Data-Inspired CMO: If Social Drives 2-16x More Sales Than Expected, What’s the Rethink on Marketing Strategy?

To adapt Geoffrey Moore’s disruption model, one could argue that “data” is well on its way to crossing the chasm.  There’s an explosion of data available to marketers and its use in inspiring better marketing decisions is increasing dramatically.

One recent example of data use is in the area of trying to make sense of social ROI, to help optimize marketing spend and mix.  With social spend growing rapidly every year (projected to reach over 20% of marketer budgets within a few years), it’s critically important to ensure there’s concrete value captured from the spend.  

Some studies have suggested social has a very small impact on sales.  That certainly doesn’t make it as attractive a purely sales-focused investment.

Recently, for example, KLM undertook a “last-click” attribution analysis (as the name suggest, this is based on considering whether or not a user’s last click before a purchase was from social media) and found that social was driving about 25 million euros in sales.  Sounds good, but that’s a tiny 0.3% of KLM’s 9.7 billion Euros in 2013 sales. And, in a much discussed analysis, Coca Cola found that that social buzz had no measurable impact on sales.  They did, however, point out that social is still crucial to the brand and continue to invest in the channel.  And, there’s constant discussion about the decline in Facebook organic reach.

How’s a CMO to make sense of conflicting social ROI data?

If a recent study is correct, the methods we use to go from data to marketing mix insights need to be revisited.

new WOMMA study conducted by econometric modeling firm Analytic Partners found that word of mouth drove around 13% of sales.  Of that, approximately 4%, was attributable to brand social discussion.  This is separate from paid social ad sales impact. 

The study used a specific type of econometric modeling to look deeper than clicks (and to look at interactions between media types).  While the study looked at social mentions (vs. the estimated reach of those mentions), the approach  looked at more measures of offline and online social discussion, categorizing social commentary on factors such as sentiment, emotion and intensity. And the study looked at a detailed set of data from a range of brands, from low to high involvement/price.

Sales impacts of word of mouth overall ranged from about 2% to over 25% (average was 13%).  Assuming about 1/3 of that is driven by social, that’s about 2-16x higher than previous studies suggest. In general, word of mouth had a more pronounced impact on higher involvement/price purchases.  Brand sponsors for the study included AT&T, Discovery Communications, Intuit, PepsiCo, and Weight Watchers.

What are the implications if this level of sales is being driven by word of mouth?  

For some categories,  they are significant and should prompt a  rethink in how spend is allocated – or at the very least prompt a deep look at the category data.  For example, the following might be 6 new core marketing pillars to maximize ROI:

  1. Use new metrics and analytic methods.  Add social/viral reach as a complement to paid reach (measure the relative ROI against brand metric and sales shifts),  optimize the impact of social content on brand metrics using surveys/panels and consider multichannel attribution to track sales impact.  And consider tracking offline word of mouth to look at at a more expansive reach metric.
  2. Track and drive reach by enabling fans to talk and advocate, online and offline.   It’s harder than ever to break through the noise.  We use social data to help brands identify the intersection between their brand benefits and consumer passions.  The best share-focused creative does just that, in a way that reinforces what’s best about the brand.  Adding a focus on enabling brand fans and influencers to share their brand love (advocacy) helps amplify the message.  In our study of 22 brands across 4 countries, we found that some of the best brands have 3x the brand social advocacy levels of the lowest advocacy brands.
  3. Enable social advocacy everywhere (consider owning your social relationships).  Viral reach is a particularly important driver of marketing efficiency through social, and it should be enabled everywhere – website, mobile, branded communities and multiple social platforms.  Consider owning your brand social relationships (for example, see the recent Sony Playstation branded community that curated over 500,000 pieces of content to help drive the PS4’s launch success).  Forrester’s also a proponent of this concept of brand ownership of social, and an expansion beyond one or two social channels.
  4. Integrate channels and scale with paid to improve brand and sales impact.  This study found interactions between paid media and WOM.  Our previous work has found that programs are most effective when combined with other types of marketing communications, and reach is scaled up to a meaningful audience.  We recommend “surround sound” strategies that are executed across multiple traditional, digital, mobile and social platforms and supported by paid media and event/experiential/sampling to encourage offline word of mouth.  New paid tools from social platforms also allow this scale to be targeted to audiences that matter.
  5. Optimize for funnel stages and the specific consumer decision journey.  We’ve found consumer/customer mentions can have impact on brand lift (over 1.5-2x brand attitude shift likelihood in our own restaurant study), and also on sales. However, the impact will vary depending on the content, the audience characteristics and the category.  The best brands test and learn.  We’ve created new tools that allow for rapid content effectiveness and virality testing – and we use advanced analytics like multichannel attribution to adjust spend where appropriate. While historically these techniques have been prohibitively expensive, newer panel- and survey-based tools allow for an estimation of attribution starting at lower price points.
  6. Apply (and track) additional social data throughout the marketing and product planning/development process.  The directly measurable impact of social and consumer advocacy content is only one part of the revenue value.  The full social/content picture should include the value of social data and insights in informing the entire product marketing process.  Social insights and data inform product development, marketing, email/CRM, loyalty and media focus.  Brands need a structured and rapid process for taking social data and insights and using them, for example, to generate more relevant emails (higher conversions),  adaptive web sites,  dynamic creative briefs, network value-focused loyalty programs and more impactful paid targeting.


