Source: Blog post from Flurry, 20th June 2017
Amazon Alexa in your car & in your life
5 years ago
Showing posts with label correlations. Show all posts
Showing posts with label correlations. Show all posts
Thursday, 22 June 2017
Thursday, 21 May 2015
Friday, 27 February 2015
WeightWatchers 'is losing users to activity trackers'
"Revenue fell 10% to $327.8 million in the fourth quarter, Weight Watchers International said in a statement Thursday, declining for the eighth straight period as FitBit, Jawbone and other activity trackers lure dieters away.
Weight Watchers, founded in 1961, has built up an ecosystem of dieting programs, food products and support centers for people seeking to slim down. With consumers paying more attention to how many calories they're burning from exercise or everyday activities, fitness gadgets have surged in popularity, with 51.2 million American adults using applications to track their health, according to Nielsen. That's making it harder for Weight Watchers to justify subscriptions starting at $20 a month, since activity trackers can be paired with free mobile apps that make it easy to analyze caloric input and output.
"Weight Watchers really has to change what they're offering -- they have to get modern," said Meredith Adler, an analyst at Barclays. "People are just more digital now than they ever were.""
Weight Watchers, founded in 1961, has built up an ecosystem of dieting programs, food products and support centers for people seeking to slim down. With consumers paying more attention to how many calories they're burning from exercise or everyday activities, fitness gadgets have surged in popularity, with 51.2 million American adults using applications to track their health, according to Nielsen. That's making it harder for Weight Watchers to justify subscriptions starting at $20 a month, since activity trackers can be paired with free mobile apps that make it easy to analyze caloric input and output.
"Weight Watchers really has to change what they're offering -- they have to get modern," said Meredith Adler, an analyst at Barclays. "People are just more digital now than they ever were.""
Source: AdAge, 27th February 2015
Note - This is not necessarily causal, or even connected...
Friday, 14 November 2014
Tweets drive time-shifted TV viewing
"Not surprisingly, the most important factor changing how many viewers watch later was an episode’s live audience tune in, accounting for 42% of variance in +7 TV audiences. So, just under half of the difference between the live and +7 audiences can be explained by the size of the original live audience. Conventional wisdom on several other variables also bore out. For example, reality series were 31% more likely to be watched live. A premiere episode, regardless of genre, was 15% more likely to be watched live.
Interestingly, all 11 variables tested proved to be statistically significant. In other words, all measurements and characteristics we looked at affect time-shifted viewing. In fact, this integrated model explains 72% of the variance in the +7 audience, significantly higher than what the live audience could explain alone. Moreover, Nielsen Twitter TV Ratings (NTTR) impressions were significant, even after accounting for the effect of the other 10 variables. Specifically, a 10% increase in NTTR impressions corresponded to a 1.8% increase in the +7 audience, indicating that social media activity around TV programming is playing a role in driving viewers to watch programming later in the week."
Thursday, 2 October 2014
Music drives the sale of hardware including smartphones and tablets
"A study published today by the BPI - the trade body that represents recorded music - has found that the UK obsession with music is not only helping to drive the rapid uptake of consumer technology, it is also generating huge additional demand for the latest smartphones and tablets.
UK consumers spend on average nearly a quarter more per head on music than their counterparts in other G7 countries. This is resulting in billions of Pounds worth of additional expenditure on music-related technology products in the UK - purchases that are well above and beyond the level that would occur if per capita spend was in line with the lower G7 average.
The independent economic study models the impact that an increase in music sales has on technology purchases. It tracks variations in spending patterns between 2008 and 2012 across G7 countries , and demonstrates that during this 5-year period the UK's relatively high consumption of music accounted for an additional £11bn in sales of technology products, broken down as:
£8.4bn additional value in the sales of smartphones;
£2.5bn additional value in the sales of tablets;
£384m additional value in the sales of mp3 players; and
£74m additional value in the sales of Integrated Audio Systems. £11.4bn total
The data confirms that consumers in the UK spend a great deal more on music per capita than in most other developed nations. In 2012 this averaged at nearly 25 per cent more, resulting in proportionately greater sales of consumer technology products. The economic modelling used by the study calculates that for every 1 per cent increase in demand for music there is a corresponding 1.4 per cent lift in sales of smartphones, equivalent to £77.6m, while tablets benefit by a 2.2 per cent rise, translating to £52.6m."
