Kissmetrics Academy

Learn how Kissmetrics can optimize your marketing

Built to optimize growth. Track, analyze and optimize your growth cycle.

How Onnit Can Use Kissmetrics to Drive Their Growth

Launched in 2010, Onnit is a fast-growing company that sells nutritional supplements, food, and fitness products.

Most products Onnit sells are their own private label, while a few are from third-party sellers. Many of Onnit’s products are consumables, meaning that customers need to reorder on a regular basis. Therefore, the key to Onnit’s success and growth relies on creating and maintaining a loyal customer base.

So, how can an ecommerce company like Onnit use Kissmetrics to drive their growth? That’s what we’ll be examining in this post. We’ll show you that:

  • Our funnel report can help identify major roadblocks to conversion
  • You can alleviate the roadblocks using people data and A/B testing
  • Our revenue report finds which advertising channels are the most valuable

Before we get into it, we need to be clear that Onnit is not a customer of Kissmetrics, and all the data you’ll see is hypothetical. We wrote this article to show how Onnit and companies like it can use Kissmetrics.

Let’s start with the funnel report.

Identify Which Areas Are Holding Back Growth with the Funnel Report

Before purchasing a product, a website visitor has to go through a series of steps. They have to visit the site, view a product, add it to their cart, and then purchase.

Marketers can use the Kissmetrics funnel report to identify roadblocks to purchase. When they identify the main roadblocks, they can run tests on those areas to increase conversions to the next step.

We’ll be going through two funnels that Onnit can use – a whole funnel and a checkout funnel. You can think of the whole funnel as a macro funnel. It’s a very high-level view of people who visit the site and end up converting.

The checkout funnel is a more detailed view of a few specific steps in the macro funnel. Since checkout is such a vital part of an ecommerce business (because of the need to close on the intent to purchase), it should be carefully tracked.

Let’s begin with the whole funnel.

Whole Funnel

We’ll set up a funnel to track how effectively the Onnit website converts visitors to customers. Since this is a macro funnel, we’ll be tracking the main steps in the conversion from visiting to purchasing.

So, here we go: A visitor lands on the Onnit website. This is the first step (Visited Site) of the whole funnel. Visited Site can occur on any page of the website, not just the homepage.

onnit homepage

The visitor selects the flagship product, Alpha Brain, and thus completes the second step (Viewed Product) of the funnel.

alpha brain product page

The visitor completes the third step (Added Product to Cart) of the funnel by adding it to a cart.

onnit added product to cart

After the visitor adds the product to a cart, they’ll go through the checkout flow and close their order, with Purchase being the final step. Remember, since this is a high-level funnel, we’re looking at only the main conversion steps. (We’ll view the checkout funnel in the next section.)

Now that we’ve gone through the steps, we can view our hypothetical funnel.

onnit purchase funnel kissmetrics

The overall conversion rate, 3.6%, represents the percentage of people who purchased after visiting the site. Within that main conversion, there are 3 steps, each with their own conversion rate:

  • Visited Site > Viewed Product (85.6%): If this were your data, you’d know that you were doing a great job of getting your traffic to look at individual products. Your navigation, homepage, and landing pages are all super healthy. This also shows high product awareness in that people who come to Onnit have at least one product in mind they’d like to view.
  • Viewed Product > Added Product to Cart (33.2%): This is where conversions take a little dip as people are more hesitant to hit the “Add to Cart” button. This doesn’t necessarily mean it’s a roadblock. Many people may be just looking for information on products and not seriously considering buying.
  • Added Product to Cart > Purchase (13%): This is the biggest roadblock to conversion in this example. Once visitors add a product to their cart, they have the intention to buy. Since only 13% of the people who put a product in their cart end up purchasing, we see this as a major roadblock that needs improvement. If this were your store, you’d want to dive into your checkout funnel in more detail to see exactly what was pushing people away. So, let’s build a more detailed funnel to take a look now.

One important distinction about this funnel report is that it doesn’t matter if the visitor views a product page, then goes back to the home page, and finally purchases a few days later. Kissmetrics tracks and counts those visitors through their entire journey down the conversion funnel.

