http://www.stephendebruyn.com Measuring the ROI of Social Media Campaigns

As social media and social media platforms increasingly become promising targets for marketing professionals, the question of return on investment (ROI) of social media campaigns is gaining in prominence, specifically in comparison with the returns generated by alternative marketing initiatives and tactics that can be employed. 

The case for ROI

Social media is obviously a new phenomenon, and therefore social media measurement models are still undeveloped, but a number of individuals in the social media arena have made attempts to address existing confusion and lay down some basic ground rules.  While many in the field have compiled lists of key performance indicators (KPIs) by which the success of social media initiatives can potentially be measured, Olivier Blanchard and Jacob Morgan, among others, quite correctly emphasize that ROI is by definition a financial equation that specifies the correlation between an investment and its financial return. 

As Greg Satell noted in his blog Digital Tonto, in business finance ROI is actually used to identify the financial return on major investments or initiatives such as whole divisions or product lines, and not for more granular initiatives such as campaigns.  Outside of finance circles there appears to be an increasing tendency, however, to use the term more loosely to relation to financial returns on smaller business initiatives.  Let’s explore whether this actually makes sense.

Key Performance Indicators (KPIs)

Key performance indicators measure various types of actions, expressing different levels of engagement, which indicate intent that in turn can lead to behaviors of various kinds which have an impact on the financial bottom line.  For example, an action such as a website visit represents a fairly low level of engagement and therefore not a lot of intent can be read into it; however, registration for a company’s software demo indicates a much higher level of engagement, and therefore a greater potential for intent that could include purchasing behavior that would impact the financial bottom line.      

While KPI measurements are relatively easy to obtain, the correlations between exposure, intent and purchasing behavior have always been very difficult to establish.  In this context, department store merchant John Wanamaker is often quoted.  He stated more than a hundred years ago; “Half of the money I spent on advertising is wasted; the trouble is, I don’t know which half.”  

Markers on the ‘path to purchase’

Since Wanamaker’s time, corporations have continued to aim for as much clarity in the investment/return equation - after all, they owe it to their stakeholders to allocate resources as efficiently as possible.  More recent technology innovations have helped to establish a more direct correlation between specific marketing campaigns and their results – direct marketing comes to mind –, but the ultimate solution to the ROI question remains elusive due to a simple reason which we as consumers can all intuitively understand – our purchasing decisions are influenced by a large variety of factors, some of which we are not even conscious of.  As the price and complexity of a product or service increases, so does the variety of factors impacting our decision, and the amount of time it takes to come to one.  After all, it’s one thing to buy a bar of chocolate in a deli; quite another to purchase a car or computer system.  Marketers refer to this process as the sales cycle, but it might actually be more accurate to refer to it as the path to purchase – and this path is unique to every individual with each individual purchasing transaction. 

The Holy Grail for marketing would be to be able to identify all the specific individual paths prospects take on their way to a purchasing decision.  In this context, social media is promising, as we tend to increasingly exchange perspectives and opinions about products and services online, thereby leaving individual traces in cyberspace which can be captured as data points.  Even though these systems are still immature, marketers today can in principle monitor online actions and intent related to specific products and services, and have them linked to business systems such as CRM.  Marketers are more than ever able to track and monitor actions and devise intent, but they will never manage to accurately identify all variables impacting purchasing decisions, or accurately weigh these variables in terms of relevance.  Say you were driving down a highway, and glanced at a billboard ad – how much did that impact the purchasing decision you eventually made for that product advertised there?   As you cannot even accurately assess that, how can a marketer be expected to do so?

KPIs defined along an ‘engagement ladder’

With this information as the backdrop, let’s see if we can come closer to a model that as accurately as possible measures the success of a social media campaign in financial terms.  First of all, let’s address the issue of key performance indicators.  KPIs can be quantitative (web visitors, tweets, impressions, etc.) or more qualitative (tone of comments, recommendations, etc.) in nature, with the last category more indicative of intent.  The graph below sorts KPIs along the typical marketing continuum of awareness, interest, engagement and action, creating an engagement ladder.   As one climbs this ladder, engagement level increases, and thereby potential purchase intent; therefore the KPIs at the upper end of the ladder are more valuable from an ROI perspective, even though they do not directly show ROI.

Which metrics to select for a specific social media campaign depends entirely on the goals and objectives of the campaign, but it is imperative to define them, early in the campaign development process.  In his excellent article on social media measurement, Chris Murdough from Mullen Communications identifies five stages in a social media measurement process, and choice of metrics is defined as part of the first step in the process. 

The correlation between marketing initiatives and revenue generation

Depending on the goals and objectives of the campaign, some metrics allow for a narrow correlation between campaign metrics and resulting revenue impact.  For example, a store that wishes to increase in-store sales via a social media campaign could encourage its audience to download a coupon online which is valid exclusively at its stores.  Or a technology consulting firm could as part of its campaign encourage its online audience to download a white paper or tip sheet, with the downloads tracked to eventual sales.   However, keep in mind that correlation doesn’t necessarily imply causation.   Many different markers on the purchase path could be at work – but at the very least, we might be able to link specific marketing activities to specific  results.

Very often the nature of the social media campaign is such that a revenue correlation cannot be identified as easily as with the store coupon example above.  The correlation appears to be more indirect, but not necessarily less valuable.  Many marketing activities are of that nature – for example, the CEO of the company speaks at an industry event.  Will that have a revenue impact?  It surely could, but except in a few rare instances, it will simply be another marker on the path to purchase for a specific subset of the attendees of that particular event.  Or to stay in the social media realm; say you roll out a program on specific social media platforms to generate ideas to improve your product or service.  Hundreds are generated, and a few dozen are implemented, exceeding the program goals originally set.  Sounds like you hit the mark, but what’s its revenue impact?  It will be indirect, take place over time, and be impossible to measure directly. 

A company’s revenue stream can simply most often not be correlated directly with specific marketing campaigns or initiatives.  That includes social media.  What can be attempted is to measure the impact that marketing initiatives in the aggregate and  over time have on revenue and/or revenue parameters such as the number of customers; the number of sales transactions per month; spend per customer; and up-sells and cross-sells.  With a revenue baseline established, a range of marketing activities unfolding over time can be compared against fluctuations in these revenue parameters for a specific time period after the marketing programs have been executed. 

The limitations of measurement

But even these calculations have their limitations, as sales are impacted by many more variables than marketing and advertising alone – there are variables such as the quality of the product or service, corporate activities, competitive initiatives, quality of service and more.  Real world dynamics are just too complex to allow for simple equations and correlations.  Econometric models can handle this complexity to a degree, but most companies cannot drum up the massive amount of data these models require and cannot afford the associated expense.  

In conclusion, can the revenue impact of social media and other marketing programs be measured?  Yes, but typically only in specific cases when a program is very narrowly defined, and results can be directly linked to the campaign.  Marketers simply have no choice but to operate in a real-world environment which is complex by nature.  Any correlation identified does not necessarily imply causation, and results are therefore by definition inconclusive and tentative.

A few suggestions…

Despite all that said, it of course still makes sense to build as much measurement into your program as possible.  Here are four recommendations for the development and success measurement of social media campaigns:

1. Always define goals, objectives and metrics early on in your campaign.

2. Dependent on your goals, aim for KPIs as high as possible on the engagement ladder.

3. Aim for KPIs that allow for a direct correlation with revenue impact – if at all possible.

4. Measure throughout the process, and incorporate learnings into future campaigns.