UK consumers spend on average nearly a quarter more per head on music than their counterparts in other G7 countries. This is resulting in billions of Pounds worth of additional expenditure on music-related technology products in the UK - purchases that are well above and beyond the level that would occur if per capita spend was in line with the lower G7 average.
The independent economic study models the impact that an increase in music sales has on technology purchases. It tracks variations in spending patterns between 2008 and 2012 across G7 countries , and demonstrates that during this 5-year period the UK's relatively high consumption of music accounted for an additional £11bn in sales of technology products, broken down as:
£8.4bn additional value in the sales of smartphones;
£2.5bn additional value in the sales of tablets;
£384m additional value in the sales of mp3 players; and
£74m additional value in the sales of Integrated Audio Systems. £11.4bn total
The data confirms that consumers in the UK spend a great deal more on music per capita than in most other developed nations. In 2012 this averaged at nearly 25 per cent more, resulting in proportionately greater sales of consumer technology products. The economic modelling used by the study calculates that for every 1 per cent increase in demand for music there is a corresponding 1.4 per cent lift in sales of smartphones, equivalent to £77.6m, while tablets benefit by a 2.2 per cent rise, translating to £52.6m."
Friday, 26 September 2014
TV Tweet levels broadly correlate with TV channel & viewing figures over time
"Based on detailed examination of a year’s worth of exclusive data from both Twitter and BARB, excluding news and live sports programmes, its key findings include:
There is a ‘Twitter TV Top 30’: In terms of volume of tweets, the top 30 TV series (excluding live sport and news programmes) account for 50% of all measured UK Twitter TV activity and 9% of viewing volume.
TV Tweet levels broadly correlate with TV channel shares and programme/series viewing figures across a broad time period, although some channels over-perform on Twitter relative to audience share.
Of the TV Tweets analysed, there was a noticeable skew towards entertainment, talent shows, constructed reality, documentaries, soaps, special events and some dramas, including Sherlock, Downton Abbey and Doctor Who, where there is a cult or younger following.
Twitter TV activity correlates with audience size at a broad level: the shows with the largest volume of Twitter TV activity tend to have higher audiences. However smaller shows can gain disproportionate levels of Twitter activity if they are ‘social TV friendly’, for example encouraging participation or skewing to a younger audience.
Twitter activity has a direct, positive influence on viewing figures during broadcast for 11% of programmes, boosting audiences by an estimated 2% during those shows."
There is a ‘Twitter TV Top 30’: In terms of volume of tweets, the top 30 TV series (excluding live sport and news programmes) account for 50% of all measured UK Twitter TV activity and 9% of viewing volume.
TV Tweet levels broadly correlate with TV channel shares and programme/series viewing figures across a broad time period, although some channels over-perform on Twitter relative to audience share.
Of the TV Tweets analysed, there was a noticeable skew towards entertainment, talent shows, constructed reality, documentaries, soaps, special events and some dramas, including Sherlock, Downton Abbey and Doctor Who, where there is a cult or younger following.
Twitter TV activity correlates with audience size at a broad level: the shows with the largest volume of Twitter TV activity tend to have higher audiences. However smaller shows can gain disproportionate levels of Twitter activity if they are ‘social TV friendly’, for example encouraging participation or skewing to a younger audience.
Twitter activity has a direct, positive influence on viewing figures during broadcast for 11% of programmes, boosting audiences by an estimated 2% during those shows."
Wednesday, 16 July 2014
Relative crime levels by district correlate with the relative number of Uber rides
"You’ll notice that we do the most San Francisco trips in the downtown and SoMa areas. These also happen to be large, densely populated regions, so that’s to be expected. So in our spatial demand predictions we clearly need to take into account population density.
But there’s a catch. While neighborhood population density might account for some of the variance in our demand, we also need to take into account where people are hanging out, going to work, etc. This is different from census data. Where people live, where people work, and where people play are (usually) in very different neighborhoods in a densely populated city.
So we needed a simple surrogate metric for where people are. We could do that by counting the number of businesses or bars or whatever in a neighborhood… but we had a better idea.
Crime.
We hypothesized that crime would be a proxy for non-residential population density.
