When Mad Men Meets Flash Boys

Written on May 25th, 2014

To download a PDF version, please click on the following link:
When Mad Men Meets Flash Boys

Advertising used to be simple business. My newspaper has a million readers. My TV program has a million viewers. Now let’s negotiate what it’s worth to you Mr. Brand to put your ad in front of my audience. The web made this dynamic a bit more complicated, but early on the process was quite similar. My website gets a million hits per month and now let’s figure out what a banner ad is worth to you. The story got more complicated rather quickly as a bunch of genius coders realized that advertising is actually the perfect domain for technology because tech is capable of bringing way more measurement and accountability than ad buyers had ever seen before. It’s not a coincidence that all of those high flying tech stocks you follow are really advertising companies – think Google, Facebook, Twitter, Yahoo, Baidu. Online advertising is a $100 billion global market, or about 10x larger than it was in the early 2000s. This segment has risen from nowhere to become second only to television advertising but growing much more quickly.

What we find compelling about advertising technology is that it continues to transform in very rapid fashion. The complexity factor took another major leap about five years ago with the rise of something called programmatic buying and real time bidding (RTB). The concept here is that every single website impression creates a live auction. From the time I click my mouse to the time that the web page opens up (a few milliseconds), computers have figured out generally who I am, what I like to browse, what purchases I’ve made online, my gender, etc. This data has been delivered to advertisers and those advertisers have bid for the right to place their ad. The winner of the auction has been selected and the ad is now sitting on my web page. And once again, all of this took place over the span of 10 milliseconds a few billion times in the past hour. The technology used to make all of this happen is very cool stuff – reminiscent of the high frequency trading platforms popularized in Michael Lewis’ new book.


The overall industry is growing rapidly, well into the double digits and there are especially lucrative pockets of growth in that mix, which are growing more than 2x year over year. These include mobile, video, and social. We at Difference Capital are particularly keen on the ad-tech vertical given the tremendous growth potential. A few of our portfolio companies that touch this theme are: Appinions, EQ, iPowow and Vision Critical.

Here are a few topics we will be exploring in this report:
1. Market segmentation and growth forecasts
2. Why do ad-tech companies overcomplicate things – understanding the metrics of ad-tech and why we don’t have something similar to TV ratings to gauge the success of online channels
3. The rise of programmatic and real time bidding – nearly 80% of marketers have never used programmatic, yet the market is exploding
4. Combating click fraud – new metrics and the rise of premium programmatic
5. Making sense of the ecosystem – why does the website only receive 20% of the revenue from an ad purchase?
6. What does a real time bidding auction look like?
7. Best of breed vs. suite – how Google owns the stack
8. Mobile, Video and Social – similarities and differences in the ad-tech model
9. Examples of failures and successes in ad campaigns

1.0 Market Segmentation

The US advertising market is worth roughly $175 billion annually. Television remains the dominant media for this spend, representing about $75 billion or 43% of the total. This segment has been the largest for some time and has shown relatively flattish growth. The Internet segment is second, worth $43 billion last year or 24% of the total. This category has been growing quickly over the past decade. Internet has grown at an 18% CAGR since 2004. Within this category, mobile has been the strongest performer over the past few years. Mobile now represents 17% of total Internet spend or roughly $7 billion. This figure was below $2 billion in 2011 and virtually zero in 2009.

AdTech2 AdTech3

Within Internet advertising, the search category remains by far the largest with 43% share last year. These are basically your standard Google-type searches on the web. The display/banner segment was second at 19%, followed by mobile at 17%. Search has been relatively flattish over the past 5 years but slipped a bit in 2013. Display/banner has also been flat to down recently. The major growth is coming from mobile and video. Mobile is more than doubling year over year and actually took the second place position from display/banner in the fourth quarter.

