Multi-Channel Attribution

add mustard Multi-Channel Attribution Modeling Thinking Beyond The Last Click Adverti sers and marketers have access to a wide range of online channels, mediums and platforms that enable them to make their brands visible to much larger audiences. A go od marketing camp aign shoul d integrate as many relevant, avail able marketing channel s as possible to ensure potential customers can be reached at every step of their journey to conversion. The number and type of interactions a customer has with your brand before a conversion varies between consumers. This means that businesses today may run into trouble when trying to analyse howtheir marketing channels are p erforming both independently and in relation to one another. Attribution Model PPC SEO Direct Social Affiliates Email Most adverti sers will only place the value of a conversion on the last interaction or click to their website, as this is the stand ard attributi on model in all web analytics tools. However, this way of reporting gives the last interaction an unfair advanta ge over all others, creating the need for multi-channel attributi on. Imagine the following scenario: ONE A user receives an email from your marketing team that promotes some of your new pro ducts. The user clicks through to your site and their interest is piqued. TWO Google Google The foll owing week, the user remembers seeing a great product, but cannot remember exa ctly where they saw it. Typing a query into Google, the user is presented with one ofyour PPC ads, clicks on the ad and spends a few minutes browsing your site. They are not ready to buy but are definitely interested in your product. THREE fO O-- f Later that evening, they are served a display ad whilst on their social networking site of choice. This remarketing ad has provided the customer with a special offer for the product they had been looking at. The customer does not have their credit card to hand at that time. FOUR Google Google The following day, ready to buy and credit card in hand, they search for your brand name on Google, click a natural brand listing in the search engine, and go on to purchase the product. This is the path they followed to conversion: SEO Email PPC Social SALE This path can be visualised as a relay race where each channel plays its part in the journey to a conversion. First click is the first runner and gives the team a head start Last click is the final runner who brings home the win Why do we count all the clicks in between? B ecause all the runners in a vwinning team deserve a spot on the podium! SOCIAL PPC SEO EMAIL Taking the standard 'Last Click' model, in this instance SEO will be given all of the credit for the conversion. While SEO was the last to uchpoint before the win, this does not award recognition to the other members of the team. Y 100% SEO 0% 0% 0% EMAIL PPC SOCIAL The Solution The need for more in depth analysis on marketing channel performance and the tremendous amount of data avail able to marketers today means that a number of attributi on modeling tools have been made available. There are many paid resources out there, but as a starting point, take a look at the Attributi on Model Comparison in Google Analytics. n.b. Vari ous standard models are available, but over time and with an understanding of your business goals, custom models are also possible. Here are some of the more common models: Last Interaction 100% SEO 0% 0% 0% EMAIL PPC SOCIAL This mo del is only use ful when your advertising is focused on attracting people at the moment of purchase, or if there is littl e to no consi deration time involved in the consumer path to conversion First Interaction 100% EMAIL 0% 0% 0% PPC SOCIAL SEO This model, like the last-click mo del above, is not recommended if you are looking to track performance across all of your marketing activities. While the first cli ck may have given you a head start in the race to a conversion, it is not the only member of the team to have helped bring home the trophy. However, this model can prove use ful if you are running educati onal campai gns whi ch look to attract new customers and the first interaction is paramount in driving awareness of your brand. Linear 25% 25% 25% 25% EMAIL PPC SOCIAL SEO This model is an improvement on those previ ously menti oned. All channels are given some credit for playing their part. This model is use ful for busine sses looking to maintain conta ct and awareness with the customer throughout the booking cycle. Time Decay 30% 40% 10% 20% SEO SOCIAL EMAIL PPC The time decay model attributes a greater weighting to the most recent touchpoints in the custo mer journey. This mo del is useful if you run regular promotional campaigns for a short, limited amount oftime and you wish to give more credit to the interactions that occur during those perio ds. Position-Based 40% 40% 10% 10% EMAIL SEO PPC SOCIAL This model is recommended when you wi sh to give more credit to the first touchpoint that introduced the customer to the brand and the final one that resulted in a sale, without discounting those in the mi ddle. Non-Direct Click 40% 40% PPC 20% SEO 0% AFFILIATE DIRECT In many ca ses, busine sses prefer not to include 'direct' traffic to a website as a mark eting channel. The rati onal e behind this is that a consumer who comes directly to your website has already made the decision to buy from you, so what has driven that purcha se is not the direct visit itself, but the ones that pre-empted it. Please remember that no attribution model is perfect; it will not account for the effects of offline activity on online, cross devi ce tra ffic and many other factors such as website performan ce, product and seasonality. Attribution, however will help at least eliminate some the speculation previously required when assessing the performance of your online marketing efforts. START CONVERSION