Volument data model
How Volument collects, analyzes and visualizes data.
The conversion funnel
Take any modern book about marketing, and you get to know the concept of “marketing funnel” or “conversion funnel” where people go through a series of steps before taking action:
While the names of the steps can vary, this generic AIDA funnel has proven its worth across the decades. It can be applied for all businesses, because ultimately they all want the same: they want to build something people actually want so they can get more conversions as a result.
The AIDA model changes the way you do analytics. You no longer need to handpick custom metrics that are specific to your business. You no longer need to build a custom infrastructure or hire dedicated people to deal with your “unique” requirements.
Volument is a simple, all-round solution for conversion optimization that helps marketers, designers, and developers build interesting and desireable products. Volument is the first one to apply AIDA funnel into the world of web analytics.
One significant difference in Volument and other analytics tools is how Volument tracks data. Instead of studying individual data points on a time range, Volument collects all visitor activity starting from the very first touch with the product, and pushes all the events on a time-series as follows:
This way, Volument can understand how the visitor starts building awareness, interest, and desire before taking action — i.e., how the person proceeds along the AIDA funnel.
Of course, the above customer journey is from a delighted visitor. The majority of journeys, however, look like this:
Market segments in Volument are a set of customer journeys that are placed on the same line as follows:
By studying all the journeys from the beginning of the relationships, we can see how certain groups of visitors, or market segments, proceed the AIDA funnel.
The most important segments are visitors that come from a particular referring domain, such as
slack.com, or all visitors that entered the funnel trough some landing page, such as the front page.
Traction is the sum of all the conversions that an individual market segment generates over the lifetime of their visitors:
This weighted score is the single best metric for tracking accumulated visitor value for a product. It ranks your market segments from good to bad. It finds your absolute best traffic sources and landing pages, and all your optimizations strive for maximum traction.
Traction is the One Metric to Rule Them All
Author of Running Lean
Visitor lifetime is the total time how long the visitor engages with the product. Volument calculates this metric by taking all visitors in a sample set, look for how they come back, and see when 99% of the people stop visiting.
A lifetime may sound like an eternity, but the web is ruthless: based on our data, your website is doing rather well if you can retain new visitors for more than three days.
The above L-shaped curve, how people lose their interest in a new product over time, is a typical pattern. It seems to work like a clock. Just think about your behavior: when you see a whole new website — when was the last time you went back to it on the next day?
The longer you can retain visitors, the more traction is generated. Because visitors who keep coming back are more likely to convert, have more chances to invite other users, and you have more opportunities to monetize them.
Visitor lifetime makes the biggest impact on traction.
Engagement refers to your website's ability to hold visitor attention. That is: the longer they stay, and the more content they consume, the more engaging the experience is.
Another big difference between Volument and others is how the page view data is collected. Volument monitors all mouse movements and keyboard usage to check whether the visitor is actively engaging with the content or whether the visitor is doing something else, such as daydreaming or making coffee.
This activity monitoring gives us a unique ability to see how the visitors pay attention to the content.
While general analytics focuses on the key conversion events: signups, purchases, and shares — Volument is the polar opposite: it focuses on what happened before those key events occurred.
That is: Volument focuses on the leading indicators, and general analytics focuses on the lagging indicators.
Leading indicators are the holy grail of optimization: they offer a way to understand why people convert and they are actionable: you can actually impact them.
You can make such changes to your site so that the product becomes more interesting and desireable. But whether the visitor takes action is ultimately their decision (as long as your action buttons are available and they are working).
The first visit is the most important source for leading indicators: it's when the visitors become aware of your product, form their first impression, and start building interest.
Perhaps the most important feature of Volument is the ability to extract engagement bottlenecks from the data — the spots where the conversion funnel is loses (or “leaks”) most of its visitors before they proceed to the next step.
These leaks are your worst-performing landing pages where new visitors leave the site in just a few seconds (3, 7, or 15 seconds), even before they become aware of the product: what it does and how it is useful.
The landing page with the largest number of these rapid bouncers is your bottleneck. You fix bottlenecks by improving the first impression so that more people become aware of the product and can proceed to the next step: building interest.
Measurement is fabulous. Unless you're busy measuring what's easy to measure as opposed to what's important
Volument is a dedicated solution for conversion optimization designed to solve all the known problems in traditional A/B testing. Here's how it's done:
Use of traction as the goal
Unlike traditional A/B testing, that tracks an individual metric over a time range, Volument shoots for maximum visitor lifetime value. Your optimizations are likely to make a much bigger impact because:
- You have a list of bottlenecks to start from where you can make the biggest wins with the lowest possible effort.
- The optimizations strive for maximum traction: maximum engagement, retention, and virality over the lifetime of the visitor.
New ways to optimize
With Volument, you are not limited to small changes on some individual pages. You have three powerful optimization methods at your disposal:
- Market discovery — find people who want your product the most.
- Page optimization — fix your bottlenecks to build more awareness and interest.
- Site-wide optimization — optimize your site for increased traction.
Volument is an order of magnitude faster than traditional A/B testing. Here's why:
- Uses leading indicators for sampling. Instead of waiting for key conversions that happen rarely, Volument pulls out results from each page view. Youll get the results faster, because these leading indicators make the bulk of your analytical data.
- Half of the data already exists. when you want to optimize your front page, for example, Volument already knows how the current version performs. And when you push out a new version of the front page, we can simply compare the original data with the new. With traditional A/B testing you start collecting data for both variants at the same time. This doubles the required time to collect samples.
- Low sample variance. With Volument, you no longer need to wait for the results to settle down after the wide gap in the early stages of the experiment. Volument A/B tests are “apples-to-apples”: both variants measure the conversion flow from the first touch to the end of the visitor lifetime. You'll get sane results early on.
Run multiple experiments at once
With Volument, all your A/B tests are isolated from the rest of the site so you can run multiple experiments simultaneously.
In fact, you'll have hundreds of experiments running out of the box, because all pages and marketing segments collect data without any manual setup. You know your best-performing landing page or referring site without any extra effort. Volument gives you a good idea about the kind of content/market combination that works in a matter of days after you start using it.
Volument A/B testing is no longer a separate task that blocks everything else. You can, for example, take product milestones from your GitHub commit history and bring them to Volument for inspection. You can check what works and optimize accordingly.