We often want the tools we are using to be perfect, and be able to give us what we want. But sadly, that isn’t the case most of the time. Every tool has its own limitations, sometimes out of its control. For Web Analytics, it’s the same. Let’s go through some of the most common Web Analytics issues and concerns that we should be aware of.
1. First-Party vs Third-Party Cookies
In most cases, vendors will set their own cookies for tracking purposes, which are considered Third-Party Cookies. The problem is, many browsers and anti-spyware applications are built to reject and remove Third-Party Cookies, which include the ones we are using for Web Analytics. This is one of the contributing reason to incomplete data.
Having said that, some Web Analytics tools (e.g. Adobe Analytics) have provided the option to implement First-Party Cookies
instead of the legacy ones (Third-Party Cookies).
2. Incomplete Data
It is important to set the expectation right for all stakeholders. There is no good way to ensure that data captured is 100% complete. There are many ways data can fail to be captured:
- Poor network connectivity
- Incognito mode
But that is not an excuse for poor Web Analytics implementation. Besides the limitations that are out of our control, we should make sure that all tags are firing correctly to ensure maximum data completeness.
3. Data Ownership
Therefore, if having access to every single piece of data collected is important to you, it is important to do your due diligence and find out more about the data retention and data export policies.
4. Customer Privacy
Customer privacy should be your top priority when collecting data. Never collect PII (Personally Identifiable Information) unless you have the explicit permission from your customers. Even so, do not collect PII using Web Analytics, as they will most likely be stored on the vendors’ servers, thus putting your customers’ privacy at risk.
5. Cost and Price-Performance Ratio
There are generally 2 cost models: one-time payments and recurring payments.
Most vendors have a business model using recurring payments, ensuring that you pay only for what you have utilized. This can be good for smaller businesses, but for companies with huge amount of traffics, it is highly encouraged to negotiate for a one-time package.
Another option will be to house the Web Analytics solution in-house. That will also mean you have to take care of the various system administrative tasks that comes with an in-house model.
Also, evaluate a tool based on its merit. There are many free tools out there that provide you with basic reports. But those reports are 99.9% of the time useless and do not contribute to you gaining any actionable insights. For paid versions, there are always additional features that might or might not be useful for you. Evaluate each feature accordingly, and decide whether it is crucial for the success of your Web Analytics. There is no point paying for a tool that is cheap but does not provide anything but fluff metrics and reports.
6. Data Reconciliation
If you are using more than one vendor to satisfy company’s Web Analytics needs, there is a high chance that you will face issues with data reconciliation. Results from different tools can vary from 10-15% (or even more!).
Reason being, each tool has its own way of collecting data and defining metrics. Therefore, if there is a need to make use of more than one tool for Web Analytics, be sure to fully understand how each piece of data is collected by different tools and be willing to accept data discrepancy.
If possible, stick with one tool for all your Web Analytics needs.
Like any other tools out there, Web Analytics is not perfect and might never be perfect. As long as we are aware of the issues and concerns lying around, we can take proactive actions to achieve an highly optimized Web Analytics implementation.