Note: Ogilvy was a sponsor of the referenced WOMMA study.

Image source: WOMMA,



Building a Passion Brand: Key Findings and Insights from our 2013 Global Advocacy Study

“All human actions have one or more of these seven causes: chance, nature, compulsions, habit, reason, passion, desire” – Aristotle

When we think of how people express passion for a brand, do emotions trump reason?

We know brand advocacy is hugely important to making marketing more relevant and effective.  And advocacy via social channels is especially valuable because of its tremendous potential to scale.

But what really drives people to express their passion for a brand through advocacy in social media?   Our newest Social@Ogilvy research – the most comprehensive study of global social advocacy to date – analyzes millions of social brand mentions to help us better understand advocacy for brands online.  The data – which includes about 7 million mentions of 20+ brands and 8 feature films across 4 countries including China, Brazil, UK and US – provide us with insights and clues on how to build brand advocacy.

Here are some key findings:

1) Brands are largely failing at driving advocacy in social media.  Most brands are driving very low social advocacy from their satisfied customers. It’s estimated that less than 5% of satisfied customers advocate publicly for the brand on social channels.  This “social advocacy gap” represents a huge opportunity to improve marketing’s efficiency and effectiveness.

2) Practicality trumps emotion.   Overall, advocates in all four countries were more likely to talk about product features than benefits, cost (or deals/savings), customer service or ads.

3) True passion is rare.  For most brands, the majority of mentions were casual. In the US, only 2 brands had over 50% of mentions falling in the most enthusiastic advocacy category (love, excitement, must-do or buy). And these 2 brands had even more enthusiastic advocacy than blockbuster movies like The Avengers and The Hunger Games.

Based on these findings, we’ve come up with 5 key recommendations for brands interested in tackling the social advocacy gap. Take a look through our study to learn more about how brands can turn advocacy into passion.

A special thanks to Mark Bonchek at Think Orbit, for providing some invaluable suggestions on an earlier draft of our study.

Get Value Now from Social Big Data

A version of this originally posted on Harvard Business Review Blogs as “Metrics Are Easy; Insight is Hard”

Big data is great. But we should consider that we’ve actually had more data than we can reasonably use for a while now. Just on the marketing front, it isn’t uncommon to see reports overflowing with data and benchmarks drawn from millions of underlying data points covering existing channels like display, email, website, search, and shopper/loyalty — and new data streams such as social and mobile engagement, reviews, comments, ratings, location check-ins and more.

In contrast to this abundant data, insights are relatively rare. Insights here are defined as actionable, data-driven findings that create business value. They are entirely different beasts from raw data. Delivering them requires different people, technology, and skills — specifically including deep domain knowledge. And they’re hard to build.

Even with great data and tools, insights can be exceptionally tough to come by. Consider that improving Netflix’s recommendation-engine accuracy by about 10% proved so challenging that only two teams — of tens of thousands from over 180 countries competing for the $1 million prize — were able to hit the goal. Or that, despite significant work to improve online display ad targeting, the average clickthrough rate (and, by implication, relevance) still remains so low that display ads on average receive only one click for every 1,000 views. That is, the vast majority of people who see the ad don’t think it’s interesting or relevant enough to click on

When they are generated, though, insights derived from the smart use of data are hugely powerful. Brands and companies that are able to develop big insights — from any level of data — will be winners.