AdTech4 AdTech5

2. Why do Ad-Tech Companies Overcomplicate Things?

Despite the rapid growth and massive size of Internet advertising, we contend that the market is actually being stifled by what should be an enormous asset. Technology is capable of giving advertising much more accountability than the industry has ever seen before. And this is probably the biggest reason why companies like Google have become global titans in relatively short timeframes. We would argue that one of the less acknowledged talents that Google possesses is simplicity. Google spends $15 billion per year on R&D and capex, yet the company’s message is quite simple. Bid on the right to place an ad when somebody conducts a certain search. I.e. I’m willing to pay 50 cents every time somebody clicks onto my ad and goes to my website.

The newest wave of techies could learn something from Google in our opinion. These new programmatic and RTB companies have over-complicated the ad business. Global advertisers spend $200 billion on TV commercials and don’t seem to care a whole lot about measuring ROI. The key driver for TV ad spend in the US is ratings and the proxy for ratings is a metric called gross rating point (GRP). GRP is defined as what portion of the audience saw an ad multiplied by how many times they saw it. So for example, if a commercial played 3 times during a show and 20% of the audience was watching that show, the total GRP would be 3 x 20 = 60. The TV networks tend to charge advertisers about $50K per GRP. Pretty simple stuff, right? So where do things go wrong? Firstly, the TV ratings are sampled from what people from about 20,000 homes in the US are watching. Of the 110 million or so homes in the US, this translates into a sample size of about 0.02%. And what about those viewers who prefer to watch their content online or via streaming service like Netflix? And is there a mechanism to track how many people who watched that Nike commercial went out to purchase that pair of shoes? No way.

Despite these limitations, advertisers appear perfectly content spending lots and lots of money on TV commercials. Why is online advertising held to a much higher standard? The easy answer is inertia. TV advertising first began in the 1940s and really entered the mainstream in the 1960s. It has been the easy and safe place to park your money over the past 50 years. The more nuanced answer is that ad-tech is too complicated. Techies love numbers and love measuring things. Advertisers are constantly telling us they get so much data, they don’t know what to do with it all. While it may be interesting that our banner ad got 3x more clicks on Tuesday morning in the state of Utah, is it relevant and what can I do with this data?

Online advertising allows ad-tech companies to introduce dozens of fancy acronyms in order to impress the marketing buyers. Unique visitors, impressions, SEO, SEM, CTR, CPM, DSP, RTB, etc. The problem is that the advertiser, newly empowered with these new acronyms, now wants to translate CTR and CPM into the most important metric of all, ROI. Unfortunately, this is very tricky. Just because I didn’t click on the ad, doesn’t mean it didn’t capture my attention and get me thinking about that tasty chocolate dip donut at Tim Hortons. After seeing 5 different ads about the new Godzilla movie coming out on my phone, tablet, computer and TV, I went to Google to check out the trailer. Who takes credit for converting me to watch that ad? Ad-tech companies have become their own worst enemies by constantly raising the measurement bar in order to differentiate themselves. Ad-tech company A says I can get you access to 10 million people who like sports and 0.1% will click your ad. Ad-tech company B now has to get you access to 10 million males who like baseball and 0.15% will click your ad. And the arms race grows. One of the last big ad-tech IPOs, Rocket Fuel, claims to have developed an “artificial Intelligence and Big Data-driven predictive modeling and automated decision-making platform”. If that mouthful of words doesn’t excite the techies, nothing will.

Why not introduce a concept like GRP for digital media? Our website gets 10 million unique visitors per month and those visitors tend to be male, between the ages of 25-34. That ad was seen by 1 million visitors so 10% of the total and it was shown an average of 10 times in the month. Therefore, the total GRP is 100. Let’s charge $50K per GRP and we, the website, starts making $5 million in ad revenue per month. Is this too simple to work? Perhaps. But there is some precedent. In fact, Google recently announced a partnership with comScore to launch the concept of “active GRPs” in display advertising. The purpose is to make adjustments to online data in order to make this metric comparable to standard GRP and thereby have an apples-to-apples way to compare online with offline campaigns.