Checkout Funnel

If your funnel had a bottleneck between adding a product to cart and purchasing, you’d want to take a much closer look at your checkout funnel. This is where you can zoom in and create a Kissmetrics funnel on those individual steps.

Before we build the next funnel, let’s review the exact checkout process so we know which events we want to include in our funnel: When a visitor adds a product to their cart, they see this notification page. This serves as the first step (Added Product to Cart) of the funnel.

onnit added item to cart new customer

After the visitor selects Continue To Checkout, they either sign in or register for a new account.

log in create account onnit

Once they do one or the other, they are taken to the second step (Payment Page) of the funnel.

onnit order confirmation page

They’ll enter their data and click Continue, which enters them into the third step (Confirmation Page) of the funnel. They’ll complete the fourth step (Purchase) by clicking the Place Order button.

onnit final purchase screen

Here’s how that funnel would look, with our hypothetical data:

onnit checkout funnel kissmetrics

  • Added Product to Cart > Payment Page (27.7%): If this were real data, we’d instantly know where the problem was. The bulk of people who add a product to their cart do not even continue to the Payment Page. This is clearly the largest roadblock in the funnel.
  • Payment Page > Confirmation Page (58.7%): There is strong conversion here, but it should be improved. Once people get this far into the checkout, few should exit. Onnit can test alternate payment pages to try to increase conversions. (In the next section, we’ll address in detail how Onnit can use Kissmetrics to resurrect these abandoning customers.)
  • Confirmation Page > Purchase (80%): Once people reach this page, the majority of them order. It’s a high conversion rate, but it could be improved since the only thing that’s required is clicking a button.

Now, in the cases of cart abandonment, one of the best practices is to email customers to get feedback and try to resurrect the order. In addition, it would be helpful to know exactly what customers were doing around the time of abandonment and shortly thereafter. With Kissmetrics, we can do just that!

Confirmation Page > Purchase

One of the greatest benefits of Kissmetrics is its people tracking platform. Each visit is tied to a person and not a pageview. Because of this, you can see who progresses through your funnel and who falls off. Since we want to resurrect our cart abandoners, we’ll hover over our Purchase step and select the people who are not in the Purchase step:

people not in funnel step

We can view the people who abandon at any step in the funnel, not just the last step.

Click the link “View the people not in this step,” and you’ll be taken to a list of all the people who bounced from the Confirmation Page step. Keep in mind these are fabricated email addresses. They are not actual Onnit customers.

list people from funnel kissmetrics

Now that we have the email addresses of visitors who bounced, we can send them a note asking if there are any problems with their order. Using the person details report, we can find which products they added to their cart, as well as any other events and properties they triggered after signing up.

For instance, if the visitor got all the way to Confirmation Page but has since been shopping around for other products, we know they may be holding off on their order for a little while longer. In this case, we wouldn’t need to email them. But, if the visitor hasn’t been seen since they left the Confirmation Page, we’ll want to send an email and try to resurrect that order.

Added Product to Cart > Payment Page

added product to cart payment page

There are many possible reasons conversions are low in the second step (Payment Page) of this checkout funnel. Here are a few common traps that prevent people from converting:

  • Customers are continuing to shop around before they log in or register. This won’t be a problem for Kissmetrics because it tracks the same person over multiple visits. Their data won’t be erased if they visit, then leave, and later come back.
  • New visitors didn’t want to register for an account before purchasing.
  • New visitors wanted to register for an account, but were turned off by the long registration page:

onnit registration page

In this case, Onnit can test alternate registration pages against their current registration page. Perhaps, breaking down each section (email, address, contact information) into smaller steps will yield greater conversions to purchase. Luckily for Onnit and all other marketers, they don’t have to leave Kissmetrics to track their tests.

With the A/B test report, you can set up your test in a tool like VWO or Optimizely and track the results in Kissmetrics. What’s more, you can track the results through the entire funnel. Let’s go through an example using a variation of the registration page.

