September 18, 2020 at 5:39 am #2677Jim EwelKeymaster
Jim: hope things are going well for you in these crazy times. My top marketing issues are:
How to measure the cost/benefit of a marketing campaign in financial terms
How to allocate results across multiple marketing campaigns that reach a customer
MikeSeptember 18, 2020 at 6:35 am #2717Jim EwelKeymaster
Good to hear from you. Except for the book and occasional consulting, I’m retired. My wife and I bought an Airstream trailer and spend about a third of our time on the road. I’m in Eureka Springs, AR today.
In regards to your questions, I don’t cover either of these questions in any depth in the book and I probably won’t before it’s published. Both are notoriously difficult for any business outside of direct marketing. I do spend quite a bit of time steering marketers away from vanity metrics and towards metrics that matter, specifically metrics that other parts of the company are measured on (like sales/revenue, profitability, new account acquisition, etc). The challenge is to separate marketing’s contribution from sales, product, etc. I state clearly that marketers need to focus on three specific things:
Aligning with sales/product/finance, understanding how they’re measured and how they expect marketing to contribute to their success, and then being transparent about the activities and results.
Picking specific measurable customer behaviors that can be show to directly lead to sales/revenue, increased ARPU, LCV or something else, and measure those customer behaviors. Classic examples are sales qualified leads (SQLs) and ABM leads.
Increasing the speed at which marketers test various campaigns/offers. I have a case study from Twitter showing how when they moved from 1 test every two weeks to 10 tests per week, their new active user acquisition skyrocketed.
The attribution issue across multiple marketing campaigns is also very difficult. We did some work on this at Adometry before being sold to Google. It’s all based on the model/algorithm you use, and there are lots of assumptions in any model that you use. I don’t know who’s doing this well today.
Thanks for the feedback and I’ll keep this in mind for a future blog post, if not the book.
- You must be logged in to reply to this topic.