Also, the Media Rating Council (MRC) in March this year announced that it has released some guidelines around what constitutes a valid ad impression. The Viewable Impression Measurement Guidelines specify that 50% of pixels must be in the viewable portion of the browser for a minimum of one continuous second to qualify as having been viewed.

We think the ad-tech industry is full of data and advertisers are getting confused by what it all means and what is truly relevant. Any ideas like GRP and viewability guidelines, which drive toward more simplicity in ad-tech, will be useful in promoting wider adoption of these technologies. Digital advertising has been growing quickly but still lags behind where it should be. Advertisers in the US still spend more ad dollars on print than they do on Internet despite the fact that people spend 4x more time on the web than they do reading newspapers and/or magazines.


3.0 Advertising Becomes Hi-Tech

Most forms of advertising, including Internet, have been relatively low tech. Ie. You as the TV show, newspaper or website share your traffic figures with a brand and then negotiate how much you will charge that brand for the right to place its content in front of your viewers. The Internet advertising model has changed rather dramatically over the past few years with the emergence of real time bidding and programmatic buying – terms that tend to be used interchangeably even though they are not the same thing. The concept is relatively simple even though the execution is quite complex. These technologies allow brands to bid on users one at a time based on the information about those users. Knowing what a web viewer likes, his browsing history, demographic, etc. will presumably lead certain brands to bid more for the right to advertise. One cannot help but be impressed when considering all that needs to happen to get that targeted ad in front of your eyeballs between the time that you click your mouse to the time that the website opens up.

The advantages of the programmatic approach are quite obvious. This has led to dramatic growth of programmatic buying and real time bidding in the market. The share of programmatic has risen from 24% in 2011 to 53% in 2013. Global programmatic buying represents $12 billion in 2013, according to Magna Global. This is expected to grow to $33 billion by 2017. Of this total, US real time bidding is a $3.9 billion market, expected to grow to over $10 billion by 2017.


What is perhaps most impressive about this growth is that advertisers are still somewhat perplexed by the concept of programmatic buying. A recent survey of US client-side marketers in March 2014 showed that only 23% of respondents understood programmatic buying and have used it to execute a campaign. 77% have never used programmatic and over 40% do not even know what the term means.


4.0 Fraud and the Push for Premium Programmatic

Since the inception of pay-per-click (PPC) advertising, click fraud has emerged as a real burden to advertisers and advertising networks alike. Costing advertisers billions of dollars per year, click fraud stands to undermine the entire digital media industry. It was reported earlier this year that up to 36% of traffic is generated from machines, not humans. The goals of this machine traffic is often to create fake clicks in order for publishers to earn advertising revenue. This can place a heavy toll on brands that need to pay for this illegitimate traffic. Both the publisher and the ad network are incentivized to drive click rates as they can lead to revenue streams as high as a $40 CPM (or $0.04 per click). Whether the players are automated robots (“bots”) or cheap offshore labour, the end result is the same. The clicks drive the advertising spend through the roof, which undermines the industry.

The industry has been responding to the click fraud threat through the adoption of various detection and prevention technologies. These can include things like blacklists (which sites to avoid) or whitelists (which sites to accept) or more elaborate forms of real-time bot detection. The result has been a significant decrease in bot traffic over the past few quarters. According to Integral Ad Science, the portion of suspicious activity on ad exchanges has dropped from 20% in the first half of 2013 to 13% in the second half. The portion of suspicious activity in direct ad sales is still well below at 2%. Viewability is another important metric as discussed above. The proportion of ads viewable for longer than 1 second is between 44% to 56% on average. A third metric that matters to advertisers is brand safety or how often ads gets placed on questionable websites (i.e. Adult content, illegal downloads, drugs/alcohol, etc.) Direct publishers continue to be the least risky according to this metric. In Q4/13, 2.8% of impressions were placed alongside objectionable content according to Integral Ad Science. The figure for ad exchanges was 7.2%.

