not true of my Android brethren and sistren. According to ad network Chitika, "people on the Android OS clicked on ads 81% more often than people on the iPhone." Chitika quickly comes to the conclusion that 'Android users [are] 80% more valuable than iPhone users." However I wouldn't equally jump to that immediate conclusion.
Ad mediator/exchange Smaato previously found, similarly, (in February) that Android users clicked-through more than iPhone users:
But later Smaato reports, as Android penetration grew, saw its CTR position decline. Compare July numbers for the US and then globally:
There may be some novelty around mobile advertising for many Android users, especially if they've upgraded from feature phones and this is their first smartphone. However, the Smaato figures suggest that the Chitika numbers will similarly change in a few months after the newbie Androids become jaded and embittered.
Once again, however, this all begs the question of whether we should be paying any attention to CTRs on mobile ads (other than paid search). If there's a lead-gen form on a landing page or a store locator, then arguably yes. But otherwise CTR is probably the wrong metric.
It's an easy, lowest-common denominator metric so people gravitate to it. But comScore has repeatedly argued and shown that CTR is the wrong metric to assess the efficacy of onlne display advertising. That should equally apply to mobile.
There are a range of metrics that mobile marketers can and should be using to assess the success of their campaigns, depending on their objectives. The generic CTR is not necessarily very helpful or meaningful in terms of brand lift, purchase intent or other actions that mean something in the real world.
Nielsen offers data culled from a year of looking at US mobile subscriber phone bills and other sources. It shows the places and populations who talk and text the most by state and ethnic group.
There are geographic differences in terms of voice minutes and texts per month. And it's a bit of a non-surprise that teens text way more (emphasis on the word "way") than anyone else -- more than twice as many texts per month than those in the 18-24 age bracket.
Residents of the south talk more than others, while the states with the greatest volume of texts are distributed:
The chart above confirms (again) that the most effective way to market to teens may be SMS-based.
Millennial's July data were released earlier this morning. They reflect what's been broadly reported elsewhere: continued growth of Android (and the iPad). Here are the headlines:
Basically, the story is that Android is now a rapidly gaining number two to the iPhone's number one on the Millennial network.
RIM has grown both in terms of representation on the network and in terms of developer involvement, but the news that Android has passed it is more confirmation that the company is in an increasingly challenging competitive position.
And now for the graphics:
A survey of 3,600 US mobile consumers by Sterling Commerce found that 15% of mobile users have made purchases on their phones.This number is broadly consistent with other studies and surveys (see links below) for the US market.
In this survey population slightly less than 50% of respondents reportedly owned smartphones, whereas the number is 23% to 25% in the overall US mobile population. So the results can't be directly extrapolated to the population of mobile users.
More findings (nearly verbatim):
A striking datapoint above is that 60% want store inventory information; while "only" 25% want offers or coupons while shopping. An equal percentage said they performed in-store price comparisons. As barcode scanning becomes more commonplace and easier to use, this latter figure will go way up.
Separately, Insight Express recently found that 82% of survey respondents used their mobile phones while shopping in stores.
These numbers were previously released but I thought it would be interesting to present them together. Below are Nielsen data regarding iPad ownership demographics and Mobclix data about engagement ad ad performance on the iPad.
iPad users appear comparable in age ranges, though perhaps slightly younger, than the iPhone population. They're more affluent than iPhone users however. Nielsen says the "sweet spot" for the iPad is "affluent 25 to 36 year olds."
Interestingly, 51% of iPad users responding to the Nielsen survey did not own an iPhone. I fall into that category as well.
According to Mobclix, rich media/video ads on the iPad have a 10.7X higher CTR than "standard" display on the iPad. In addition, iPad apps generate a 5X eCPM vs. the iPhone.
It's clear that the iPad audience is desirable for brand marketers and that users are much more engaged with media and ads on the iPad than on smartphones or the iPhone.
Nielsen offers some very interesting data on China's mobile market, based on almost 5,000 face-to-face consumer interviews, and compares it to mobile behavior in the US. The finding: a higher percentage of Chinese subscribers use the mobile Internet than their US counterparts.
