It’s no secret that having a correct and complete data recording methodology is one of the main desires and interests of most online businesses, no matter how big or small. After all, collecting qualitative data not only saves you money, but could also bring you good business when used efficiently.
Over the past many years, I have encountered a number of extremely common issues across multiple companies, whether they are large, or small, and that are found regardless of whether the decision-maker is a web analyst, or a marketer. In order to truly have -relevant- data being tracked, and being tracked correctly, these issues need to be resolved every single time a new tracking or analysis project is begun.
THE RELEVANCE OF RELEVANT DATA
The first issue I am referring to is “knowing what and why you want to track something”. Or more simply put, relevance.
Sure, it may sound like this is a minor issue, and that -everything- that gets to the planning stage of a tracking or analysis project should automatically be assumed to be relevant, simply through the virtue of it being proposed by experts in their field.
Is it though? An extremely common issue that I’ve personally run into, and that I have seen others run into multiple times, is wanting to simply track -EVERYTHING-. Even when that has not been the case, and a less extreme situation has occurred, I’ve still been privy to many companies investing man-hours in the implementation of data tracking that led to just about nothing.
Does it matter? Yes, it does. Because those man hours cost the company money. It matters because simply having that tracking option that is truthfully irrelevant for the business will either lead to it never being used, in which case those hours spent thinking it up and implementing it were wasted, OR, even worse, analyses that have no lasting impact are then performed, leading to even more wasted hours. Reports become mottled by extra columns, analysts lose focus of the main goals that they should be looking at, and deep dive into non-actionable insights, and so on.
So how do we avoid these pitfalls? How do we make sure that when a department head, or a product owner, or a business owner asks us to track “everything” or “all the major elements” of all product pages, for instance, or all footer links, or any other myriad of elements that we can prioritize and only act on those that matter?
As with most things in life, there is, unfortunately, no one simple universal answer. That being the case, the best I can do is tell you my personal process of elimination, which assists me in the task of identifying the most useful tracking projects. In order to make these steps more tangible, we’ll use an example of an e-commerce website that we’re going to call matt.com, that sells everything from shoes to dog food, to high-end electronics.
EG: The department owner for the “Refrigerators” category reaches out to you, wanting to find out how many users check the Technical Details tab on the product page, for Products belonging to a certain brand, and then Download the PDF containing the Technical Specs.
Q1: How is this going to be used? The current setup for the page may require that the Description tab be always “on top”, and so might the brand. Is changing the elements going to be something realistically achievable?
Q2: How much traffic do those pages even receive per device? How long will it be before you get statistically relevant results?
Q3: Who is going to analyze the data, so as to know if it’s not you, if they have the necessary knowledge to find those reports, and interpret them correctly?
Q4: Most importantly, is there nothing else related to that category of products that would be more easily quantifiable and that would yield faster, more relevant results?
In the above example, you may reach the conclusion that the time spent implementing, reporting, and analyzing the requested data, is simply not worth it, when compared to an alternative.
Here are some of the steps and questions you need to ask in order to assess the viability of this tracking project.
- Ask what the future analysis should entail.
- Is it doable, or would it require additional work on the data collection infrastructure?
- Why does that department head/product owner NEED that element tracked? Don’t just ask “why”. That will easily be answered by “so that we know how many people do X or Y”. Instead, ask what you’re going to do with the information moving forward.
- Who is going to analyze it (process data), and when, and what would be actionable about it?
- Furthermore, could it be handled by a segment? If yes, it probably doesn’t require additional work.
KEEP IT CLEAR AND SIMPLE
Moving on, the second issue that needs to be resolved, in order to make your efforts worthwhile, is clarity.
Put simply, clarity refers to as detailed a brief as possible, with information ranging from what the end goal is, what eCommerce metrics to improve, as well as a working knowledge of the website and the functions that are going to be affected by your efforts.
This is mandatory in order to ensure that the results of your efforts will be usable, and that you don’t find yourself with a ton of data that is outright wrong, or incomplete, 3 months from now.
Ensuring the clarity of the request is not best done by asking what, but rather why. By understanding your co-workers’ end need, you will be able to better understand how to do things. With a clear understanding of the end goal, you can then identify opportunities to improve upon the request, and more often than not, correct it.
A department owner, for instance, may have a very singular desire, that is achievable either via the product page, or a separate area of the website. They may not be fully aware of all instances of this option on the website, and therefore, their request may be incomplete, as your tracking project may collect data from across the website.
EG: The department owner for the “Refrigerators” category reaches out to you, wanting to find out how many users check the “Reviews” tab on the product page.
Q1: Is the Reviews Tab accessible strictly via that tab button? Can it also be accessed via a link next to the product price, as is the case on most Ecommerce websites?
Q2: What about from the Listing page? If the general grade of a product (EG: 3.5/5) is visible and clickable from the listing page that the product belongs to, and on-click, that link takes users directly to the Reviews are of the product page, should that also be tracked, though separately?
Q3: If a product is relatively new, and hasn’t had the time to gather any reviews, do you not report it, or does it show up as a product that has received no interest related to the Reviews section?
This all culminates with the actual question of:
Q4: Is the interest in finding out how many users see the Reviews section (as an absolute number), or what the INTEREST in the Reviews section is (in the form of a percentage)?
With Q4 answered, you can then decide which elements should be tracked, and which shouldn’t, as well as how to then analyze the given data.
Here are some basic questions that you should be able to answer in order to decide how the tracking project should be implemented:
- What are we interested in analyzing? A user’s active intent, or just the consequence of an action? (EG: clicking on the Reviews tab vs. scrolling down in the product page and eventually reaching it)
- Are we interested in absolute values, or rates? (EG: Number of users that have clicked on a certain element? Or percent of users?)
- Is the element that we’re interested in always available for that user action, or is it only available in certain cases or situations? (EG: We are interested in the Click through rate for the Reviews Tab, but not all products have Reviews).
- Is the element or action that we are interested in available on all devices and screen resolutions?
- If the answer is yes, does it have the same level of accessibility, or should you record it separately, in order to better allocate a weight based on the difficulty to reach/perform that action?
By focusing your attention on these elements of Relevance and Clarity at the starting phase of any tracking project, you will be able to avoid innumerable wasted hours and ensure a higher level of quality for the overall output.
The importance of having correct tracking in place has been argued and preached over and over again, across probably hundreds of blog articles, presentations, meet-ups, and so on.
Similarly, there is a veritable cornucopia of health checklists available online, put forth by various experts and persons of interest in the field of web analysis. These can range from simply which settings you should watch out for, in your Google Analytics account, to how to create that tracking itself using tools such as Google Tag Manager.
With the advent of these articles, websites have seen an overall increase in the quality of data collection over the past years, leading to more clarity, and an easier and more complete understanding of who their users are, what they do on their website, what they enjoy, and what turns them off.