Questions around click fraud and brand safety have led to the rise of something called premium programmatic ad spending. Programmatic and real time bidding began as a way to get remnant publishing inventory sold so that websites could increase their fill rates and monetize on all available traffic. The industry is now looking for a hybrid way to marry the best of both worlds – the speed and efficiency of programmatic combined with the safety and quality of direct ad sales. The term premium programmatic has emerged to fill this gap. Spending in this area is expected to grow exponentially in the coming years. eMarketer forecasts this segment to grow from $200 million last year to $1.6 billion next year and nearly $6 billion by 2017.


5.0 Making Sense of the Ecosystem

How hard can it be to put that Big Mac logo on my web page as I’m perusing through the NBA playoff scores before I get to work? It’s probably trickier than you would think, at least judging by all of the parties in the ecosystem. An industry consultant a few years ago estimated that out of a $5 ad purchase, about $1 may go to the publisher or the website that attracted my eyeballs to the screen. The rest of the $4 gets split between the agency, ad network, exchange, data provider and various other intermediaries. We offer a glimpse of this ecosystem below and attempt to explain what everybody does in plain old English.


Advertiser – The company that wants to put their brand in front of you so that you buy their stuff. Advertisers can work through agencies to place their ads or can circumvent the agency for the execution of the ad.

Agency – The marketing geniuses that come up with campaigns to most effectively differentiate your stuff from other stuff in the marketplace.

DSP (Demand Side Platform) – For an advertiser, good eyeballs are in demand. Brands demand eyeballs and websites (publishers) supply those eyeballs. The demand side platform (DSP) helps advertisers purchase eyeballs in a centralized fashion with the right business rules. Not all eyeballs are created equal. Eyeballs attached to big wallets tend to be more coveted than eyeballs attached to small wallets. It would be overly cumbersome for advertisers to purchase eyeballs from websites directly and so they run through DSPs. DSPs fill two primary roles. Firstly, they offer a centralized place for advertisers to purchase this virtual shelf-space. Secondly, they allow advertisers to set the rules of what types of eyeballs they want to target. In effect, an advertiser like Ford could ask its DSP for 1 million people to view its new truck ad. Ford wants to deliver this ad to males, ages 35-45, with annual income above $60K, who have shown some interest in vehicles over the past 30 days. The DSP translates this request into an algorithm and then scours a list of hundreds of websites for people that fit the profile. EQ is a DCF portfolio company that fits into this category. The company helps brands and agencies reach their target audience with a technology platform that connects into multiple exchanges into the PC, mobile, video and social segments. Other Canadian companies that fit this category are AcuityAds, adgear, Chango, and eyeReturn.

Ad Exchange – This is where the trade happens – where advertisers bid on eyeballs and the winning bidder gets to put his ad on your desktop or phone. The main purpose of the exchange is to facilitate an automated auction-based process in real time. In order to establish an efficient auction, the bidders need good data on who is looking at the web page. The publisher will typically try to provide as much of this data as possible based on cookies and other sources of information. Third party data providers will also try to supplement this data with their own proprietary sources. The auction process is described in a bit more detail below.

Ad Network – these were created to aggregate publishing inventory in order to make the process of buying for an advertiser easier. An ad network would negotiate with multiple publishers (websites) in order to be able to bundle different types of eyeballs together. An ad network that has relationships with 50 top websites is able to more granularly slice the inventory by demographic, geography, etc. The ad network can then purchase a large campaign and be confident that it can place the ads in front of the right eyeballs.

SSP (Supply Side Platform) – these are similar to DSPs except they are used to help publishers or websites manage their inventory. Publishers would use SSPs to optimize their selling prices in real time in order to effectively balance revenue maximization with fill rates (ie. How many opportunities for an ad actually get placed with an ad?)

Publishers – these are the websites that attract the eyeballs.

