Digital performance reporting is a major focus for most businesses with an online presence today and for good reason. If you do not have clear visibility into the digital activities you are engaging in, how will you know which are working and which are not? It is the aim of digital performance reporting to provide the data necessary to make these types of assessments. Thus the backbone of any good performance reporting program is the data it produces. Too sparse and there won’t be enough information to make educated decisions, too saturated and the truth may be buried and difficult to find.
When creating a report and determining KPIs to report on, you first need to ensure that you know your audience, goals of the program, and the types of information that relevant members need to make informed decisions. The reports that you present to upper management will most likely differ from the reports that you deliver to an analyst who may be interested in the more granular, lower-level data. An executive might be interested in viewing a high-level report on a monthly or quarterly basis, while another member finds data broken out on a daily or weekly basis valuable. Depending on your audience, you may have to create multiple reports to ensure the viewer is gaining actionable insights that they can act on.
Segment was deployed across the website and app, generating events based on user activity. These events were then used to build a comprehensive reporting system to track user behavior, the number of insurance policies sold, and total revenue generated. This allowed us to gain visibility into ROAs and help them allocate funds for paid and other marketing efforts more effectively.View customer story →
A performance reporting effort should focus on what actually matters with the end goal being to concisely convey how the business is doing to all relevant stakeholders. A website, app, or other digital property can have hundreds if not thousands of actions a user can potentially take. Too often we see tools such as Google Analytics, for example, cluttered with event tracking that is logging every micro-action on a site such as page scrolls, link clicks, etc. with no real structure or strategy in place. This can make it extremely overwhelming for a user to navigate in order to gain insights that really matter. When working through a performance reporting project, ask yourself these questions whenever you consider adding a variable to your data pipeline:
Amplify the value of your analytics and reporting tools with data that is accurate and impactful to your business.Learn More →
These questions are aimed at really drilling down to one of the core tenants of reporting: simplicity. Remember that the whole reason for reporting is to effectively convey information in order to provide insights needed to take action. It may take 20 different metrics to do this, but could it really only take 10? Often more information inhibits action. Not only does it take longer to process this additional information, but there is also a greater risk of misinterpretation. By limiting the information collected and processed and only focusing on what matters, you are actually able to improve time to action and thus remain more agile.
In close concert with the tenet of simplicity lies consistency. Think about paid campaigns that you may be running in multiple platforms like Facebook and LinkedIn, for example. Are you promoting the same offers to the same audiences in both, but have the campaigns named very differently in each? Taking the time to come up with consistent naming conventions that are descriptive will making future reporting and cross-platform data consolidation much simpler. By starting every project with the goal of having consistent data and naming conventions across all relevant sources, you are setting yourself up for success.
Often forgotten during inception but always felt down the line is the concept of how easy it is to generate or alter a report once everything is set up and in place. This does NOT mean making reports easy for everyone to generate, data experts exist for a reason. What it does mean is setting the system up in a way that allows for things to be added and changed without blowing up the whole system. For example if you track A, B, and C and all of a sudden someone says “we need to add “D” is it an easy add? What if the next week they say “add E”? It’s apparent that if these additions require an entire rework of the reporting set up they are not only less likely to get done, but also more prone to errors. This is not to say every variable is a worthy addition. As discussed earlier, keeping things simple is important but all businesses change over time and so will reporting needs. By keeping the reporting system flexible enough to accommodate these changing needs you are setting yourself up for success in the future.
“If only we’d known sooner”. The most terrible sentence to hear if you’re the one who has to suffer the consequences. Will you have to rebuild something? Will you have to rework the entire system? No matter the severity, this is something you are going to want to avoid. Not only does it mean wasted time and effort, but may also result in unusable data and a very frustrated marketing team. API changes, software bugs, and server issues, just to name a few, are all possible causes of data quality issues. To safeguard against this, we recommend that any tracking/reporting system have a robust alert system in place letting relevant users know of any major changes that occur.
Your reporting efforts are meaningless if you can’t convert the data that is collected into actionable plans. It is important to consistently monitor your data and look out for trends signifying that a program is working or needs to be further optimized. As mentioned earlier, depending on the program and audience it may make sense to monitor metrics often on a weekly basis or rather on a monthly or quarter basis to glean significant and meaningful trends. For example, an analyst managing PPC campaigns will want to review data more often where for an SEO program, key metrics should be monitored over months since SEO optimizations can take some time to affect performance. A reporting model that is helpful to use is the following:
After a meaningful amount of data is collected in the reports that you have set up, it is time to analyze these KPI’s and draw a conclusion on whether an initiative is working or needs to be further optimized to achieve the desired goal. A hypothesis should then be generated on why something may be working or not working and that is something that you will probably want to test. It could be that your messaging isn’t clear on your PPC landing page or that a CTA is non-existent. This is the time to take action and test your hypothesis. Once you make some adjustments and collect more data, you then observe your reports again and determine whether your hypothesis was right and KPI’s have improved or whether you need to test a different hypothesis. Setting up organized reports in an easy to digest manner is crucial so that the viewer can get the necessary insights they need to take action.