According to the data from San Francisco Crimespotting (HUGE shout-out to Stamen Design for the data; you guys are awesome!), there were 75,488 crimes in San Francisco since Uber’s launch on 2010 June 01. These crime data are broken down into 12 categories: murder, robbery, aggravated assault, simple assault, arson, theft, vehicle theft, burglary, vandalism, narcotics, alcohol, and prostitution.
If it looks kind of like the trips map to you, that’s because the two are decently correlated (r = 0.56, p < 0.001). (For you math sticklers, crime and trip data are log distributed by neighborhood, so all correlations are Spearman rank correlations, but log-log Pearson correlations give approximately the same results).
Neighborhoods with more crime (more people hanging out) have more Uber rides.
But we also wanted to know if any specific crimes might be better predictors of rides than others.
To examine this we looked at the correlation between the number of each type of crime and the number of trips we’ve done in each neighborhood. All types of crime except murder, vehicle theft, and arson were positively correlated with number of trips. After correcting for multiple comparisons, four crimes remained significantly correlated (p < 0.05, Bonferroni corrected):
Prostitution
Alcohol
Theft
Burglary
In other words:
The parts of San Francisco that have the most prostitution, alcohol, theft, and burglary also have the most Uber rides! Party hard but be safe, Uberites!"
But there’s a catch. While neighborhood population density might account for some of the variance in our demand, we also need to take into account where people are hanging out, going to work, etc. This is different from census data. Where people live, where people work, and where people play are (usually) in very different neighborhoods in a densely populated city.
So we needed a simple surrogate metric for where people are. We could do that by counting the number of businesses or bars or whatever in a neighborhood… but we had a better idea.
Crime.
We hypothesized that crime would be a proxy for non-residential population density.
According to the data from San Francisco Crimespotting (HUGE shout-out to Stamen Design for the data; you guys are awesome!), there were 75,488 crimes in San Francisco since Uber’s launch on 2010 June 01. These crime data are broken down into 12 categories: murder, robbery, aggravated assault, simple assault, arson, theft, vehicle theft, burglary, vandalism, narcotics, alcohol, and prostitution.
If it looks kind of like the trips map to you, that’s because the two are decently correlated (r = 0.56, p < 0.001). (For you math sticklers, crime and trip data are log distributed by neighborhood, so all correlations are Spearman rank correlations, but log-log Pearson correlations give approximately the same results).
Neighborhoods with more crime (more people hanging out) have more Uber rides.
But we also wanted to know if any specific crimes might be better predictors of rides than others.
To examine this we looked at the correlation between the number of each type of crime and the number of trips we’ve done in each neighborhood. All types of crime except murder, vehicle theft, and arson were positively correlated with number of trips. After correcting for multiple comparisons, four crimes remained significantly correlated (p < 0.05, Bonferroni corrected):
Prostitution
Alcohol
Theft
Burglary
In other words:
The parts of San Francisco that have the most prostitution, alcohol, theft, and burglary also have the most Uber rides! Party hard but be safe, Uberites!"
Source: Uber's blog, 13th July 2014
Note - the key point is that crime levels are a good proxy for non-residential population density
Monday, 14 July 2014
People who are careful with money have fewer car crashes
"The [insurance] industry has started quietly trawling customers’ finances and other sources of “big data”, after proving that those who are careful with their cash have fewer car crashes.
The correlation is so strong that Lloyds Banking Group has begun offering prudent account holders savings of as much as 20 per cent on their car insurance.
Lloyds’ insurance arm Scottish Widows has not established why consumers who stay within overdraft limits or avoid bounced debit card payments have fewer accidents, but insiders suggest the thrifty tend to be more responsible.
The development shows how access to so-called big data are revolutionising the insurance industry. Companies are now accessing a wide range of information on everything from financial probity to shopping habits to determine the risk premium for individual customers.
Tesco, which tracks consumer behaviour through its Clubcard loyalty scheme, offers those it deems less risky based on their shopping habits discounts of as much as 40 per cent on home and car insurance.
Aviva has calculated the optimal distance from the street at which a property is least likely to be burgled – not too secluded nor too exposed.
Elsewhere, executives at Aviva’s Canadian business have established that houses within a radius of a few hundred metres from cinemas are more likely to be vandalised.