6.0 Visualizing the Auction Process

What does the auction process look like in an ad exchange? We offer a pictorial depiction below. Step 1 has a user going to a website. Let’s assume ESPN in this case. The website publisher typically collects information on the user and his surfing habits based on cookies and other third party sources. This data is delivered to the DSPs in step 2. The DSPs have been programmed by advertisers and agencies to bid on users that match a certain profile. Step 3 has the DSPs evaluating the criteria and bidding on the right to place their ad through the ad exchange. In our example below, Bidder A is representing Canadian Tire, either directly or through an agency. Canadian Tire sees that the user was recently on its website browsing bicycles. This would be deemed a very relevant and targeted ad placement. As such, the DSP has been instructed it is able to make a $3 CPM bid on this user. CPM is cost per thousand impressions. So effectively, Canadian Tire is bidding $0.003 to place that one ad on that one person’s web page. Bidder B represents Walt Disney Company. Based on the user’s web surfing habits and interest in ESPN, the DSP representing Disney believes this is a male, aged 22-29. This information is pretty good but not as targeted as Bidder A had. As such, it only makes sense for Bidder B to make a $2 CPM bid for the right to place the ad. Bidder C doesn’t really care much about the targeting data. Bidder C represents Nike and Nike is simply interested in promoting its new line of shoes on any sports related website. In this case, Nike is only willing to bid a $1 CPM. In Step 4, the exchange selects Bidder A as the winning bidder. In Step 5, the winning ad is placed on the ESPN web page. This whole process took about 5 milliseconds from start to finish.


7.0 Leaders of the Stack

That ecosystem picture we provided above gives a high level overview of the ad-tech value chain. If we rotate that picture by 90 degrees, we get something that techies like to call a “stack”. There is one primary company that seems to be trying to acquire its way into the entire stack and that company is, to no surprise, Google. Others like Adobe and AOL are making their fair share of acquisitions as well – Adobe with Omniture, Auditure and Neolane. AOL, with the acquisitions of Convertro, and Gravity, has repackaged its ad-tech offerings into something called “One by AOL”.


The best of breed vs. suite discussion pops up in most tech segments and ad-tech is no different. One of the top areas experiencing convergence is with agencies that are bringing more tech in-house. There was a time when ad agencies were in charge of strategic vision and creative content while ad-tech companies were responsible for the execution of that vision. An agency comes up with a great campaign and then the tech company makes sure it gets delivered the right way. However, both segments are trying to keep a higher portion of the margin by encroaching into each other’s territory. Ad-tech companies are adding a creative component to their offering and agencies are bringing much of the DSP function in-house. Converged players argue that controlling everything leading into the ad exchange ensures good quality control and flexible execution. Best of breed players argue that independence is what ensures innovation and there is greater transparency in having the agency function and the DSP function separated. Some industry players have speculated that agencies over-inflate the DSP cost in a bundled offering, which has led some brands to explicitly dictate that they do not want their agency utilizing its own DSP for ad placement.

8.0 Content Marketing

Content marketing is how brands try to communicate with consumers in a more subtle fashion than through the typical ad. It is meant to offer real and tangible value to consumers through information or entertainment. Through this process, brand recognition is meant to improve and hopefully lead to positive feelings at some point. These positive feelings should eventually lead to sales. Content marketing is this report. Hopefully you find it useful, which helps build some goodwill down the road so that good companies invite us to invest in them. We could have just placed an ad on the TV, radio or web but we decided that generating this publication was probably the better route. According to Demand Metric, content marketing is 62% less expensive than traditional marketing but leads to 3x as many leads. Prominent examples of content marketing throughout history include the Michelin Guides, which are over 100 years old. The Michelin brothers in 1900 published their first car guide, which they gave away for free. The concept was intended to boost demand for cars and therefore car tires. The idea worked out pretty well as Michelin became and still is one of the largest tire makers in the world. A more recent example is Proctor & Gamble’s, a website devoted to providing puberty advice to adolescent girls. The site’s popularity has grown since being created in 2000 and now gets over 2 million visitors per month. This form of advertising was found to be 4x more effective than typical ads.


