Drowning in Numbers: Making Sense of Customer Success Metrics


Share on LinkedIn

One of the biggest problems plaguing the customer success terrain is the abundance of data floating around. The general assumption is that the more data, the better. Supposedly, the more you know about what your users are doing in your website, the more you’ll be able to control traffic trends, improve conversion rates and impact the session duration/ activity ratio. But the fact is that the more data we are presented with – the more time we spend trying to analyze it.

It’s true that in 2014, an online marketing professional or account manager needs to wield a whole set of analytical skills that our predecessors knew nothing about. Then again, most of us (online marketers) haven’t the faintest idea how to approach graphic specs for print. That said, most of us aren’t exactly Data Scientists or statisticians, and we do need to know which figures to focus on when analyzing browsing patterns, and which figures determine our customers’ success.

Defining Goals
Always helpful, I’m sure you’ll agree. In this particular case, it is important to determine which stage of the funnel you need to look at, to help you decide what to do next. The differentiation is fairly straightforward, but not altogether intuitive:

    1. Inbound traffic – at this point, you are checking what works. What is it that brings traffic to your website, who your users are. These are mass-metrics, indicating trends, traffic waves and overviews. The parameters you will need to look at: (borrowing the terminology from Google Analytics) Acquisition, keywords, referrals, Audience.

    2. User engagement – you can now proceed to investigate what goes on within the 4 walls of your website. What engages your users the most, where your ROI efforts should be directed, what your users love. The main parameters: Behavior, landing pages, content drilldown, session duration, segmentation by events and timeframes.

    3. Customer Care – we now drill down to the tip of the funnel: figuring out who the leads worth cultivating are and what you need to do to convert them into paying customers. This metric group is much more user-centric analytics oriented, and the resolution you are looking at is on a much larger scale. The parameters for this metric group: user-centric data, individual browsing patters, statistics for conversion successes, drill-down churn rates data.


Focusing on the right metrics
Having determined what metrics we do NOT need to invest in, we can make room for the information we should focus on: customer success metrics. In essence, customer success is about helping your customers through the funnel by the path of least resistance.

This also means changing the way you look at customer service. Instead of resigning oneself to a traditional, reactive form of customer support – a healthy customer success approach should include pro-active customer service, i.e. using user-centric data to find out where in the software flow customers got stuck – and reaching out to them before they get a chance to practice their democratic right to cart abandonment.

So… what ARE the right metrics?
As previously mentioned, the customer success metrics are based on the need to identify user distress or activity stagnancy at the right moment. Timing is important, as it is part of a momentum that drives the activity flow. You can still send your users onboarding emails after the momentum has passed, but they are less likely to be as effective.

The main customer success metrics you will want to focus on are derived from the relation between user-centric metrics and browsing trends on your website. Or, simply put: you need to analyze an individual user’s activity in light of collective user activity patterns – to be able to predict if they are on their way to Churnville. Like so:

customer success

Most of us don’t have the time or skill to write an algorithm that can analyze those browsing patterns for us, but there are a few patterns that should raise red flags. Consider the following guidelines:

    Frequent visits to the knowledgebase coming to a halt. Trips to the knowledgebase are great: they indicate a healthy resourcefulness and a willingness to learn more, meaning – high engagement potential. It makes sense that at some point, the customer’s need to consult the knowledgebase cools down, as they become more familiar with the software. But it may also mean they have given up on figuring out the software and are resigned to, well, whatever. Be wary of knowledgebase quitters.

    Incomplete micro-conversions. The roadmap to conversion is paved with micro-conversions: specific actions, button clicks, certain page visits, small tasks executed. You will have created segments of these micro-conversions and are probably tracking them anxiously. In terms of user-centric analytics – create a roadmap of micro-conversions and track specific users’ progress on it. If they are not treading the obvious route, they are less likely to make the end-point of the journey.

    Email metrics. It sounds trivial, but it really isn’t. Mailing systems are one of those wonders that let you track an individual user’s actions through and through. Take advantage of that – check to see how far back their click-through rate goes, what they DID click, back when they still did, and try to figure out if there is a pattern: occasional clicker, used-to-click-but-does-not-anymore, clicks only very specific content.

Ultimately, conversion rates adhere to patterns just like any other human behavior on the planet. We are simply still too slow in recognizing these patterns and deciphering them. The technology that is meant to do the deciphering is pretty much here. But it is still up to us to teach it what to look out for, which activity patterns to pay attention to and what is it that defines customer success.

Noa Dror
Content Manager for iridize.com - I manage iridize's content strategy, blog and social media presence. Extensive background in data management, systems analysis and NGOs. Academic, with a flare for in-depth research and cautious disclaimers. Data geek. Closet misanthropist.


  1. Hi Noa: this post helped me think differently about ‘you can’t manage what you don’t measure.’ Of course, this assumes that what you’re measuring is worth managing – a trap that many people fall into.

    In managing revenue risk, it’s useful to examine the funnel, and to figure out what gives the funnel its taper. In a ‘perfect’ world, the shape approaches a cylinder, in which every lead that enters the top follows the use-case ‘happy path’ right to a purchase. That is seldom the case, so it’s worth answering what causes loss out of the funnel. In many cases, the top of the funnel can be needlessly large, which translates to wasted marketing spend, or a too-vague definition of qualified lead.

    Companies can run probability analyses to determine what might happen at specific points of prospect’s buying pathway. When these are modeled mathematically, companies can ask a range of questions that are invaluable for planning, including, “what’s the likelihood that we will have the required number of leads in the 3rd quarter? or, what is the probability that we will make our revenue goal if our qualified lead volume is 75% of plan?”


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here