The funnel has become a key part of the modern marketing toolbox. This is the first report they read on Monday morning, and the last report they check before leaving for the weekend on a Friday.
All serious analysis tools have a version of a funnel report, with varying degrees of flexibility and features. Some are narrower, allowing a limited number of use cases, while others are more complete with advanced conditions, date ranges, and advanced segmentation.
Kissmetrics has its own report on the funnel. It is flexible and has many features that you can use right out of the box.
The Kissmetrics Activity Report is a bit of a natural extension of the Funnel Report. It's great to segment into one or more events.
This article will show how marketers and growth teams can use the funnel report and activity in tandem to gather even more accurate information through segmentation. Let's see how.
Funnel Conversion Report: Identifying Opportunities
The use case of the standard funnel follows the key steps of the first visit to the conversion. A funnel for a SaaS company may have these steps:
- Visited website – They visited your site but they are not converted yet.
- Signed for the test – Visitors have registered for a free trial of your product.
- Activated – An activation of the product has occurred. They installed JavaScript, added a member of the team, etc.
- Billed – They decided that they wanted to keep the product and started paying for it after the expiry of their lawsuit.
Fallout can occur anywhere in this funnel. Most sites get a conversion rate of about 3%, so the biggest "trough" will be 1-2. But if steps 2-3 also have a big dip, this could signal a problem with your integration or your marketing is listing unskilled people.
An e-commerce funnel may look like this:
- Website visited – Same as the funnel above – they visited but did not convert.
- Product seen – This may or may not be in most funnels, but it is a necessary step in the funnel.
- Added to Cart – they showed enough interest to add a product to the cart.
- Purchased – Ordered product (s).
Fallout in this funnel may occur in steps 3-4, and here it is important to "zoom in" with a funnel and visualize each micro-step that occurs between 3-4. There are usually a lot of steps that customers must go through after adding an item to the cart and before they buy, so this can help create a funnel just for that stream.
Example
You are the marketing manager of a SaaS company, and we are Friday afternoon and you analyze your analytics data for a week. You pull out the Funnel report and spot your opportunity:
This is really not a bad funnel. Most SaaS companies would like a conversion rate of 7% +. And further down in the funnel, there is a solid conversion rate to activate the product (these conversion rates will depend on the complexity of the product and the conditions of the event activated ) . The number of test users who perform the upgrade is good at 7.7%. A better test experience, attracting the right customers, achieving a product / market fit, and tackling the biggest problems faced by customers can improve this conversion step.
But there is always room for improvement.
For now, we will focus on improving our listings. Most conversion rates are improved by the a / b tests, but in this post we will explore this event signed .
Activity report: Exploration to understand what drives fallout
To use the activity report, we will simply select the date range and the event. Since we want to retrieve our funnel report, we will select the Signed event and use the same date range. We will launch the report and here is what we get:
This shows the daily entries for this week. We can see that it reaches its peak in the middle of the week and then falls, reaching its lowest on Friday.
Now, let's explore to see what motivates these numbers. Although we have many properties, only a few properties can be the cause. Since we are dealing with an event marketing (a person who registered), we will use a property marketing . I like our property Channel . It divides visitors into 6 different channels according to their origin.
An extension of Channel is Channel: Origin . This list lists the channel and the corresponding URL or the name of the campaign. If a visitor came from nytimes.com then they will have the Channel: Origin is Sponsor: nytimes.com .
Now that we have this, we will use the Channel: Origin property as the "first step" to break down listings for this week.
Here is what we have:
The lower part of this image is the top 3 canal: origins . We see that organic search is where most people come from, then AdWords, and the third is direct. The graphic is visualization – showing us how they interact with each other.
Just at this level, we can already see that our listings are directly correlated with the organic search traffic we get. If you scroll upwards, you will see that the global and organic inscriptions: google are linked together.
Under this section we obtain the figures:
We see that there are 103 different channels : the origins . We only look at our first few so we can get an idea of what is sending us registrations.
Let's evaluate the Copyblogger referral further. To do this, we will click on Add a value and add this property:
We will click on the Copyblogger option and get the visualization:
Compared to our others, it does not add much. Our first three channels are the ones that really remind you of bacon. And since organic is our main channel, we know that we must continue to circulate our organic traffic, otherwise our business will be affected. This can be considered as a risk factor for the company.
Conclusion
The right analytical tools do more than push the data.
They help you understand what drives these numbers by segmenting each metric and KPI you have.
Kissmetrics has a segmentation on every applicable report. The funnel report already had top-level segmentation, but with the Activity Report, you can do a deeper analysis to really understand what drives (or removes) the conversions.
About the author: Zach Bulygo (Twitter) is the blog manager for Kissmetrics.