Disparate Systems
More people will understand and rightly be focused on the benefits from unifying attribution across marketing and advertising channels. This means we as the experts should be prepared to tell the story on how we got here and at the same time explain the challenges and barriers that stand in our way. The good news is that in answering these questions we can articulate the difference in data sets and systems for ourselves.
The short, easy answer is that budgets have been historically siloed for decades between print, direct mail, TV, radio and billboard as well as between local and national. Additionally the internet allows for more interactions direct with consumers as well connections with CRM systems.
Since e-commerce began in the late 1990s different technologies have grown up over time providing value propositions and unique, custom built platforms for email marketing, display advertising, ad serving, etc. With programmatic advertising new data opportunities arrived and yet new and vital services appeared on top of it for verification, remarking, etc. The large array of devices that have grown up along with walled gardens and now privacy laws has created a very disparate environment to accomplish unification.
However there is no good reason to avoid unifying marketing and advertising data as it not only will result in cost savings on marketing but it will enable companies to better understand what they are as an organization and how they continue to provide value to customers. Removing noise in the data can create clarity on marketing initiatives where none existed before and importantly this can be done even if we are only able to unify 50% of the data.
Kicking the can down-the-road on attribution unification means that you are putting off the opportunity to understand the data better which is the key to success in selecting the best system and data vendors for your firm. It also helps you understand your staffing needs and skill sets needed right now. The future will arrive whether or not management is asking for it today. It might be smart to have answers and be able to share long and short term plans when your executive team asks what we are doing.
There are a surprising number of organizations right now building out brand new mar-tech systems for the full marketing funnel but that does not mean they understand the use cases for a unified approach to marketing. In fact they can only choose the systems available today which as mentioned are not compatible for unified marketing. Understanding the disparate systems and data sets specifically how we can align data to answer questions and develop strategy is key to success for organizations and individuals in this space.
Data Sets — From the digital advertising standpoint attribution had traditionally been done at the placement level by day as this was the most granular way to evaluate performance. The phrase “placement level” refers to the media plan which has been in use in advertising for decades. The media plan includes the publishers, placements and volume of media running on each along with the dates the media runs.
With programmatic advertising however the preferred evaluate level is at the impression level as this is also at the user level. Additionally the impression can be qualified from verification services to determine whether the interaction is valid (viewable and fraud-free). Impression level data also means that the data is not something a person can evaluate easily in a report because one day would have thousands or millions of rows of data but systems can evaluate the data and store data in CDPs (user level). Ongoing analysis can be done from a data lake holding all the data in this format.
Email marketing data is evaluate normally based on the “blast” which can be done in association with specific marketing promotions. While the data can and often is evaluated at a user level the most basic evaluation is based on the blast as a whole. Savvy marketing departments have automated email systems reconnecting users who visited the site or had some other interaction. Email marketing is a powerful producer for e-commerce clients and has similarities with re-marketing initiatives in display media but with email there are AI operations in place and basic automated response mechanisms. The systems have been developed more around the customized, unique and automated approach to bring in sales. This data could be aligned with advertising data on date but that would likely be the most granular level.
Looking at email marketing and programmatic advertising it is easy to see why CRM data has been the next step clients want to connect to these systems because everything can all be connected back to the user. Looking at the business more closely from a marketing standpoint it is powerful to have user data to craft strategy and direct products where they want it to go. Real time performance metrics from a unified marketing engine on each initiative and customer segment is a beautiful view just coming into view for many executives.
Going back to the data our list of systems in use the question now is if we can pull the data together and align it based on date? This should be possible to do with data exports from email marketing, direct mail, digital display/video , broadcast and walled gardens (Facebook, Google). While not the ideal granularity to evaluate what worked and what didn’t for each marketing initiative this is a solid start for a unified view of marketing performance.
The end product from this effort should be a report of media vendor (i.e. publisher), target information (i.e. sports content), date, media volume and attribution (i.e. clicks, conversions). As manually produced report this could be an extensive effort but at least you proved that it can be done. You can also find many data visualization tools such as Datorama to automate this effort. But it is important to keep the reporting apples-to-apples so that the data is actionable rather than just reports pushed together. This is why I recommend sticking to date as the most granular level.
One of my favorite quotes is from Lao-Tau “the journey of a thousand miles begins with a single step.” The effort described here aligning data sets and attribution from multiple channels is certainly more than one step but it is also only the beginning of very significant work we should see build over months and years to come. The main takeaway is your understanding of the disparate data sets, attribution points, marketing mechanisms and systems.