Here’s a four-step marketing data-centered process that doesn’t stop at the data, but focuses instead on generating insights relevant to specific segments or affinity groups:

1. Collect. Good data is the foundation for the process. Data can be collected from sources as varied as blogs, search, social network engagement, forums, reviews, ad engagement, and website clickstream.

2. Connect. Some data will simply be useful in the aggregate (for example, to look at broad trends). Other data, however, is more actionable if it’s connected to specific segments or even individuals. Importantly, the linking of social/digital data to individuals will require obtaining consumer consent and complying with local regulations.

3. Manage. Given the speed and volume of social interaction online, simply managing big data requires special techniques, algorithms and storage solutions. And, while some data can be stored, other types of data are accessed in real-time or only for a limited time via APIs.

4-5. Analyze / Discover. These part of the process works best when they are broadly collaborative. Using statistics, reporting, and visualization tools, marketers, product managers, and data scientists work together to come up with the key insights that will generate value broadly, for specific segments of customers and, ultimately personalized insights for individual customers.

Consider these insights — drawn from detailed studies and data analysis — that are being used by us and others to deliver value today:

Friends’ interests make ads more relevant. Based on the evaluation of social-graph data and clicks, companies such as 33Across have found that showing ads based on friends’ similar interests can substantially raise ad click/conversion rates.

Sometimes it’s okay if people hate your TV show. A television network commissioned Ogilvy to look at the relationship between social media buzz and ratings. An analysis of thousands of social media data points and Nielsen ratings across 80 network and cable shows identified ways to help predict ratings changes and find the specific plot lines and characters that could be emphasized in marketing to drive higher viewership. One insight was that it’s critically important to look at data differently by show and genre. As an example, for some reality and newly-launched cable shows, both love and hate — as long as there was lots of it — drove audience ratings.

Social media works best in combination. Measuring the actual business impact of social media and cross-media interactions (beyond just impressions) is in the early stages and could have perhaps the most profound impacts of all on making marketing better and more efficient. For example, by exploring panel-based data on brand encounters by socially-engaged customers in the restaurant industry, Ogilvy and ChatThreads found that social media was very effective in driving revenue from this segment. However, this effect was strongest when social media were combined with other channels such as traditional PR and out-of-home media. Exposure to these combinations drove 1.5x to 2x increases in the likelihood of revenue gains.

Each of these insights works because it is actionable and generates value. Each one provides a concrete road map for making marketing more effective and efficient. And applying each insight creates value that both brands and consumers can appreciate.

The infographic below is exclusive to

Yes, Social Does Impact Sales


Recent research from Forrester suggests that social media has an insignificant impact on sales. While this may be true within the specific context of the study, the study’s methodology makes it impossible to draw broad conclusions around the impact of social on sales across all of the “buyer journey.”

Here are three key reasons why:

  1. More comprehensive tracking of content engagement paints a very different portrait. The Forrester study tracked social as a driver of sales only if someone clicked a link on a social property and made an online purchase within 30 days. In fact, brand social strategy is about engaging people with the brand with the intent to increase sales in the future —  both online and offline. In a quick service restaurant study we did with partner ChatThreads last year, exposure to social media was a significant drive of sales increases. And, when combined with other media (for example, editorial and billboards) social exposure resulted in a 1.5-2x higher likelihood of purchase across all 5 restaurants in the category. Further, two more studies support a social-sales link: Edison Research’s study last year showed that 28% of social media users cited social networks as influencers of their purchase decision. And in a 2011 ROI Research study, just over 50% of respondents reported they would likely purchase a product after following the brand on Facebook or Twitter.
  2. In some categories, there’s a sales impact over a longer time than 30 days. With purchase taking several weeks to months depending on the category, a 30-day window doesn’t capture the impact of many longer-lead-time purchases (e.g., air/hotel, some automotive, some apparel, many technology purchases, non-food CPG items and much of B2B). Forrester’s study authors point this out as a potential limitation.  This is a more complex area of study, but data from companies like Ford (see below) suggest a good trend.
  3. “Non-social-network social” drives impact. For many categories, bloggers and review sites are very influential in the purchase journey (but weren’t specifically looked at in this study). Mom blogs, budget/finance blogs, tech/gadget blogs are among those blog categories that are widely read and trusted; in travel, Expedia found a 1 point increase in review score correlated with a 9% increase in average daily rate the hotel could charge; and finally in many categories, reviews via and many other sites are similarly important.

