What is your favorite way to track advertising ROI?
I’d love to know people’s favorite way to track marketing ROI for a startup.The scenario is that it may take several visits to get revenue out of the
customer.Ad networks can add a UTM-type of code so your startup's WebApp can track campaign to users who are created.
What system do people like to use the best to track user accounts in their WebApp across many visits. Then map that revenue back to which campaign in their ad network.I’d love to hear what people like to use best for this end-to-end ROI tracking on marketing ad spend. Specifically when it takes several return visits to get revenue from the customer.
Consider looking into Segment or MixPanel. With proper implementation, both should enable you to track activity for individual users. The first time a new user visits your site a unique id will be assigned to their browser/cookie and all events triggered by that user moving forward will be aggregated under that id.
If your WebApp requires account creation to perform certain actions, you can set up these tools to tie unique Ids and their corresponding activity to the user accounts.
As long as you have UTM parameters present in all of your Ad URLs those UTMs will be recorded by the tools when users trigger events on your site.
Once you have all of the above you will be able to track and store the data and run different attribuion models against it to measure ROI across your advertising channels. At this point, the attribution model you choose is up-to-you, depending on specifics of your business. Here is an example of some of the most common attribution models.
I suggest to always run several models to get a better understanding of each advertising channel's contribution. For Ex: Remarketing would have more weight if your attribution is Last Touch and will have less weight if it is First Touch.
And Last but not Least: When it comes to calculating marketing ROI (Return on Investment) it should be calculated against you Gross Margin or Gross Profit, and not revenue. When you calculate marketing spend against revenue the metric is cold ROAS (Return on Ad Spend) and is often confused with ROI (as many people think it's the same thing).
How you track ROI really
depends on the type of product or service you are selling. One time
purchases are obviously easier and you can set up most conversion pixels to
dynamically pass the cart value to whatever tracking system you are using.
When it comes to
tracking lifetime ROI for a reoccurring customer or a subscription service, it
becomes a bit trickier but is certainly doable. I would recommend
creating a unique user ID that is passed via a URL parameter or through the
specific user to the method of promotion that brought them in and the (as long
as that information is also passed at the same time).
In terms of it taking
several visits before a customer converts, Google Analytics is great at
tracking this and quite easy to use, but there are plenty of other, more
advanced options that people swear by such as Convertro and Adometry. For
Google Analytics you could simply include a URL parameter for each promotion
method. If you wanted to get more granular you could also include
parameters for specific campaigns or even specific ads.
That said, being able to
track customer journeys is only part of the picture. The most important
question is to what extent each touch point influenced the sale. If you
know your customer saw a banner ad, then searched and clicked an organic link,
then finally purchased through a Facebook ad, it could mean a million different
things. It is important to look at different attribution models (last
click, first click, time decay, etc.) to see different views on what
combination of marketing tactics is the most effective. There is
unfortunately no on-size fits all answer this problem, which even fortune 100
companies are struggling with.
I would ultimately
recommend getting your feet wet with Google Analytics if you haven’t attempted
attribution already. There is a great deal of literature online on
setting it up to track attribution and can do the job well for most small