If we’ve integrated Kissmetrics with Optimizely, we can set up a test there and track it in Kissmetrics. (The A/B test report does not set up tests. You can still use your favorite tool for that.) The advantage of tracking the test in Kissmetrics is that you can run a report for any part of your funnel to see how the tested change affected it. All you have to do is select the conversion event and the name of the test:

ab test report kissmetrics start

Whatever we name the test that will be its name in the A/B test report. Let’s call it the “Payment Page Test.” The conversion event is the outcome we’re testing for. In this case, we’re using Payment Page because that’s what people see after they add a product to a cart and sign in or complete the registration page.

ab test setup event test

We’ll select Run Report and view our hypothetical data:

ab test report onnit kissmetrics

There are three sections to this report layout:

  • Across the top, highlighted in blue, we have the summary. If the test has run long enough and reached certainty, Kissmetrics will give you a recommendation to either use a variation, stick with the control, keep running to test, or move on to another if it hasn’t reached certainty after a long enough time. Below that, we have a summary of the test specifics: how long it ran for, how many people were in the test, the total conversions (for both the control and variation(s)), the percentage improvement the variation has given, and the statistical significance. In our test, we see the variation beat the control by nearly 32%.
  • The bulk of the report is the visualization in the middle. Track how the control or any variation did along the time span of your test. Throughout the lifetime of this test, the variation consistently converted more people than the control.
  • Across the bottom, we have a summary of the control and variation(s), including each metric for the control and variation(s).

Great, at this point, we know the impact of the variation on getting people to the next page of the checkout process. But, now, let’s check its impact on purchases (our real goal) by setting up another A/B test report.

While the test variation broke the registration page into shorter steps and converted more people to the Payment Page, it may not have actually led more people to purchase. This is why the A/B test report is so useful: It allows you to run a report for any part of the funnel. We simply change the test to the conversion event “Purchase,” and all the people who were in this test are then measured for the Purchase outcome.

payment purchase onnit kissmetrics

Here’s how that would look:

ab test report purchase onnit kissmetrics

Now that we have a report for Purchase, it looks like the variation still converts more than the control, albeit at a smaller rate than the Payment Page conversion. It still hasn’t reached certainty, but considering that more people have converted after 2 months, we should go with the variation. Since this is the Purchase outcome, any gains we can make should be kept.

This is why it is so important to check multiple parts of your funnel. What if we had run a report for only the Payment Page and saw it was a big winner, but actually fewer people had converted to Purchase? We would be hurting our growth if we started using the variation.

You can test any part of a funnel. Would you like to see if a homepage redesign increases purchases? Run the test to find out! Run a test at the top of the funnel and see if it impacts the bottom of the funnel.

Gain Insight into Which Products and Promotions Bring the Most Revenue and ROI with the Revenue Report

The Kissmetrics revenue report gives an overall picture of your revenue by displaying various metrics and graphing revenue over time. The best part is that it allows you to segment your revenue. Continuing to use Onnit as an example, we’ll segment by product and discount code in this section.

Before we get into that, we need to set up our report criteria:

revenue report ecommerce set up kissmetrics

We’re calculating revenue from a property that records the amount of money received from each order. (A property is a characteristic about a person, such as where they came from or the first product they purchased.)

We’ve set the date range to the entire 2014 year, and we’re calculating churn by 30 days. This means that if a customer hasn’t reordered the same product within 30 days, they’ll be considered churned. We hit Run Report and get our hypothetical revenue. Remember, these are not actual revenue figures from Onnit.

revenue report visualization kissmetrics

Below this visualization, we get the option to segment people by a property. We’ll first break down the revenue by the first product that people purchased:

revenue report segmented by product onnit kissmetrics

With each product purchased, we have a row of metrics that includes the revenue we’ve received, the average revenue per person, the lifetime value we can expect out of that customer, the total number of customers each product has, and the churn rate.

Do you remember the 30-day churn limit we set when we first created the report? That means that a cancellation will trigger if a customer hasn’t repurchased the product within 30 days. We’re setting it to 30 days because many of Onnit’s products have a 30-day supply, and we want to view customer loyalty for each of those products. For the products that have a larger supply, we can set churn to 90 days.

Also, we can see if there is a measurable difference in churn for the 30-day products (i.e., a higher churn rate for the 30-day limit vs. the 90-day limit for the same product). If churn is significantly higher, we will know that customers order the 30-day products on a less-frequent recurring basis than the 30 days’ supply would indicate.