Many leading brands, such as Starbucks and our client Ford, have continued to increase their investment in social because of the long-term payoff. And great campaigns continue to deliver: Ford, for example recently found that not only did its Focus Doug social media program drive 61% higher consideration for Focus, but it also was cited as a driver of purchase.

To help CEOs and CMOs make better spend-and-mix decisions in an environment where social content takes an increasing share of consumer attention —  and influences decison-making —  it’s important we work harder to tie social to business impact, specifically sales.  While many CMOs – 58% in one survey by Wildfire and 76% in a Bazaarvoice survey find social does impact sales, there’s still uncertainty about the magnitude and the factors that drive impact.

We must move beyond solely reporting activity such as Likes, Engagement and Tweets. And we must find business-impact-focused ways to compare social to other media, avoiding the temptation to fall back to easy metrics that don’t tell the whole story (e.g. ad equivalency or rating points).

We have started taking the first steps towards this goal by rolling out a business-impact model called Ogilvy Conversation Impact that looks at social against key business impact areas such as Reach, Positioning, Preference and Action (Sales). We will also soon be formally announcing a new advanced cross-media social and digital impact attribution tool. This tool can determine the link between exposure to social content and offline or online sales  moving us and our clients one step closer towards real data-driven social spending & allocation decisions.

Source: The Purchase Path of Online Buyers in 2012, Forrester Research, Inc., September 2012.

The Promise of Gamification for Brands — And Some First Steps

Gamification can drive people, organizations and even societies to achieve goals. Ordinary life already has many examples of how gamification drives people. The game of competing for the highest net worth or visible expressions of wealth drives many to pursue wealth far beyond what’s required for a comfortable life.  In a different space, the game of competing for the best citation score drives some researchers to pursue specific publishing strategies.

At a very basic level, most of us use some type of score to track how we’re doing. Think about some of the subtle and not-so-subtle motivations that play a role in life -whether it’s citations, salaries, mile run times, or-for a much smaller group-number of Olympic medals. Creating experiences that embody rewarding “gamification” elements like these can dramatically influence behavior;  and, research suggests the enjoyment and memory of experiences persists much beyond the purchase of a product.

The promise of creating gamified experience for brands is not around simply adding points and badges. There’s much thinking that we and others are doing on how to move gamification towards a genuinely rewarding, valuable experience built around core human behaviors-while avoiding as much as possible the temptation to simply add gimmicks.

Built well, these experiences can and will provide dramatic benefits to consumers and brands.  A good early example of this thinking is embodied in the highly successful area of airline and hotel loyalty programs–and more recently, frequent buyer/shopper programs. There are newer, evolving programs around enterprise gamification from companies like Badgeville, Bunchball and RedCritter that are looking to smart gamification to improve employee productivity. Look for additional approaches to gamification across the spectrum, from employee wellness to customer service, CRM and loyalty.

The following are some early stage examples that we’ve executed successfully specifically around marketing (see linked presentation below from

  • Transforming the traditional test drive press release into an engaging Ford Fusion mobile test drive app.
  • Generating casino resort sales and consideration by providing a new layer of financial and non-financial incentives for sharing and discussion around Caesars / Harrah’s Atlantic City properties.
  • Educating people about changes to bank transfer procedures with a shocking (literally) teller deprogramming feature by Kiwibank.
  • Joining with consumers to discover food pairings for Tabasco.
  • Making it fun to learn about such topics as solar power (SunPower), anklyosing spondylitis (Pfizer), CPUs (Intel) and social influence (Ogilvy).

These are just the beginning steps. There’s deeper games and gamification work on the way that’ll transform some of the duller relationships between brands and consumers.