“This is not just a backroom theoretical exercise,” said Maurice Tulloch, who runs Aviva’s UK’s general insurance business. “If our analyst finds something new on a Monday, that can be live and impacting our prices on the Tuesday.”
Simon Douglas, director of AA Insurance, said: “The winners in the insurance market will be the ones that have got the data insights that others don’t have. It could be supermarkets, banks or social media companies.”"
Source: Financial Times, 12th July 2014
The correlation is so strong that Lloyds Banking Group has begun offering prudent account holders savings of as much as 20 per cent on their car insurance.
Lloyds’ insurance arm Scottish Widows has not established why consumers who stay within overdraft limits or avoid bounced debit card payments have fewer accidents, but insiders suggest the thrifty tend to be more responsible.
The development shows how access to so-called big data are revolutionising the insurance industry. Companies are now accessing a wide range of information on everything from financial probity to shopping habits to determine the risk premium for individual customers.
Tesco, which tracks consumer behaviour through its Clubcard loyalty scheme, offers those it deems less risky based on their shopping habits discounts of as much as 40 per cent on home and car insurance.
Aviva has calculated the optimal distance from the street at which a property is least likely to be burgled – not too secluded nor too exposed.
Elsewhere, executives at Aviva’s Canadian business have established that houses within a radius of a few hundred metres from cinemas are more likely to be vandalised.
“This is not just a backroom theoretical exercise,” said Maurice Tulloch, who runs Aviva’s UK’s general insurance business. “If our analyst finds something new on a Monday, that can be live and impacting our prices on the Tuesday.”
Simon Douglas, director of AA Insurance, said: “The winners in the insurance market will be the ones that have got the data insights that others don’t have. It could be supermarkets, banks or social media companies.”"
Source: Financial Times, 12th July 2014
Wednesday, 9 April 2014
'The less Americans know about Ukraine’s location, the more they want U.S. to intervene'
"On March 28-31, 2014, we asked a national sample of 2,066 Americans (fielded via Survey Sampling International Inc. (SSI), what action they wanted the U.S. to take in Ukraine, but with a twist: In addition to measuring standard demographic characteristics and general foreign policy attitudes, we also asked our survey respondents to locate Ukraine on a map as part of a larger, ongoing project to study foreign policy knowledge. We wanted to see where Americans think Ukraine is and to learn if this knowledge (or lack thereof) is related to their foreign policy views. We found that only one out of six Americans can find Ukraine on a map, and that this lack of knowledge is related to preferences: The farther their guesses were from Ukraine’s actual location, the more they wanted the U.S. to intervene with military force."
Source: Washington Post, 7th April 2014
Source: Washington Post, 7th April 2014
Friday, 14 March 2014
“If it rains in New York, people around the country become miserable”
"For about 1.23 billion people on this planet, Facebook is there, in sickness and in health, in good times and in bad, in joy as well as in sorrow. But venting to Facebook creates a significant emotional ripple effect: In the social network, keeping your feelings to yourself is not an option, and curiously, positive vibes are more contagious than negative ones, according to a new study.
To measure how the emotional content of a person’s Facebook status updates might affect other users, the group of researchers (including two Facebook employees)--who published today in PLoS One--collected anonymized status updates from the 100 most populated cities in the U.S. between 2009 and early 2012. They then ran them through a software program called the Linguistic Inquiry and Word Count, a generally reliable auditor of basic feelings.
To measure how the emotional content of a person’s Facebook status updates might affect other users, the group of researchers (including two Facebook employees)--who published today in PLoS One--collected anonymized status updates from the 100 most populated cities in the U.S. between 2009 and early 2012. They then ran them through a software program called the Linguistic Inquiry and Word Count, a generally reliable auditor of basic feelings.
Rainy days, somewhat unsurprisingly, created an outsized emotional effect: Rain increased the number of negative posts by 1.16%, and decreased the number of positive posts by 1.19%. That observation set the stage for a natural experiment: How far would emotions motivated by rain spread through the social network? If New York City’s rain-induced pathos could affect users in New Mexico, it would say something remarkable about the power of online emotional contagion.
