Gone are the days of one-touch marketing, where a customer saw one advertisement and immediately was convinced to buy. With the number of online marketing channels at an all-time high, shoppers often interact with 3-5 different channels before making a purchasing decision. In many companies, the reporting has been set up so that every channel claims 100% of the final sale, even when their marketing efforts were only partially responsible. This “double-counting” of converted traffic has these companies overpaying for sales/leads and has inflated their individual channel marketing reports. Performance attribution allows companies to properly measure the impact of each channel on sales, preventing overpayment and report inflation. We consider this the breakout topic for 2013, and hope to provide plenty of education on the subject through monthly blog posts about attribution modeling.
Attribution models help determine the impact and contribution of each channel on sales, offering a more accurate picture of marketing spend and allowing companies to budget their marketing dollars accordingly. Think about it: how many channels does a typical company currently utilize for marketing? Website, organic search, paid search, social media, referrals, affiliate marketing…the list goes on and on. Often times, a customer will touch some or all of these points before finally converting.
Many companies determine attribution through a “last in” policy – where the customer last entered before buying a product. This approach made more sense a few years ago when there were fewer channels, but today it can provide a misconstrued view. For example, if a customer finds a company through organic search and checks them out on social media, those channels should receive some credit, instead of just the affiliate whose link led the customer to finally commit to the sale. Overvaluing the last channel creates a messy picture of attribution for businesses, and last-click only attribution needs to be revisited.
Conversely, most companies currently have no problem letting each group count their sales as one whole sale, inflating each individual channel’s performance and hurting the overall bottom line. Consider this comparison: if a company was comprised of a commissioned sales force and every salesperson took credit for a sale where multiple salespeople had called on the same customer, commissions could cost more than the profit generated. Different channels need to be held accountable for their role/cost; this is the basis of performance marketing, and one of the key reasons why intelligent attribution makes such a huge difference in the e-commerce world.
Developing a successful attribution model gives companies a clearer view of the success of various marketing efforts, allowing marketing managers to increase revenue by better allocating their marketing spend. One much more accurate and widely preferred method is the multi-touch model, which accounts for every channel associated with a sale. This attribution method ensures that no channel is taking credit for too much of a sale, assigning weight to each touch based on how much it impacted a final sale. In future posts, we will further define attribution and the multi-touch model, show how it has worked in the past for successful companies and help you create a basic model for your business.
To get started on the road to attribution success, check out a few of our favorite reads on the subject:
- Why You Should Care About Attribution – This Inc.com read provides an excellent overview of attribution modeling, painting a clear picture of the benefits of creating detailed models.
- Marketing Attribution: Valuing the Customer Journey – Take a deep dive into this whitepaper from Google Analytics and Econsultancy, and see how other companies are using attribution to revise their marketing ideas for better results.
- Last-Click Attribution Models Give Marketers Incomplete Picture – eMarketer breaks down the data from the previous whitepaper and goes a step further, proving that last-click attribution modeling is far less effective than linear and custom models.
- Multi-Channel Attribution: Definitions, Models and a Reality Check – Avinash Kaushik gives a high-level analysis of attribution modeling, defining problems with current models and offering suggestions to improve your own models.
Peruse these articles and start to understand attribution modeling. Be sure to check back here often, as we will continue to discuss the topic of attribution and help you use it to enhance your company’s marketing spend.