The Marketing Technologist.

We talk about analytics, code, data science and everything related to marketing technology. Backed by the tech geeks of Greenhouse.

Improve your customer journey with data collection

During my talk at Growth Hacking symposium in March, I presented a guide for setting up data collection for any project, startup or website. In this post, I’ll discuss that specific guide and help you set up your basic metrics.

Why should you collect data?

Let's start with the most important question: why should you set up a data collection solution? Well, it helps you make better decisions for whatever you’re doing. Without it, you might have an idea of what’s going on, but you won’t know any actual numbers. With data collection, you know exactly what’s going on and are able to improve your product in the best possible way. Data changes thinking something might work, into knowing something actually works.

The important metrics

For every website, app or other projects, there is a basic set of metrics that will always be helpful:

  1. Main goal
  2. First interaction
  3. The journey & micro Conversions
  4. Value

In the world of startups and growth hacking, metrics are often referred to as AARRR metrics or pirate metrics:

  1. Acquisition
  2. Activation
  3. Retention
  4. Referral
  5. Revenue

Though my metric list mainly focuses on acquisition, activation and revenue from the AARRR list, I think the biggest potential for improving any kind of project lies in the ‘journey & micro conversions’ step from my list.

Let’s elaborate on the four steps:

1 Main goal

For every project you start, there is always one important thing that users should do: they should achieve the main goal. For some projects, the main goal is easy to define:

  • Webshop: users should buy stuff.
  • App: users should install the app.
  • Service provider: users should get a contract.

For these main goals, it’s easy to capture the corresponding data. But there are also projects where your main goal might be less clear, e.g. a blog. At TMT we want people to read our content, but how do you measure read content? We’ve currently defined a 75% scroll and 60 seconds time on page as a full read. It's not perfect, but it allows us to capture the moment that people probably have read the article.

To clarify the way the metrics relate to one another, I'll add them to a simplified visualisation. The main goal is our first dot:

simplified main goal

A simplified view of the main goal.

2: First interaction

The second important metric is the first interaction. This captures all the first interactions users have with your brand or product. Think about a user who sees one of your advertisements or a search result on Google. The first interaction shows you how many people were exposed your brand, product or startup.

We’ll add it as a yellow dot to our visualisation:

simplified first interaction

A simplified view of the main goal and the first interaction

3: The customer journey and micro conversions

Now we know how many people see your brand and achieve a goal, it’s time to look at the customer journey: what happens between those the first interactions and the main goal?

simplified customer journey

A simplified view of the customer journey.

To analyse your website more effectively, you should identify key moments in the customer journey and set up metrics to capture data for these steps: these are called micro conversion. Micro conversions are not as important as the main goal, but they help you understand how far users get to achieving the main goal. If we take a standard webshop as an example, the micro conversions could be:

  • Website visit
  • Category page (orientation)
  • Product page (interested orientation)
  • Basket page (buying indication)
  • Checkout (strong buying indication)
  • Purchase (main goal)

Just as with the main goal, the micro conversions may be harder to set up for a website where the goal is less clear. If we look at this blog, we use the following steps to measure content performance:

  • Website visit
  • Article open (interest)
  • Article scroll (triggered to start reading)
  • 25%, 50% and 75% scroll (the content is good enough to continue reading)
  • 75% scroll + 60 seconds time on page (main goal)

Measuring content performance isn't as obvious as measuring a shop's performance, but the idea is the same: identify the substeps in your customer journey and measure them.

Let’s add the micro conversions a blue dots to the visualisation:

simplified customer journey with micro conversions

A simplified view of the customer journey with micro conversions.

4: Value

With the main goal, first interaction, and micro conversions set up, you’re all set to measure performance and see how new customers or users are coming in. Though this is valuable information for any startup or company, it’s important to track what happens after the main goal:

  • How many users that have installed your app keep using it?
  • Do new customers of your store return for more products?
  • How many people come back to read more content on your blog?