So where does the technology come into play? The first place is measurement. Similar to many forms of online media, content marketing effectiveness is tricky to measure. Transaction-based advertising campaigns are easier to measure (ie. Sign up to receive something or click here to purchase). Content marketing is based on the premise that brands need to develop relationships with consumers over time. Relationships are tough to measure and do not correlate well to page views or clicks. So companies like Contently have begun to offer tools that measure engagement time as an example. They will track scrolling and mouse movements in order to determine whether a person is actively engaging with a piece of content. Or they will track what portion of consumers actually finish reading a story. The second important tech category for content marketing is content creation. Contently addresses this issue as well by offering a marketplace that connects brands with freelance publishers willing to blog or otherwise create new content for an average of $275 per blog.

Appinions is a DCF portfolio company that extends the content marketing theme. Appinions’ technology is used to identify and message leading influencers on any topics. Influencers are perceived as being knowledgeable and independent. Hence, their views go a long way in swaying purchasing decisions. Appinions will help content marketers reach the right audience by identifying who the key influencers are.

9.0 Mobile

Mobile advertising is in theory similar to the rest of digital advertising, except the ads are designed to be seen on your phone or tablet rather than your PC. There has been great enthusiasm for this segment because mobility tends to be more contextual and geo-specific. This means that if you are searching for movie info on your phone, there is a higher probability that you are probably planning to see a movie in the near term. If you’re searching on your PC, perhaps you’re simply curious about a trailer or maybe you simply enjoy reading movie reviews.

Mobile ad revenues more than doubled in 2013 to $18 billion and are expected to grow another 75% this year according to eMarketer. Google is still the market leader with about 50% share in 2013; however Facebook is growing quickly having only turned on its mobile ads last year. It is projected to increase its market share from 18% last year to 22% this year.


Advertising on mobile devices comes with a few key differences to the standard PC ad model. Firstly, PC advertising generally relies on cookies from your web browsing history being used to understand your preferences. Mobile devices don’t really have cookies so targeting becomes a more complicated process. Consumers spend about 80% of their smartphone time on apps and only 20% on the web. Traditional online ad companies don’t have any visibility into the consumer while he’s playing with his apps. In-app targeting requires a different infrastructure where firms build back-end technology that integrates into the app developer’s first party data. A few Canadian companies that fit the mobile theme are Addictive Mobility, Juice Mobile and Go2mobi.

The second difference between mobile and PC advertising is that the mobile screen tends to be a lot smaller and so the ad needs to be less obtrusive. A banner ad on the mobile device can take up half of the screen, which doesn’t make for a very worthwhile experience for the user. Facebook and Twitter have been very careful on how and when to launch their mobile advertising networks. Below are examples of how Facebook sponsored posts and Twitter promoted tweets are integrated into the user’s experience.


We stated above that apps are hugely important for the mobile ecosystem. People spend 80% of their mobile time on apps and gaming is by far the most popular activity taking up that time. Flurry in 2012 suggested that 52% of mobile sessions were gaming related. The number 2 category was social networking, representing 22% of mobile sessions. The intriguing thing for developers is that people actually pay a lot of money to play these free games through in-game purchases. Finnish game developer, Supercell, grew its revenues last year to nearly $900 million from $100 million in the year prior. The success was primarily driven from two games – Clash of Clans and Hay Day. A second major game developer, King Digital, recently IPO’d and last quarter generated over $600 million in sales and nearly $250 million in EBITDA because it has 3 of the top 10 mobile gaming apps on the key platforms. The average paying user spends $18 per month playing “free” games like Candy Crush. These are serious dollars, which are undoubtedly attracting serious attention from other game developers. The problem with the app world is it is an extremely pointy pyramid where the top 10 games make 90% of the revenues and the vast majority of games die a quiet death.

This phenomenon has created a major new mobile advertising segment called “app discoverability”. Mobile app analytics companies are suggesting that the price charged for an app install has risen dramatically over the past year. SuperData Research reported that the average cost per install had risen to $2.73 in late 2013 from $0.97 in the year ago period. Some are forecasting that this figure will grow to $7.00 over the next year or two. Companies like Tapjoy and Flurry are leading this new segment. Tapjoy creates an exchange that allows users to earn virtual currency and points if they engage with ads (ie. Watching a video). Flurry allows game developers to target consumers based on their habits so for example if a user downloads a new game, he may be asked if he wants to try out another game that is similar in style.