And that’s precisely what the researchers found. “If it rains in New York, people around the country become miserable,” says study co-author Nicholas Christakis, professor of sociology and medicine at Yale University. This works in every direction, though given New York's population, it has an outsized influence. But weirdly enough, it wasn’t the negative posts that carried the most viral ability: Each positive post yielded an additional 1.75 positive posts, whereas negative posts only spread an additional 1.29."
Source: FastCoExist 12th March 2014
Tuesday, 18 February 2014
Tracking love on Facebook
 |
| Click to enlarge |
"During the 100 days before the relationship starts, we observe a slow but steady increase in the number of timeline posts shared between the future couple. When the relationship starts ("day 0"), posts begin to decrease. We observe a peak of 1.67 posts per day 12 days before the relationship begins, and a lowest point of 1.53 posts per day 85 days into the relationship. Presumably, couples decide to spend more time together, courtship is off, and online interactions give way to more interactions in the physical world."
Wednesday, 15 January 2014
Friday, 29 November 2013
Monday, 12 August 2013
Twitter activity drives TV viewing, and vice versa
"Today Nielsen released findings, which, for the first time, provide statistical evidence of a two-way causal influence between broadcast TV tune-in for a program and the Twitter conversation around that program. Nielsen’s Twitter Causation Study included time series analysis to determine if Twitter activity drives increased tune-in rates for broadcast TV and if broadcast TV tune-in leads to increased Twitter activity. This latest study follows research released earlier this year that quantified the correlation between TV ratings and Twitter.
Analyzing minute-to-minute trends in Nielsen’s Live TV Ratings and Tweets for 221 broadcast primetime program episodes using Nielsen’s SocialGuide, the findings show that Live TV ratings had a statistically significant impact in related Tweets among 48 percent of the episodes sampled, and that the volume of Tweets caused statistically significant changes in Live TV Ratings among 29 percent of the episodes. The time series analysis methodology used for this study was developed by Nobel Prize-winning economist Clive Granger, and is widely used in the fields of econometrics, physics, and neuroscience, among others.
“Using time series analysis, we saw a statistically significant causal influence indicating that a spike in TV ratings can increase the volume of Tweets, and, conversely, a spike in Tweets can increase tune-in,” said Paul Donato, Chief Research Officer, Nielsen. “This rigorous, research-based approach provides our clients and the media industry as a whole with a better understanding of the interplay between Twitter and broadcast TV viewing.”"
Source: Press release from Nielsen, 6th August 2013
Analyzing minute-to-minute trends in Nielsen’s Live TV Ratings and Tweets for 221 broadcast primetime program episodes using Nielsen’s SocialGuide, the findings show that Live TV ratings had a statistically significant impact in related Tweets among 48 percent of the episodes sampled, and that the volume of Tweets caused statistically significant changes in Live TV Ratings among 29 percent of the episodes. The time series analysis methodology used for this study was developed by Nobel Prize-winning economist Clive Granger, and is widely used in the fields of econometrics, physics, and neuroscience, among others.
“Using time series analysis, we saw a statistically significant causal influence indicating that a spike in TV ratings can increase the volume of Tweets, and, conversely, a spike in Tweets can increase tune-in,” said Paul Donato, Chief Research Officer, Nielsen. “This rigorous, research-based approach provides our clients and the media industry as a whole with a better understanding of the interplay between Twitter and broadcast TV viewing.”"
Source: Press release from Nielsen, 6th August 2013
Monday, 20 May 2013
The inverse relationship between Netflix & AOL subscribers
Click to enlarge
Source: Business Insider, 9th May 2013
Note - US only, Yellow bars show AOL dial-up subscribers
Monday, 13 May 2013
There is a correlation between page views on Wikipedia and stock market performance
"A flood of views to a company's Wikipedia page may be a sign that their stock price is about to plummet.
That's the implication of a study published on 8 May in Scientific Reports, which looked at Wikipedia page view data from 2007 to 2012 and correlated it with changes in the stock market price for companies listed on the Dow Jones Industrial Average.
An uptick in the number of Wikipedia page views, as compared with the average weekly views, was followed by a fall in a company's share price.
The research suggests that Wikipedia pages, which can be edited by anyone, are part of the information-gathering process for financial transactions -- potentially useful knowledge for people wanting to play the system.
