A Simple Template For Running Marketing Experiments
Earlier this year, we talked about How and Why Movers Should Run Marketing Experiments. From that article, we wrote:
The next step is to make a list of potential new marketing channels to experiment with. These may include new or growing paid lead sources like MyMovingReviews or HomeAdvisor, established lead sources you haven’t tried in a while like Moving.com or Equate Media, or other performance-based programs like P4P’s Premium Program.
In this article, we will provide a very simple template you can use to run these experiments. Our goal with this article is to give you a framework you can use to try out a new channel, and compare the resulting business from that channel with your other marketing channels.
Below is a screenshot, or here is a link to a google spreadsheet you can make a copy of to manage this for your own business: https://docs.google.com/spreadsheets/d/1WmgktYF554I5w8rOyJLJTdN5mrrCAx6Pu9c3PCOypMY/edit?usp=sharing. The same spreadsheet is available in Excel for download here.
r each new experiment, make a new section on the spreadsheet. Set a budget, such as $1000 per month, and turn it on. At the end of the month, you’ll collect all the metrics here to decide whether the channel has been effective or not. Let’s look at Experiment 1: EXAMPLE in the spreadsheet here:
Row 7, “Amount spent” (INPUT): At the end of the month when you receive a bill from the lead provider, plug that number in here for how much was spent throughout the month.
Row 8, “Total leads” (INPUT): The total number leads generated. For an Adwords campaign, this would be the number of people who called the number on your website or filled out the form that you can trace back to Adwords. For Moving.com or Equate Media, this is the number of leads they sent your way. For P4P Premium, this number is not applicable because P4P books the appointments directly with your salespeople and doesn’t charge on a per-lead basis.
Row 9, “Total Appointments” (INPUT): The number of in-home or video survey appointments generated by those leads.
Row 10, “Total Booked Moves” (INPUT): The number of moves that ultimately get booked from those appointments.
Row 11, “Closed?” (INPUT): While you will likely know at the end of the day on the last day of the month how many leads were generated and how many appointments were scheduled, you likely won’t know how many appointments actually happened and you definitely won’t know how many moves have been booked. Our goal is to collect data on the full cohort of leads that came in, so it may take up to two weeks (or longer if customers aren’t moving for a while and want more time before booking). Leave this cell as “no” until you have accounted for every lead and appointment from this cohort. Once you have accounted for every lead, change it to “yes”.
Row 12, “Cost per lead” (OUTPUT): This formula is the total spend divided by number of leads. While this is a good number to keep track of, it is not necessarily actionable.
Row 13, “Cost per appointment” (OUTPUT): This formula is the total spend divided by number of appointments. This is a very useful number to know and you can use it to compare across lead channels. If customers are all properly (and equally) qualified, this is can be an apples-apples comparison for marketing channel efficiency.
Row 14, “Cost per booked move” (OUTPUT): This formula is the total spend divided by the total booked moves. This is the most important number, because it accounts for everything in the stack, including differences in how your contact center may be qualifying different lead sources. Again, note that you won’t fully know this number until maybe 2-3 weeks into the following month so it is important not to make decisions until you have “closed” the month by accounting for every lead. There is much more detail on this metric in our Mover Metrics: Cost per Booked Move post.
With this template, you are ready to start running marketing experiments. As always, P4P is here to help answer any questions on this or suggest new lead sources you can test out to see which may yield results.