The three big metrics you’ll want to focus on are total revenue, average revenue per person, and churn rate. Total revenue will tell you which products bring home the bacon. Average revenue calculates the average amount of revenue you’ve received per person. It’s a simple calculation: total revenue / paying customers.

Churn rate (which triggers if a customer hasn’t reordered the same item within 30 days) is a super important metric to keep an eye on, especially for an ecommerce company like Onnit. These companies need loyal customers. Too many one-order customers, and their business will suffer. This is why Onnit set up recurring purchases.

They promote their recurring purchases feature pretty heavily. You see it on their product pages:

onnit recurring subscription purchase

You also see it on their Payment Page:

onnit offers checkout recurring

Churn is the best tool for gauging customer loyalty through recurring purchases. Since the revenue report breaks down each product by churn, Onnit knows which products customers stick with.

Let’s quickly break down the data:

revenue report segmented by product onnit kissmetrics

  • In this hypothetical case, we can see that customers love the egn (Earth Grown Nutrients) product. We know this because it has the lowest churn rate. Once people buy it, they stick with it and keep reordering it. If Onnit could get more customers to buy this, they could be pretty confident that those customers would become long-term customers.
  • Alpha Brain is the mothership of Onnit. It brings a large amount of revenue, has a huge customer base, and a great lifetime value. Also, the churn rate is one of the lowest.

Track Advertising Effectiveness with the Revenue Report

Onnit advertises on about a dozen podcasts. Each podcast advertises its own unique discount code, which customers apply on the Confirmation Page. Using the revenue report, we can segment revenue by each discount code to learn which podcasts bring the most valuable customers.

To get the data, we just change the segmentation from “Product” to “Discount Code.” (This data is not automatically tracked in Kissmetrics, but it can be set up with a little help from a developer.) Here’s how it would look:

revenue report tracking advertising kissmetrics

Please note that these may or may not be legitimate discount codes. We are using them only as examples with our hypothetical data.

Let’s break down the data:

  • The bulk of customers who applied a discount code did so using the rogan discount code. It brought the most revenue, the largest number of customers, and the highest lifetime value of all the discount codes.
  • The most loyal customers (lowest churn) came from the carolla discount code. However, that code has the lowest revenue per person metric.
  • The wodcast code has a strong revenue per person, but the highest churn rate. If we could get these customers to stick around, the code could become one of the best performing advertising channels.
  • To know more about which podcasts bring the best and worst ROI, we’ll need to look at what we’re spending.

How Kissmetrics Fits Companies Like Onnit

Let’s quickly review the main points:

  • Onnit is an eCommerce company that sells nutritional supplements, food, and fitness products. Kissmetrics is a tool that gives online marketers data they can use to optimize their marketing.
  • Using the funnel report, Onnit can discover which pages are a roadblock to purchase. Once the roadblocks are discovered, Onnit can obtain a list of the customers who abandon their carts and then email them to try to resurrect the orders.
  • To improve a roadblock, Onnit (and any other Kissmetrics customers) can run an A/B test in Kissmetrics and track any conversion step in any part of the funnel. Be sure to track more than just how many people convert to the next step. Also, start tracking more impactful areas of your funnel.
  • The revenue report graphs revenue over time. You also can use it to segment your revenue. Make better business decisions by learning which products drive your revenue and which advertising brings you the most valuable customers.
  • We didn’t even get into all the things you can do with the metrics dashboard, the path report, or the cohort report.

Do you want to really know which online marketing activities and website configurations are actually impacting your bottom line? Don’t lose another dollar – give Kissmetrics a try right now.

  1. Mansal Denton Nov 08, 2016 at 9:03 pm

    Wow! This is insanely comprehensive. Also, seeing some of their stats (both 85k+ hitting add to cart and 23k+ making it to checkout) is ridiculous. They have some huge volume – no wonder they’re doing well.

    At this kind of scale pretty much any business can make some significant gains from the small tweaks on Kissmetrics and similar sites. Kudos

2 comments

Please use your real name and a corresponding social media profile when commenting. Otherwise, your comment may be deleted.

← Previous ArticleNext Article →