"We were really intrigued by the idea that data from usage of Internet information resources such as Wikipedia might help us understand how traders gather information before making these decisions," lead author Suzy Moat, Senior Research Fellow at Warwick Business School, told Wired.co.uk.
"The connection we find between views of Wikipedia pages and stock market moves suggests this may indeed be the case. This suggests that data on people's usage of online information services, such as Wikipedia or Google may be used to anticipate decisions they might later take in the real world."
Previous research has shown that the volume of financially-related searches on Google and also the "mood" of tweets on Twitter can be linked to changes in the stock market. Big data approaches to real world trends have also shown links between web searches and flu infections."
That's the implication of a study published on 8 May in Scientific Reports, which looked at Wikipedia page view data from 2007 to 2012 and correlated it with changes in the stock market price for companies listed on the Dow Jones Industrial Average.
An uptick in the number of Wikipedia page views, as compared with the average weekly views, was followed by a fall in a company's share price.
The research suggests that Wikipedia pages, which can be edited by anyone, are part of the information-gathering process for financial transactions -- potentially useful knowledge for people wanting to play the system.
"We were really intrigued by the idea that data from usage of Internet information resources such as Wikipedia might help us understand how traders gather information before making these decisions," lead author Suzy Moat, Senior Research Fellow at Warwick Business School, told Wired.co.uk.
"The connection we find between views of Wikipedia pages and stock market moves suggests this may indeed be the case. This suggests that data on people's usage of online information services, such as Wikipedia or Google may be used to anticipate decisions they might later take in the real world."
Previous research has shown that the volume of financially-related searches on Google and also the "mood" of tweets on Twitter can be linked to changes in the stock market. Big data approaches to real world trends have also shown links between web searches and flu infections."
Source: Wired, 8th May 2013
Friday, 22 March 2013
Twitter is one of three statistically significant variables that correlate with TV ratings
"U.S. TV viewers are taking to Twitter to talk about TV, and the digital chatter is building steam. According to SocialGuide, 32 million unique people in the U.S. Tweeted about TV in 2012. That’s quite the confab, but what does it all really mean for the TV industry? Should networks and advertisers be paying attention? Early research on the subject from Nielsen and SocialGuide says yes.
By analyzing Tweets about live TV, the study confirmed a relationship between Twitter and TV ratings. It also identified Twitter as one of three statistically significant variables (in addition to prior-year rating and advertising spend) to align with TV ratings.
“While prior-year rating accounts for the lion’s share of the variability in TV ratings, Twitter’s presence as a top three influencer tells us that Tweeting about live TV may affect program engagement,” said Andrew Somosi, CEO of SocialGuide. “We expected to see a correlation between Twitter and TV ratings, but this study quantifies the strength of that relationship.”
Much of the correlation is being driven by the rise in media consumption across multiple device screens. We know that 80% of U.S. tablet and smartphone owners who watch TV use their device while watching at least several times a month. We also know that 40% of U.S. tablet and smartphone users visit a social network while watching TV.
How well does Twitter align with TV program ratings? The recent Nielsen/SocialGuide study confirmed that increases in Twitter volume correlate to increases in TV ratings for varying age groups, revealing a stronger correlation for younger audiences. Specifically, the study found that for 18-34 year olds, an 8.5% increase in Twitter volume corresponds to a 1% increase in TV ratings for premiere episodes, and a 4.2% increase in Twitter volume corresponds with a 1% increase in ratings for midseason episodes. Additionally, a 14.0% increase in Twitter volume is associated with a 1% increase in TV program ratings for 35-49 year olds, reflecting a stronger relationship between Twitter and TV for younger audiences.
Further, the study found that the correlation between Tweets and TV ratings strengthens for midseason episodes for both age groups. An increase in Twitter volume of 4.2% and 8.4% is associated with a 1% increase in ratings for 18-34 year olds and 35-49 year olds, respectively. Moreover, by midseason Twitter was responsible for more of the variance in ratings for 18-34 year olds than advertising spend.
“The TV industry is dynamic and it was important for us to analyze multiple variables to truly understand Twitter’s impact on TV ratings,” said Mike Hess, Executive Vice President of Media Analytics for Nielsen. “While our study doesn’t prove causality, the correlation we uncovered is significant and we will continue our research to deepen the industry’s understanding of this relationship.”"