10.0 Video

Video advertising is expected to grow materially in the coming years. Brands love to put videos in front of consumers because sight, sound and motion are useful in establishing an emotional connection with the audience. This is the likely reason why the $500 billion global advertising market continues to be dominated by TV. According to ComScore data below, online video ad views have grown to about 25-30 billion views per month so far this year from less than 5 billion monthly views in early 2011. The market is below $3 billion annually currently but expected to grow to $5 billion by 2016. The popularity of video ads is partially driven by the high click through rates. The CTR is nearly 2% relative to most other ad formats at sub 0.5%.

Several new ad tech IPOs geared towards video have come to market of late. Tremor Video, YuMe and TubeMogul all aim to help advertisers efficiently place the right video ads in front of the right audience. The key to success in this market appears to be to make the buying process as similar to traditional TV advertising as possible. According to a recent WSJ article, major advertisers such as MasterCard and Verizon Wireless have in the past year begun to shift a portion of their TV advertising budget to online video. The same article reports that Starcom MediVest, a large US ad agency, has shifted $500 million out of TV over the past 12 months, with three-quarters of that moving to online.

A slightly different perspective on the video theme comes from another Canadian tech company named Vidyard. Vidyard helps companies manage and monetize their video content by tracking usage data and integrating into other CRM platforms.

iPowow, a DCF portfolio company, is helping both online and traditional TV broadcasters become more social and boost their engagement levels by offering something called participation TV. iPowow allows these content providers to seek audience input from viewers. This has opened up many new opportunities for differentiated content creation and greater advertising inventory.

AdTech17 AdTech18

11.0 Social

The use of social media has skyrocketed from relative obscurity in the mid-2000s. Facebook has reached 1.3 billion monthly active users (MAU) from a figure of less than 200 million in early 2009. The average tweets per day on Twitter have jumped to over 500 million, which is up more than 10-fold from levels in early 2010. The surge in social media has necessitated an equally rapid adoption of technology tools that help companies understand and manage what all of this new content actually means. Just as companies were finally starting to get comfortable with crafting their own websites and controlling their email content, the landscape has dramatically changed. The discussion of brand and products is no longer centralized at the company’s owned properties (ie. website, email, pamphlets, paid advertisements). With social media, the discussion and sharing of content moves externally to blog sites, forums and portals like Facebook and Twitter. At this stage companies notice that they have lost a great deal of control over their messaging. Millions of conversations are happening that affect how a company is being perceived in the marketplace, which will ultimately impact revenues and the bottom line. Some companies are aiming to better understand the nature of these conversations ahead of full product launches so they are soliciting feedback from users early in the product lifecycle. One of our portfolio companies, Vision Critical, offers access to online groups of customers and/or prospects so that brands get a better sense of how people will react to products, packaging, pricing, etc. before, during and after the campaign.

So now you are a corporation that has identified a need to invest some time and money into social. Where do you spend? According to an Altimeter Group report, the average large corporation spent $833K per year on social in 2010. We believe this figure is likely well above $1 million today. The variation depending on the size of company is significant. Firms with less than $250 million in revenue spent $230K per year on social. Those with more than $10 billion in revenue spent $2 million. But where does this money go? According to the same Altimeter Group report, nearly 40% of the social budget was directed at customer facing initiatives such as paid advertising, agency driven campaigns and blogger programs. About 32% was spent on internal soft costs such as hiring staff and training & development. The final 29% was devoted to technology, mainly management platforms and brand monitoring. All three of these categories are expected to grow rapidly but technology is projected to expand at the fastest pace at over 50% per year. While the rapid adoption of monitoring technology is expected to generate significant data around mentions, sentiment and engagement, companies are still going to be challenged to generate reliable ROI data. According to the Altimeter Group, “48% of social strategists said their primary internal focus is to develop ROI measurements”.

