Multi-attribution modeling: A way to enhance your school’s marketing measurability
Date posted: April 15, 2013
Digital marketing has grown rapidly over the last couple of years but measurability has led marketers to question the efficiency of all the effort that is being put into the process. Colleges and universities reach targeted audience in a variety of ways across a wide range of channels (e.g. social, search, affiliates, print etc.). To effectively compete in today’s competitive educational marketplace, colleges and universities must improve on their marketing strategy and on their commitment to measure their marketing impact. These enhancements in strategy and measurement will produce changes in the marketing channels used to reach specifically targeted audience groups (e.g. prospective students, alumni donors, the local community etc.) based on their intimate understanding of their marketing channels, tactics and touchpoints.“93 to 95% of all touch points are ignored when you attribute conversions to the last click”
Microsoft Atlas Institute Knowledge of which channels are most crucial provides some direction for analysts to develop the periodic evaluation of the institute’s marketing efforts and suggests areas of the improvement or changes might be necessary.
Astute analysts have come to realize that simple marketing channel attributing methodologies, such as “first click” and “last click” attributions, are flawed and result in misallocation of school marketing budgets and efforts (which further affect the overall campaign ROI) because most of them fail to provide full insight into how a user interacts with marketing touchpoints before the conversion happens.
What is multi-attribution?
Let’s consider a scenario of a prospective student trying to enroll in a diploma program.As a very first step, the student searches for “Accounting diploma” and sees Adwords ads from several accounting schools.Let’s say one of the schools that comes up in the search is McGill University. The student then then searches “McGill University,” visits McGill’s website by clicking on an organic result, leafs through a few web pages, and
leaves the site before making a decision. McGill University retargets this student a few days later through ad banners and the student comes back to the site, identifies a program of interest, leaves to search for program reviews, and then comes back through some review portal and eventually completes the enrollment
In the above scenario, a prospective student is exposed to the following marketing channels before enrolling into a course.
As we all know, most of the analytics tools use last click attribution by default. In this case, using last click would result in the review portal getting full credit for generating a new conversion. All of the previous touch points would not be acknowledged as being a part of the entire process. In contrast, multi-attribution tools track all the touch points that the prospect was exposed to before the conversion, and allocate credit among all these nodes accordingly. Understanding the role each touch point has can further help you make vital marketing decisions to manage advertisement budget at a macro level.
There are a few different ways to distribute credits when using a multi-attribution tool. Here are some of the most used ones:
Funnel Position: Classify the touch points as Introducers (the initiator), Influencers (touch points in the middle of the funnel) and Closers (last known touch point before conversion) and distribute credit based on organizational preferences.
Time decay: Assigns the maximum credit to the touch point that occurred nearest to the time of conversation.
Algorithmic: Use statistical and predictive models while taking time decay and a linear model into account.
Even credit: Simply divide the credit evenly among all known touch points
Choosing the justified multi-attribution tool
Picking the right tool is perhaps one of the most vital decisions your college marketing team will make. Each tool is unique, they have their own algorithm, but for the sake of simplicity and effectiveness, here is a comprehensive set of parameters you should consider while choosing a multi-attribution solution.
Data accuracy and integrity:
The tool is nothing more than a way to visualize data. Quality data is persistent and continues to provide insights way beyond its projected life. In order for any multi-attribution solution to board your college’s needs, it is essential that it should be able to answer affirmatively to following questions:
- Does the tool account for cookie deletion and the use of multiple browsers?
- Does the tool retain historical data?
- Can the tool track user interactions across both online and offline channels?
- Does the tool store data at a granular level (including clicks)?
Customization and flexibility
Make sure the tool is flexible enough with well-reasoned and built-in attribution models that you can easily pick from. Also ensure that the tool “Be clear about your goal but be flexible about the process of achieving it.”allows for enough customization to suit your organization needs and goals. Additionally, ask your vendor the following questions to make sure it’s the right fit:
- Does the tool allow you to easily build custom attribution models?
- Does the tool allow you to use custom cookie windows?
- Does the product allow you to change attribution methodologies?
- Is the tool flexible enough to allow multiple models at the same time?
IT resources at most colleges are already tight and don’t have the flexibility to allocate dedicated resources. In order to reduce (or avoid) extra overload, the proposed tool should meet the following criteria:
- The tool should be easy to deploy.
- The tool should play nicely with existing web ecosystem and tag management solutions.
- The tool should be easy to maintain on an ongoing basis when new updates are deployed or changes are made for conversion optimization.
There are hundreds of multi-attribution tools in the market right now, and they all have their own pros and cons. While choosing the multi-attribution solution, keep in mind that data quality in particular is vital. In order to truly evaluate the performance of your marketing effort, your tool needs to have a robust tracking mechanism that can account for all user touch points on- and offline.
My Top 5 Multi-Attribution Tools:
Over the course of my digital marketing career, I have used many different analytics/attribution solutions. Some worked, some didn’t, and some are just plain AMAZING. Here are my favorite five attribution tools:
Attribution modeling has become an inevitable essence in today’s competitive media market place, where audiences are bombarded by multi-advertisement messages. Using a multi-touch attribution model allows your college to track all user touch points and pull more precise information about user interaction with their brands.
Proper attribution still poses a number of challenges to marketers and institutes alike because there are so many parameters involved. There is no clear-cut answer to queries about weighting each channel or the impact that has on the value of a conversion. As technology and digital analytics continue to reinvent themselves, new answers will continue to appear.
In the second part of this post, we will discuss multi-attribution solutions using Google Analytics. Stay tuned, and in the meantime, please tell us about your multi-attribution solutions and share your feedback.