By analyzing Tweets about live TV, the study confirmed a relationship between Twitter and TV ratings. It also identified Twitter as one of three statistically significant variables (in addition to prior-year rating and advertising spend) to align with TV ratings.
“While prior-year rating accounts for the lion’s share of the variability in TV ratings, Twitter’s presence as a top three influencer tells us that Tweeting about live TV may affect program engagement,” said Andrew Somosi, CEO of SocialGuide. “We expected to see a correlation between Twitter and TV ratings, but this study quantifies the strength of that relationship.”
Much of the correlation is being driven by the rise in media consumption across multiple device screens. We know that 80% of U.S. tablet and smartphone owners who watch TV use their device while watching at least several times a month. We also know that 40% of U.S. tablet and smartphone users visit a social network while watching TV.
How well does Twitter align with TV program ratings? The recent Nielsen/SocialGuide study confirmed that increases in Twitter volume correlate to increases in TV ratings for varying age groups, revealing a stronger correlation for younger audiences. Specifically, the study found that for 18-34 year olds, an 8.5% increase in Twitter volume corresponds to a 1% increase in TV ratings for premiere episodes, and a 4.2% increase in Twitter volume corresponds with a 1% increase in ratings for midseason episodes. Additionally, a 14.0% increase in Twitter volume is associated with a 1% increase in TV program ratings for 35-49 year olds, reflecting a stronger relationship between Twitter and TV for younger audiences.
Further, the study found that the correlation between Tweets and TV ratings strengthens for midseason episodes for both age groups. An increase in Twitter volume of 4.2% and 8.4% is associated with a 1% increase in ratings for 18-34 year olds and 35-49 year olds, respectively. Moreover, by midseason Twitter was responsible for more of the variance in ratings for 18-34 year olds than advertising spend.
“The TV industry is dynamic and it was important for us to analyze multiple variables to truly understand Twitter’s impact on TV ratings,” said Mike Hess, Executive Vice President of Media Analytics for Nielsen. “While our study doesn’t prove causality, the correlation we uncovered is significant and we will continue our research to deepen the industry’s understanding of this relationship.”"
Tuesday, 12 March 2013
There is little connection between download volumes and mobile game revenues
"Something fascinating has happened in the app industry over the past two years; download volume performance has decoupled from revenue performance almost entirely. A few years ago, the original Angry Birds spent 22 months in the Top 20 chart of biggest revenue generating apps in America. The latest Angry Birds game struggled to stay 2 months in the Top 20.
The app industry revenue generation is now utterly dominated by free downloads that lure consumers into paying for in-game features month after month. The top-grossing iPhone app in America, Clash of Clans, is merely #70 on the download chart. Ironically enough, this emblem of the new era of mobile gaming was created by Supercell, a Finnish company now located in the old Nokia headquarters. Just 50 feet from the Rovio HQ.
The extremes in the mobile app market are only growing more pronounced. Rage of Bahamut, an addictive card battle game, no longer makes it to the Top 1’000 of iPhone apps. Yet it remains the #14 app in America when it comes to revenue generation, far above the new #1 download app, Temple Run: Oz. Very, very few people download the Rage of Bahamut. But those that do end up enslaved by it. Industry rumors peg the daily average revenue per active user of this game to be as high as 80 to 90 cents. That is an amount of money that translates a free download to a massive cash cow even if it gets only 0.5% of the Angry Birds download base."
Source: Forbes, 6th March 2013
The app industry revenue generation is now utterly dominated by free downloads that lure consumers into paying for in-game features month after month. The top-grossing iPhone app in America, Clash of Clans, is merely #70 on the download chart. Ironically enough, this emblem of the new era of mobile gaming was created by Supercell, a Finnish company now located in the old Nokia headquarters. Just 50 feet from the Rovio HQ.
The extremes in the mobile app market are only growing more pronounced. Rage of Bahamut, an addictive card battle game, no longer makes it to the Top 1’000 of iPhone apps. Yet it remains the #14 app in America when it comes to revenue generation, far above the new #1 download app, Temple Run: Oz. Very, very few people download the Rage of Bahamut. But those that do end up enslaved by it. Industry rumors peg the daily average revenue per active user of this game to be as high as 80 to 90 cents. That is an amount of money that translates a free download to a massive cash cow even if it gets only 0.5% of the Angry Birds download base."