The social ad market is growing relatively quickly but there continues to be some trepidation over how effectively companies are able to incorporate ads in their networks. Social ad revenues are split into two main categories – display and native. Display advertising is similar to banner ads on your PC. Not very exciting and not very popular. Native ads are more integrated into the social experience, similar to the Facebook and Twitter examples we cited above (ie. sponsored stories and promoted tweets). The total social ad market is worth about $6 billion annually, split roughly 60% display and 40% native. This is expected to increase to about $11 billion by 2017.


12.0 So Much Potential for Greatness…and for Failure

Technology in advertising has the potential to do some amazing things with creativity and interactivity. But as always, there is the downside. Let’s start with the bad news first. Below are some examples of how automation in banner ads can damage brands. Most are rather comical if you’re not the brand on display. These examples come courtesy of




Now, let’s end things on a high note. These examples come courtesy of the IAB awards. The common theme below is that video and mobility create greater engagement and emotional resonance than the boring and stale banner ads above. Technology in combination with creativity has the potential to do some amazing things. We hope you agree.

Coca-Cola built two interactive vending machines, one placed in India, the other in Pakistan. The machines allowed users to stand face-to-face and interact with each other on both sides of the border. This type of a campaign makes me want to go out and buy a Coke even though I don’t even like the beverage.


Easy Way Language Center in Brazil built an app that translated local TV shows in real-time to any language over a tablet or smartphone. This type of a campaign demonstrates clear utility for a consumer. People will use it because it adds value in their lives and for those that are interested in learning the language, there will be only one obvious choice.


Unilever created a campaign for its Axe brand, which used augmented reality, mobile, social and traditional media to enter the public consciousness. This branding helped me become an Axe customer.


Dushan Batrovic, Vice President, Difference Capital

Mr. Batrovic brings 10 years’ experience in capital markets and equity research. He is recognized as a technology industry thought leader, having published reports related to software, networking, wireless and digital media and making regular appearances in news media including the Financial Post, Globe and Mail and BNN. Prior to joining Difference, Mr. Batrovic was a co-founder of 4Front Capital Partners and was a Vice-President of Equity Research. Previously, Dushan was a senior technology analyst at Dundee Securities and an equity research analyst at Canaccord Capital. Dushan graduated with a Bachelor of Applied Science and Engineering from the University of Toronto where he ranked top of his senior year class. Dushan has a Master of Business Administration from the University of Toronto where he earned Deans’ List Honours status in both years of the program.

About Difference Capital Financial Inc.

Difference Capital Financial Inc. invests in and advises growth companies. We leverage our capital markets expertise to help unlock the value in technology, media and healthcare companies as they approach important milestones in their business lifecycle. Difference Capital Financial Inc. is traded under the Toronto Stock Exchange under the symbol “DCF”.

This report is for informational purposes only and does not constitute an offer or solicitation to sell shares or securities in any of the mentioned companies or any related or associated companies. None of the information or analyses presented are intended to form the basis for any investment decision, and no specific recommendations are intended. Accordingly, this presentation does not constitute investment advice or counsel or solicitation for investment in any security.

This report does not constitute or form part of, and should not be construed as, any offer for sale or subscription of, or any invitation to offer to buy or subscribe for, any securities, nor should it or any part of it form the basis of, or be relied on in any connection with, any contract or commitment whatsoever. Difference Capital expressly disclaims any and all responsibility for any direct or consequential loss or damage of any kind whatsoever arising directly or indirectly from: (i) reliance on any information contained in the presentation, (ii) any error, omission or inaccuracy in any such information or (iii) any action resulting therefrom.

Difference Capital has investment positions in the following companies that are mentioned in the report: Appinions Inc, EQ Inc, iPowow Inc, and Vision Critical Communications Inc.

Back to Blog

    Search the Blog