Brands, Games and Gamification

View more presentations from Social@Ogilvy

Twitter’s Way Forward: Analytics and Interest Graphs

With just a few days remaining before Facebook’s IPO, it’s clear that Mark Zuckerberg and his team in Menlo Park have built a substantial business opportunity around the world’s social graphs.  And, based on Facebook’s recent data showing a 45% year-over-year quarterly revenue growth, they’re monetizing at a respectable pace.

Is there a similar opportunity for Twitter?

The platform has been nothing short of a world phenomenon, popular in countries around the world, an undisputed factor in significant political change and a star for customer service. It continues to grow rapidly, with a recent eMarketer study projecting that Twitter’s growth will be up to 4x higher than Facebook (on a lower base) in 2014. And, there are many of us who greatly value the discovery and connections we achieve through Twitter.

Our own research suggests exposure to social content has a significant impact on brand perception and sales impact (2-7x increased probability of purchase lift based on a restaurant study).  So, the general outlook for social content is bright.

The Problem

Yet, even with its success in some high-visibility areas and its fantastic growth, it just doesn’t seem like Twitter has quite hit its stride yet.  It has been overshadowed by Facebook and, at least for now, new entrants, like Pinterest.

  • Active user base lags Facebook. It’s the number-two social network, based on visits, but its current total user active base is still significantly lower than Facebook’s (about 1/6 of Facebook in the US, according to eMarketer*).
  • Time spent is remarkably low. Monthly average time spent on Twitter (36 minutes) is 91% lower than time on Facebook (6 hours, 33 min) and 54% lower than Pinterest (1 hour, 17 min), according to Mediabistro/Statista.
  • Brand revenue growth has been slow. Partially as a result of the above factors, monetization has lagged, with Twitter’s estimated 2011 revenues at $139.5 million (eMarketer), or about 1/25th of Facebook’s $3.71 billion actual 2011 revenues (S-1 filing).

It seems that these issues are among those threatening Twitter’s ability to increase its momentum with brands and organizations.

The Way Forward

Based on the data above, some of our own research, and some thinking based on work with brands, it appears that there are three steps that could make a dramatic difference for Twitter:

  • Provide more business impact data and allow paid targeting around more business impact factors.  With their focus on engagement, Twitter’s analytics don’t demonstrate business value for brands (both for paid and earned engagement).  Facebook has built a useful brand lift analysis partnership with Nielsen (although that focuses on paid) and their newer ad products provide some additional ad targeting capabilities.  It would be very useful to see more Paid/Owned/Earned analytics around the core business impact areas of Reach, Positioning, Preference and Action/Sales.
  • Create improved experiences around interests.  The platform’s focus on a real-time stream of short data chunks has made it well-suited for lighter engagement, as reflected in the time-spent stats.  It’s great for checking out what’s trending, or shooting a quick @reply to someone, or Tweeting a customer service issue.  What about creating more of an experience around interests? From kayaking to cloud computing, it seems that there must be a better user experience for conversation and discovery around interests. And, make it a core experience element, not one provided by a third party. There’s unique value for brands and organizations if this happens.
  • Improve the visibility of relevant Tweets.  In our study, we saw a ratio of brand-related Tweet exposure that was significantly below that reported for Facebook brand-related stories, even when adjusted for the relative sizes of the active user bases. Some options for improving the visibility of Tweets from people one follows include word clouds, a squareified tree map, or just an optional Important Tweets view, drawing on signals like RTs and frequency.  Hopefully, some of those relevant Tweets would include brand or organization Tweets.

In summary, there are some important analytics improvements that would be great to see.  Additionally, it seems like discovery and engagement around interests are key underleveraged benefits of Twitter.

Much like Facebook has built an experience around social graphs, Twitter can and should build a better experience around interest graphs.


*Internet users who access their Twitter or Facebook accounts at least once a month, eMarketer March 2012

Exposure to Social Media Linked with Changes in Sales and Brand Perception

With various types of social media (Facebook, Twitter, message boards, blogs, private social communities, YouTube and other social sites) taking up an increasing portion of consumer attention, we wanted to understand just how relevant social content exposure is to changes in sales and brand perception.

Much of the work to date has looked at direct channel impacts; for example, do direct clicks from a social media site result in sales? This study of restaurant consumer’s attempts to understand the more complex factors that lead to consumer purchase and perception changes.