Source: Forbes, 6th March 2013
Monday, 11 March 2013
The shutdown of Megaupload caused an increase in digital movie sales and rentals
"This week researchers from Wellesley College and Carnegie Mellon University released a comprehensive study that evaluates the impact of Megaupload’s shutdown on digital movie revenues.
Titled “Gone in 60 Seconds: The Impact of the Megaupload Shutdown on Movie Sales,” the paper compared digital movie revenues across 12 countries. These countries vary in the relative number of Megaupload users, allowing the researchers to estimate the effect of the cyberlocker’s demise on movie sales.
“We were interested in studying the effect of a major piracy site shutdown on demand for digital movie sales since we’ve seen the argument that such efforts could be like a game of whack-a-mole, with a new file-sharing site springing up as soon as one is closed,” assistant professor of Economics Brett Danaher tells TorrentFreak.
“We saw the logic of this argument, but could also imagine a world where shutting down such a large site could change the behavior of some types of consumers,” he adds.
After controlling for a wide range of country-specific trends and other variables the researchers conclude that the latter is the case, Megaupload’s shutdown had a significant effect on digital revenues. The data suggest that the income of two major Hollywood studios was boosted by up to 10 percent.
“Our analysis across 12 countries suggests that, in the 18 weeks following the shutdown, digital revenues for these two studio’s movies were 6-10% higher than they would have been if not for the shutdown,” the researchers write in their paper.
The table [click here to see the table] shows that Megaupload “penetration” was relatively high in Spain and France, where 17% and 11% of Internet subscribers used the site. With less than 2% it was least popular in the United States.
The researchers used these differences for their statistical model and found that movie revenues were affected positively in countries with a high Megaupload penetration.
“For each additional 1% pre-shutdown Megaupload penetration, the post-shutdown sales unit change was 2.5% to 3.8% higher, suggesting that these increases are a causal effect of the shutdown,” they write.
The shutdown of Megaupload caused a 7-10% increase in the number of digital sales and a 4-7% increase in digital rentals.
The results are based on sales numbers reported by two major Hollywood studios, and the researchers suggest that the effect will be similar for other film companies. Whether the effect will remain over time has yet to be seen though."
Titled “Gone in 60 Seconds: The Impact of the Megaupload Shutdown on Movie Sales,” the paper compared digital movie revenues across 12 countries. These countries vary in the relative number of Megaupload users, allowing the researchers to estimate the effect of the cyberlocker’s demise on movie sales.
“We were interested in studying the effect of a major piracy site shutdown on demand for digital movie sales since we’ve seen the argument that such efforts could be like a game of whack-a-mole, with a new file-sharing site springing up as soon as one is closed,” assistant professor of Economics Brett Danaher tells TorrentFreak.
“We saw the logic of this argument, but could also imagine a world where shutting down such a large site could change the behavior of some types of consumers,” he adds.
After controlling for a wide range of country-specific trends and other variables the researchers conclude that the latter is the case, Megaupload’s shutdown had a significant effect on digital revenues. The data suggest that the income of two major Hollywood studios was boosted by up to 10 percent.
“Our analysis across 12 countries suggests that, in the 18 weeks following the shutdown, digital revenues for these two studio’s movies were 6-10% higher than they would have been if not for the shutdown,” the researchers write in their paper.
The table [click here to see the table] shows that Megaupload “penetration” was relatively high in Spain and France, where 17% and 11% of Internet subscribers used the site. With less than 2% it was least popular in the United States.
The researchers used these differences for their statistical model and found that movie revenues were affected positively in countries with a high Megaupload penetration.
“For each additional 1% pre-shutdown Megaupload penetration, the post-shutdown sales unit change was 2.5% to 3.8% higher, suggesting that these increases are a causal effect of the shutdown,” they write.
The shutdown of Megaupload caused a 7-10% increase in the number of digital sales and a 4-7% increase in digital rentals.
The results are based on sales numbers reported by two major Hollywood studios, and the researchers suggest that the effect will be similar for other film companies. Whether the effect will remain over time has yet to be seen though."
Source: TorrentFreak, 7th March 2013
Subscribe to:
Posts (Atom)