Predicting the Growth of Your Social Network

The ability to gauge a marketing campaign in action is essential and exists for most traditional channels. The most glaring exception is social media. Organic’s Steve Kerho unveils a new way to examine an initial campaign splash to determine whether to expect a tidal wave or just a few ripples in the water.

The ability to gauge a marketing campaign in action is essential and exists for most traditional channels. The most glaring exception is social media. This inability to gauge or predict how well a social media campaign is performing means marketers can do little more than hope the effort is successful.


At this point, any marketer worth his or her salt has sat through a number of webinars and conferences that all claim to solve the issue of social media measurement. Yet time and time again, the solutions offered are little more than KPIs that tell little about how growth in the social space relates to the true underlying business objectives. Even worse is that current social media metrics–downloads, total comments, friends/fans, etc.–are measured after the fact, when the campaign is finished. This leads to a “fire and hope” approach to campaigns because marketers don’t have an early gauge of success or failure.

Standard social KPIs suffer from this measurement dilemma because you can’t measure until you know what you have. This leads to a rear view mirror approach to campaign evaluation. While this is an essential book-end for any initiative, it doesn’t help marketers to take actions needed to correct the course of a campaign while it’s still live. This entirely defeats one of social media’s greatest strengths: providing immediate feedback.

What is needed is a metric that enables proactive marketing actions by assessing marketing efforts in real-time so they can be adjusted as needed. At Organic, we have developed this metric so companies can now examine the initial campaign splash and determine whether to expect a tidal wave or just a few ripples in the water.


The Metrics: Velocity and Acceleration
Rather than simply measure the cumulative total of social media metrics, we applied the principles of calculus to examine the rates at which these metrics change. By examining the first and second derivates of cumulative social media metrics we defined two new measures, the velocity and acceleration of social media. These metrics have great predictive power in assessing the final cumulative impact of an event in the social space.

Velocity represents the first derivative of the cumulative social media function. It is the rate at which the social media metrics change. Acceleration represents the second derivative of the cumulative social media function. It is the rate at which the velocities of social media metrics change.


Using these metrics we can understand the impact of specific events on social media. There are three phases to most events in the social space: the baseline, the hot zone, and the fallout.


The “hot zone” is where we see the velocity for the increase in activity quickest; it is the area where our campaign must be most efficient. Additionally, this curve shows us the impact that incremental changes in investment will make against the reach in a lower funnel effort.

In other words, first you have a baseline cumulative social media measure (typical number of Facebook fans and regular amount of people joining) then you have an event (marketing event, PR event, crisis) that makes it change its normal rate (the volume of fans grow for period X then levels off again). Using the initial acceleration of the change lets us determine the final cumulative impact of the event.

The initial shift away from baseline creates a curve that allows a reasonable prediction of future growth. That is to say that the initial acceleration of the event predicts the ‘ceiling’ of the measure.


How We Validated This
Organic was supplied with anonymous data from a number of Facebook fan pages in order to investigate the rate at which people became ‘fans’ of the page. Each of these pages had a distinct marketing event that drove traffic to the fan page. This sudden growth in fans was used to forecast the future level of fans on the fan page. Looking at the acceleration in the rate of growth on the day of the marketing event, and the acceleration (or deceleration) in the rate of growth on the day after, let us see the growth curve between them. The ‘viral momentum’ of the event gave us a ratio that allowed us to predict where the fan count would land weeks after the event.


This information will let marketing managers react quickly if the forecast is short of expectations (i.e. create another marketing event that drives fans to fan pages).

U by Kotex
Recently, Organic partnered with Kimberly-Clark (K-C) on the launch of its new U by Kotex brand. The brand marketing site is unique in that social is at the heart of the experience. Two weeks after the site launched in early March the word of U by Kotex was spreading rapidly. By the end of March there were already:

• More than 360,000 visits to the brand site
• More than 230,000 sample requests
• 3,400 tweets on “Kotex”
• More than 4,300 Kotex discussions taking place in the social space with over 13.6 million impressions through various channels


This clearly demonstrated the desire of consumers to learn more about the product and share this information with others in their social media circles.