Just over 37% of China's 755 million mobile phone subscribers in China access the Internet on their handsets. That represents roughly 279 million people overall. In the US there are 77 million mobile Internet users, representing about 27% of subscribers, according to the most recent Nielsen estimates.
The size of the Chinese market is currently about three times larger than the US market. However there's much more growth potential in China vs. the US, which is already a mature market in terms of subscriber penetration. By contrast just over half of China's population has a mobile phone. And China is one of those developing markets where primary Internet access is very quickly likely to be via mobile vs. a PC.
The mobile market in China is almost equally split between men and women and different age groups. Almost 90% (87%) are pre-paid subscribers. Below is the breakdown of mobile activities by category and then a comparison with US mobile user behavior.
Confirming earlier data, the NPD Group said today that Android is outselling the iPhone:
For the first time since the fourth quarter (Q4) of 2007, RIM fell to second position, as Android took the lead among operating systems in handsets sold to U.S. consumers. NPD’s latest wireless market research reveals that Android accounted for 33 percent of all smartphones purchased in Q2, ahead of RIM (28 percent) and Apple (22 percent).
Based on U.S. consumer purchases of mobile phones in Q2, the top 5 Android smartphones were as follows:
- Motorola Droid
- HTC Droid Incredible
- HTC EVO 4G
- HTC Hero
- HTC Droid Eris
Earlier Nielsen reported that in the first half of this year Android handsets had outsold Apple's iPhone. However, in the market as a whole the iPhone still has 2X share of the US smartphone market vs. Android devices.
That share probably goes up if all iOS devices (iPod Touch, iPad) are factored in.
If Android maintains sales momentum like this it will eventually (2011) capture market share and move into second position after RIM, which just introduced a new OS (6) and device (Torch) to blunt the rise of Android and the iPhone. It's not clear, however, it will succeed.
Based on a third party report of unreleased Nielsen mobile marketshare data this morning, I incorrectly described what the data reported. So did almost everybody else.
The original GigaOM discussion of the Nielsen data is technically correct -- it refers to sales rather than overall market share -- but it failed to provide context and all the relevant data. I was also swept up in what seemed like a dramatic 10 point gain in Android's market share.
The headline asserted that Android had surpassed the iPhone in the US. But that's not what the data actually show it turns out. The accurate statement is that in the last six months Nielsen says that Android handsets have taken a greater share of all smartphones sold in the US than the iPhone. However the iPhone still has considerably more share in the US market than Android.
Nielsen later in the day posted with more context. The first chart below is reflective of the overall market: Android has 13%, while the iPhone has 28% share -- more than double.
The chart below is the one that made its appearance this morning and caused the problem. It focuses not on the entire market but only on "recent acquirers." The chart is somewhat misleading because in the absence of the one above it appears to be a snapshot of the overall distribution of handset marketshare (notwithstanding the slide title).
I hadn't seen the first chart until this evening.
What we take away from this is what we already know: Apple has saturated its single carrier market (AT&T), while Android, with multiple handsets and carriers, is gaining momentum. Again, the iPhone's market share is more than double Android's but more recently Android handsets appear to be outselling the iPhone.
The most important datapoint released by Nielsen however is that smartphones now represent 25% of the total US handset market. That's a psychologically significant milestone. The next one is 50%, which Nielsen projects will happen in Q2 next year. I'm skeptical it will be quite that soon but we're definitely on our way.
With multiple handset OEMs and multiple carriers the "Android army" of devices has reportedly overtaken the iPhone in terms of market share in the US smartphone market according to Nielsen (via GigaOm). The data show the following Q2 (survey based) US smartphone market share distribution:
The Nielsen data reflect that Android's share in Q1 was 17%, which would mean an amazing 10 point leap in a single quarter.
We may see some sort of Apple push-back on or attempted spin of this data. For context and comparison purposes, here are Gartner global smartphone data share figures for Q1, 2010:
The Nielsen data, however, continue to show greater loyalty among iPhone owners vs. Android and other handset types:
Android has clearly emerged as a viable alternative to the iPhone in the US, which remains locked up with AT&T so to speak -- a strategic blunder by Apple that has now apparently cost the company some of its credibility and potentially its mindshare in the smartphone segment. While Android continues to be not as "elegant" or polished as the iPhone (it isn't), it's much more widely available and for a large number of people a perfectly acceptable alternative.
