TrenDemon Content Attribution Model
TrenDemon Content Attribution Model – A Look Under Our Hood
How TrenDemon’s Content Attribution Score is calculated, which factors we take into account and why we chose them
We get many questions about how we attribute the influence of content on goals, how we calculate our attribution score and how we know which content to recommend. To help clear some of the fog, here’s an overview of our attribution methodology and principles:
The basic question attribution systems are designed to answer are which sources or events at certain points of the customer journey have caused other events later on that journey. TrenDemon is designed to answer a more specific kind of attribution question – how well is content influencing users to move forward in their journey towards the desired business objective and which items and assets are the most influential.
Divide and Conquer – Identify your content areas from the rest of the site. This first step helps segment your site to its different areas. This distinction helps us focus our content marketing related questions to the relevant parts of the site. For example, your site may contain a blog, product pages and a pricing area. Analyzing all of these pages, in the same way, will make it harder to draw specific conclusions regarding the performance of the blog.
Define Clear Objectives – Set which events in the customer journeys indicate a successful one. In some cases, this means connecting to another system which lays beyond the website, like the CRM. The deeper the dots which you can connect, the deeper the insight you can derive.
Content to Goals – If the content is king then context is a god. We measure every piece of content in relation to every business goal set. The attribution score of a page is calculated for each of those goals since a given article can be great for getting people to sign up for a newsletter, but less effective for influencing a purchase decision.
Journeys, Not Funnels – We look at the customer journey as a way of connecting information about how users engage with content on their way to a goal. By looking at people, not only outcomes, we can ask deeper questions on how content assets are influencing behavior, from the micro level to the very macro.
Now, that we’ve laid the foundation, we can move on to scoring content. In our model, we take into account the following factors:
The 5 Key Factors we consider:
> Quantitative Presence – In how many successful journeys did the page participate in?
> Qualitative Presence – What is the success rate of the page in journeys?
> Distance from Goal – How long before reaching the goal was the page read?
> Read Score – Which % of people that visited the page, actually read it?
> Proceed Score – Which % of people who visited the page, also went on either to read or act?
Calculation – Dividing the influence pie:
We first take all the participating pages in successful journeys that passed the minimum consideration threshold. We then allocate weights to each page, based on the factors above, for a total of 100% for all participating pages. This allows presenting the relative weighted influence of each page for every selected goal.