We released our final report yesterday at Pivot Conference 2011 in New York. We found that in the real world, social content exposure – by itself and more broadly when combined with other types of media exposure such as out-of-home, PR or TV ads – is linked with 2-7x higher likelihood of consumption and actual spend increases. And, social content exposure alone is associated with the largest impact on week-to-week brand perception changes.

The infographic below the fold provides a summary of the results. Here’s the full study:

Continue reading Exposure to Social Media Linked with Changes in Sales and Brand Perception

The Coming Influence Marketing (R)evolution (Wherein a Small Percent of Your Audience Drives Outsized Value)

What if, instead of targeting 5,000 people, you could achieve the same bottom line results by engaging 500 or even 5 people, at a lower total cost? That’s the potential of influence marketing. Is it living up to that promise and how can this type of marketing be scaled in 2012?

What we do know:

  • Changes in consumer attention mean marketing is changing, and the change is dramatic
  • It’s becoming clear that one of key players in this change is the individual consumer
  • Individuals are playing a central role because they trust each other often a lot more than they trust companies
  • In today’s environment they’re able to better communicate and share with each other on many things, including products, services and causes

So, do we then target all individuals engaged in social media? Our thinking is that individuals who are influential can create outsized value. There’s been a lively debate around this (see, for example, Paul Adams excellent discussion and his comprehensive collection of relevant research links). It seems to me that most of the debate seems to center around the definitions of who is an influencer. To us, an influencer is not defined solely by the number of people they connect to. Quite simply, an influencer is someone who is capable of and wants to – bring about changes in awareness, perception or action in a group of people, around a specific topic. Below, we present 3 real world data points assessing the value of different types of influencers.

Continue reading The Coming Influence Marketing (R)evolution (Wherein a Small Percent of Your Audience Drives Outsized Value)

New Ogilvy-ChatThreads 2011 Study Preview: Exposure to Social Media Associated With Sales Increase

Can social media drive sales? One recent study found that about a quarter of social network users report having had their purchases influenced by social media.  On the other hand, another recent study found that direct clicks from social media sites didn’t generate notable sales.  But this type of click study doesn’t really capture the full extent of media influence on consumer behavior.

To take a more comprehensive look at the impact of social media, we tracked a sample group of Quick Service Restaurant (QSR) consumers’ exposure to a variety of brand ‘touchpoints” and obtained pre- and post-tracking data on purchase.

What we found

According to results we are previewing from this new Ogilvy-ChatThreads study entitled Integrated Social Media Sales Impact, there is a strong link between an increase in sales and exposure to social media.

In fact, we found that in the real world, social media exposure – by itself and more broadly when combined with other types of media exposure such as out-of-home, PR or TV ads – is linked with 2-7x higher likelihood of consumption and actual spend increases for some QSR brands.

Specifically, here are some key results, summarized in the diagram and below:


Continue reading New Ogilvy-ChatThreads 2011 Study Preview: Exposure to Social Media Associated With Sales Increase

Sustainability Innovation: Less Gloom, More Fun? Happy Earth Day!

The world’s demand for resources (food, energy, water, minerals) seems to be outrunning our ability to supply these needs in a way that doesn’t dramatically impact our lives. We are increasingly exposed to threats such as rising core inflation (described recently by Chinese officials as a long-term, not short-term challenge), financial burdens of securing energy supplies (and the cost of investing in new ones), and the multiple economic and social risks of long-term climate change on our planet and way of life.

The trouble is that, according to “Mainstream Green“, a new report released by Ogilvy Earth, all this doom and gloom hasn’t been particularly effective in driving mainstream consumers to make (and push for) changes to reduce demand for energy and other resources.

In the US, there’s a 30 percentage point gap between people’s stated importance of living sustainably (80% say this is important), and their action (50% engage in sustainable behavior, e.g. taking public transportation or hiking/biking to work; using eco-friendly products and recycling). (It’s less in China — 14 points — for reasons covered at length in the report.) The mainstream consumer isn’t adopting or championing behavior change.

The possible solution?

Continue reading Sustainability Innovation: Less Gloom, More Fun? Happy Earth Day!