Millennial Media released its June SMART report this morning. As usual it's "chock-a-block" full of data. I'll point to a few of the things that interest me. You can get the entire report here.
All data are based on Millennial's network and not necessarily representative of the broader mobile Internet.
The company cites Nielsen data for the proposition that the mobile Internet in the US is now 77 million people. The company asserts that it can reach to 81% of that audience.
Here are Millennial's top DMAs (by ad requests):
The fact that "HARLINGEN, TX" is ahead of "NEW YORK, NY" is suggestive of the idea that the company's network is not synonymous with the mobile Internet per se.
This month's SMART report focuses on the entertainment vertical (one of Millennial's top 3) and its campaign tactics. I'm not going to drill into those numbers. Rather I'll show Millennial's list of the top verticals:
Millennial shows that almost 60% of its campaigns are national, with "broad reach." However of the 41% that are targeted; the top two strategies are geo and demo targeting. When the audiences get large enough these layers will be married for geo + demo + other. But geo can also be a proxy for demographic targeting in the right circumstances.
TechCrunch in an interview with Millennial CEO Paul Palmieri says that the company is planning an IPO in 2011.
Data released today by comScore and the Yellow Pages Association show that local search on mobile devices is growing significantly. The data focus on use of directories and IYPs on mobile (and online). Here are the top-line mobile figures:
Mobile IYP users are more affluent and better educated than the general population:
ComScore utilized "both 1) a unique survey sample of over 10,000 U.S. respondents each month, resulting in a sample in excess of 35,000 when running 3-month average reports and 2) a meter-based smartphone mobile browser measurement tool based on a representative panel of approximately 2,500 devices."
ComScore has always been conservative regarding how it defines a "local search." These data too likely undercount activity online and on mobile devices directed toward offline purchases and other local commercial behavior.
The slide above also reflects activity on IYP and directory sites exclusively and doesn't look at search engines or locally focused apps in the aggregate.
The latest Opera report is out. The focus of the report this month is on Africa and the World Cup. Opera said it had 59.4 million users of the Opera Mini globally. I'm now using it on my Android Evo at least as much as the native browser (which often tries to send me to the carrier portal annoyingly).
Below are the top sites and handsets for China, the UK and US. The iPhone is now the top handset in both the UK and US, while Nokia continues to dominate in the aggregate.
Compare the data from roughly a year ago, July, 2009. While there is some shuffling among the top sites (e.g, MySpace has gone from #3 to #10 in the US) the data reflect mostly the same top 10 in these countries over the past year.
Here are the "big four" in the US:
Total: 264 million (not counting smaller carriers). CTIA says there are 285.6 million total US mobile subsribers.
The gap between the two leaders and the smaller carriers certainly argues in favor of some sort of combination, which has previously been rumored and even supported in principle by Sprint CEO Dan Hesse.
The current state of mobile commerce -- purchases made on mobile devices -- is not dire but most users aren't buying things over their handsets. There are a number of semi-contradictory surveys and studies in the market about the health of m-commerce:
In contrast to buying there is a great deal of "shopping" and price comparison lookups going on in stores on mobile devices. However there are two companies in particular that already have very healthy m-commerce businesses. The first, eBay, said that it was on track for $1.5 billion in sales from mobile devices in 2010. Amazon is the other.
Yesterday Amazon reported second-quarter earnings that included more than $1 billion in global sales from mobile devices:
In the last twelve months, customers around the world have ordered more than $1 billion of products from Amazon using a mobile device," said Jeff Bezos, founder and CEO of Amazon.com.
Are these two companies simply leading indicators of where the market is going or are they exceptions, suggesting that it may be a long time before others achieve comparable mobile success. My view is somewhere in-between.
These are two strong and, in Amazon's case in particular, trusted brands. In addition Amazon (like iTunes) has stored credit card information, removing fricition from the m-commerce process. The credit card and the challenge of entering 16 digits on the small screen is by itself enough to kill 99% of mobile transactions.
Over time consumers will feel more comfortable buying a wide range of things on their handsets, but it will take time for retailers, etailers and others to assimilate best practices and properly address the needs and use case of consumers on the go.