How to measure the ROI of your social media marketing campaign

The main aim of social media is to engage with your customers, improve brand awareness, and essentially build a loyal fan base online. The end result should have effects financially, but how can this be proved? Typically customers don't purchase solely online, or if they do, not off social media platforms.

Like most marketing campaigns a project proposal will need to be constructed for social media marketing. Establishing the aims, objectives, content and progression of the campaign can be established like other medias. However, measuring the effects of social media can be harder. This makes pitching the idea to a manager difficult if they don’t fully understand the concept of SMM (Social Media Marketing).

Reasons as to why measuring the results is tricky

Social media is the ‘transmission of multiple messages to multiple users who extend conversations that invite responses from others’. The image below shows the transactions of a message on a social media site, as you can see the message reaches multiple people who allow more people to see it. This is the snowball effect displaying an exponential growth in exposure using E-Wom.

However, any form of internet based marketing involves digital footprints. Unlike radio and TV digital marketing leaves trails which can be measured to greater effects. Through these digital footprints a ROI can be established.

Olivier Blanchard (2009) discussed this idea in his book 'Social Media ROI'. He noted several steps which offer an ROI;

1)   Establish Baseline – Start date of operations, this baseline will be the start date for all the data. This includes financial data for that time and data separate from the social media campaign.

2)   Create activity timelines - For all recordable factors, log all data regarding uploads and content updates from the campaign onto a graph with the same x-axis as the baseline.

3)    Monitor the volume of KPI’s - Plot the KPIs onto a graph. These KPIs should include all social media leads, website referrals, comments etc.

4)    Measure the amount of sales - Plot sales (£) onto a graph, this can be replaced by number of sales itself or number of customers.

    

    5)    Overlay all the data - Collect all the data which has been measured. 

  

   6)    Look for patterns - Try and isolate changes in the data since the baseline.

   7)    Prove results - From these changes you will need to establish what caused them. 

    As shown in the graph above there may be changes in your sales or number of customers. You will need to look for correlations, what happened previously to the change, maybe you introduced a new ad campaign on Facebook or used Twitter for effectively. It is advised to try and disprove your findings, there is no point making conclusions at this point if you are unsure.

  There are certain limitations to this method; The type of industry and business can have an effect. This method will favour business with large amounts of their operations done digitally, as there will be more data available. This includes data for all sales, advertising and promotions.

For many business this method can be over-complicated, some business choose to measure ROI from traceable promotions such as coupons or vouchers which are downloaded online. Other marketers choose to save money via social media, website hits from social media are free for example. Adding up these advantages and allocating a cost from what it would cost you if you were paying for it can show a rough ROI.

  it is important to remember that your social media strategy needs to align with the objectives of the business as a whole. Your KPI's by which you measure the effects of the campaign need to be SMART (Specific, Measurable, Achievable, Realistic and Time-Based), this is to ensure that they are suitable for basing conclusions on.  
  
Also, remember that not all the information you need will be available. Albert Einstein stated; ‘not everything that counts can be counted, and not everything that can be counted counts’.