For value, we'll add a final purple dot to our visual:

simplified view of the customer journey with user value

A simplified view of the customer journey with value added at the end.
Tracking valuable users allows you to transform a large user base, into a healthy user base.

Bring in the calculations

With the data collection setup in place, it's time to start adding calculations. I’ll discuss two calculations:

  1. Performance
  2. Cost Per Action

These two calculations will help you grow your startup or company and increase marketing effectiveness.

1: Performance

It’s important to know how well your website is doing, to know where users drop off the most in the customer journey. Luckily we’ve set up micro conversions and with them, we’re able to calculate conversion rates.

The main conversion rate, how many users achieve the main goal divided by users that have had a first interaction, tells you how well you’re doing on the highest level:

  • How many users that see my app install it?
  • How big is the chance that someone will buy a product after seeing my web shop for the first time?
  • How likely are people to read a blog post after seeing it appear on Google?
    Though these are interesting metrics, calculations really shine when you apply them to micro conversions. We can calculate a conversion rate for each step in the funnel.

The main conversion rate can be calculated with the first interaction and the main goal:

simplified conversion rate

The simplified view of a conversion rate.

For detailed insights, you can calculate conversion rates for every step in your funnel:

simplified conversion rates per micro conversion

The simplified view of conversion rates per step in the customer journey.

It’s clear to see now what the biggest drop off point in your funnel is. In the example, the biggest drop off is highlighted in red, where only 10% of users convert to the next step. This is the best place to start optimising your website. If you double the conversion rate here, you’ll double the number of customers in the end.

2: Cost per action

The second important calculation is the cost per action. For all you’re marketing efforts you’ll spend money: the cost of running AdWords, Facebook Ads or any other kind of advertising. The calculations you can make are easy:

  • Cost per main goal
  • Cost per first interaction
  • Cost per visit to your website
  • Cost per micro conversion
  • Cost per valuable user

To clarify the value of this calculated metric, you should look at the following example:

simplified view of cpa's

A simplified journey per marketing source.

By calculating the cost per action on several steps you can see that different marketing sources serve different purposes. Based on the example we can say that:

  • Google has the lowest cost per visit
  • Google has the lowest cost per main goal
  • Email has the lowest cost per valuable users

Though banners are performing quite bad according to CPA’s, it will probably impact the other sources. Users might not have clicked on the ad, but if users saw the banner they have had a visual impression that might trigger them to click on a Google Adwords ad later on. So keep in mind that the mix of ads is important:

  • Prime users with banners.
  • Get new visits and users through Google.
  • Increase the number of valuable users through email.

Tools for data collection

There are several tools that allow you to collect the data you want, and implementations can be as complicated as you want it to be. To give you an idea of the variations, here are three tools that differ from one another quite a bit, but are all powerful in their own way:

1 Google Analytics

Google Analytics is a powerful and free to use data tracker. Implementations can be a simple as installing the standard pageview tag, to complex implementations with scroll reach tracking and enhanced e-commerce insights.

2 Heap analytics

Heap Analytics is a retroactive analytics solution. You install one simple tag on all pages, and you’re able to define and analyse every page view and click event retroactively.

3 Snowplow

Snowplow is an open source data logger and captures raw data. It doesn’t offer any kind of dashboarding and you should do all data modelling yourself. It takes more time to set up than the other two examples, but you’ll be way more flexible with the data you capture.

Keep in mind that these tools are just three examples, and you can use any data collection solution that you want. I personally suggest to use multiple trackers, as Google Analytics is powerfull for day-to-day analysis and offers aggregated reports, and Snowplow excels in advanced data analysis on a user level that is harder to do with Google Analytics.

Start capturing

In the end, it all comes down to knowing what’s important for your project, startup or website. If you know that, you can setup your data collection. How you capture the data is up to you. Just make sure that as soon as you start capturing data, you’ll make data part of every big decision you make.