Tying In Tweets
Using internal K-C data, we have been able link behaviors in the social space, specifically Tweets, with a measurable ROI-based goal–online sample requests–for U by Kotex. The total number of Tweets had a strong relationship with the number of sample requests the company received in the days following. Using the relationship between the acceleration and the baseline level of social media metrics, we are able to project the ‘plateau’ level of Tweets following any major event in the campaign. We then tied in the relationship between Tweets and sample requests to forecast the cadence of sample requests.

For instance, in week two of the campaign launch an email blast went out to consumers who had previously opted in to Kimberly-Clark communications. This caused an increase in Twitter chatter.


Based on the initial acceleration in Twitter traffic, we expected that event to bring in four times the number of Tweets we saw on the first day. We were also able to create a forecast for the remainder of the month.


As you can see, the results from the following weeks were very much in line with our forecast.

The real value of this projection lies in utilizing these leading indicator properties of Tweets on sample requests. We utilized this relationship to project an expected number of sample requests due to the ripple effect of the email blast. Based on the initial acceleration in Twitter traffic, we expect that event to bring in four times the number of sample requests seen on the first day. The results of the next week showed K-C to be almost two thirds of the way to our expected sample plateau. Once again our projections were tracking well.


Moving forward, with every event that K-C has planned for U by Kotex–from Chelsea Lately to Khloe Kardashian in Times Square–we have been able to provide measures from the social space on the impact of the event and relate that to ROI-based activity goals.

Positive or Negative Sentiment
While acceleration gives us a leading indicator to the size social metrics, it doesn’t tell us whether or not it is a good thing. If sentiment is overwhelming positive (e.g. Dove’s Real Beauty campaign) or overwhelmingly negative (e.g. a safety recall) then the acceleration alone will tell you the size of your success or the size of your fallout. In cases where feelings are mixed, combining acceleration with measures of social media sentiment allows marketers to predict the size of the positive and negative trends separately.

The combination of acceleration and sentiment goes beyond what either brings on its own. With sentiment alone we lose the understanding of the baseline level of social chatter. By accounting for the building levels of chatter, the combination of acceleration with sentiment delivers clear expectations of the size of positive and negative groups and helps marketers decide what level of pressure they need to exert in order to gain their desired outcome.


With velocity and acceleration, Organic has developed one of the first predictive metrics for social media. It can tell marketers whether they will hit the event goal. This allows companies to react and adjust their marketing efforts in real-time to better achieve their objectives. Also, the uses of velocity and acceleration are not limited to company-initiated efforts. They can help monitor the online conversation for complaints, highlighting which issues are big and which aren’t.

The true value of these metrics lies in their simplicity. By capturing the velocity and acceleration of social media, marketers can understand the scope of their impact on the social space. These measurements capture the initial size of a social spike and then simply predict the total social impact. It is likely that marketers already have the data on hand and the computations are straightforward, so there’s no reason not to get started today. The predictive value derived will have marketers looking at all their existing social metrics in a new light.

Calculus is the mathematics of change and one of the biggest changes in marketing has been the advent of social media. And isn’t it beautiful that a modern phenomenon like social media can be made a more powerful marketing tool through the application of some of the most powerful math created more than 300 years ago that was originally used to explain the motions of planets. So raise a glass to Sir Isaac Newton and gives a “thumbs up” like to his Facebook page–yes, he has one and perhaps after this article we will estimate the growth in his fan club.


If you are interested in learning more about how to drive Velocity and Acceleration for your company, I encourage you to attend this webinar on December 21 at 2pm ET, where my colleague Jason Harper will explain the ins and outs of predicting the growth of your social network.

Steve Kerho is the SVP, Analytics, Marketing Optimization at Organic.


About the author

Steve has over 24 years of agency and client side experience leading CRM, interactive marketing, sales and media practices for brands including Nissan, Bank of America, Visa and Procter & Gamble, to name a few. In 2011, he was named an Adweek Media-All Star for his innovative work measuring earned and owned media content and developing predictive analytics models to optimize digital ecosystems