It’s hard to overestimate the value of proper analytics configuration. A thoughtful approach will result in detailed insights into the day-to-day behavior of your clients, data-based ideas for improving your business, and many more positive outcomes – things everyone can benefit from.
- Fix Poor Website Performance with Piwik PRO Analytics Reports
- How to Successfully Incorporate Analytics Into Your Growth Marketing Process
- How to Leverage Piwik PRO Pages Report for Conversion Optimization
- How to Maximize Your AdWords Campaign ROI with Piwik PRO
and many, many more.
However, skillful implementation of analytics is no walk in the park. You have to take many factors into consideration – like your business objectives, the type of data you want to track, and legal restrictions involved in dealing with particular types of information about users. Also, there will be a lot of writing, testing, and double-checking.
Don’t worry, we’ve got you covered! In this blog post we’ll walk you through all the stages of analytics implementation.
The tips gathered in this article will benefit both people starting their journey with product analytics and those who want to measure their website’s performance (web analytics).
You’ll also find a useful spreadsheet to help you create your own tracking plan.
So, settle in and let us show you the steps you should take to create a bulletproof analytics implementation plan.
1) Define your business goals and KPIs
The first thing to do is defining what you want to achieve using analytics.
At this stage, you should think about why your website or app exists. What are the primary goals? What subgoals are necessary in order to achieve the desired results?
For example, if you’re a healthcare provider, your main goal might be to sell more medical packages via online channels. And if you’re a bank, you might be aiming for an increase in online mortgage sales.
If you want to read more about setting effective business goals, we recommend these blog posts:
Track Piwik PRO Goals and Improve Your Online Business:
After you have your business goals in place, it’s time to assign your KPIs.
It’s important to note that your KPIs should help you not only connect the dots between stages of the customer journey, but also between every touchpoint where your clients connect with your brand, like your app or website. Your KPIs should reflect the complexity of this issue.
Here are some examples of appropriate KPIs:
- Task rate completion
- Conversion rate
- Cost of client acquisition (CPA / CPS)
- Customer retention rate
- Number of active users
- Revenue / revenue per user
- Percent of new / returning users
However, your choice will depend on your own needs and requirements – the industry you’re in, the purpose of your business, etc.
Psst! If you operate in the banking industry, we think this ebook will serve as great inspiration in determining your KPIs:
15 KPIs to track for E-Banking and Mobile Banking.
It’s normal that at the beginning of your journey with analytics your KPIs and goals will be quite high-level. After all, you don’t have a lot of data that would allow you to come up with more in-depth ideas. But don’t worry, you’ll get there in a couple of months.
2) Analyze your website/product/app structure
Next it’s time to investigate the structure of your website, app, or product. At this stage, you should list all its features, then break them down into every action your users perform while using them.
Treat each feature as a separate funnel and make a list of everything your user has to do in order to send a report, create a playlist, photo album, contact your sales representative, etc. Start from the essential features or website areas, then move to the ones you consider less important.
For instance, this is how we described what users tend to do in our Analytics “Email reports” section to complete the task. They:
- view “Email Reports”
- click “Create and schedule a report”
- write a description
- choose a segment
- schedule an email
- send a report at X o’clock
- send a report via mail/mobile
- report format
- send report to…
- display options
- select statistics
After they’ve done all those actions, an email report is created.
Also, when it comes to web analytics, it’s advisable to use this step to analyze every section and diligently examine their environment.
Seek out any potential hurdles like multi-domain structures, the usage of ajax, frames, Flash, or any other technology that requires the creation of custom analytics scripts.
Write it all down – this list will come in handy in the next steps of the implementation.
3) Operationalize your goals and KPIs
Now it’s time to translate your business objectives and KPIs into the things you can do and measure using analytics software.
In the case of product analytics, you’ll have to translate every user action into an event and decide what exactly is going to trigger each of them.
Then you’ll have to choose which events are more important than others and designate them as your goals. Every micro-conversion will also be a valuable source of information, but it’s your goals that will tell you if users have completed vital tasks in your software.
If you want to learn more about event tracking, we recommend you check out this exhaustive blog post: Ultimate Guide to Piwik PRO Event Tracking.
- Happiness: measures attitude or satisfaction (typically through some sort of user survey)
- Engagement: measures the frequency, intensity, and depth of user interactions with a product
- Adoption: measures how easily, often, and quickly users adapt to the new features of the product
- Retention: measures how many existing users retain in a certain amount of time
- Task success: measures the effectiveness and efficiency of the tasks users complete within your application
The main advantage of the HEART framework is that it’s extremely user-centric and allows you to measure the quality of the user experience. What’s more, it’s a very flexible approach. It can be used to evaluate and measure overall advancements made towards the main objectives, or simply to evaluate the performance of each feature separately.
Sounds interesting? You can read more about it here.
With web analytics you’ll have to decide which metrics are best aligned with your business goals and KPIs.
Of course, you can measure virtually anything that happens on your website, but should you really do that? You’d better take another close look at your online business objectives. Make sure that you’re going to track only things that are consistent with your goals and KPIs.
Some of the best ways to track your goals in web analytics:
- URLs – keep track of specific URLs. Each time someone visits a particular URL, they trigger the goal. These are ideal for thank you pages, confirmation pages, and PDFs.
- Time (visit duration) – track how many people stay on your site for a certain amount of time.
- Pages per visit – instead of tracking how much time people spend on your site, this goal tracks the number of pages each visitor sees before they leave.
- Events – these come handy in multiple places, especially where you can’t really settle for anything else. With Event Tracking you can record clicks on both clickable and non-clickable elements of your website, usage of video and audio players, scrolling down, leaving the page, filling out a form without submitting it, interacting with Flash elements, and downloads from your website. Even though these actions don’t generate changes in the website’s URL, they can be still recorded.
4) Create your tracking plan
After that, it’s time to write down everything you want to measure with your analytics and include it in your tracking plan. You can do this in various ways, but creating a spreadsheet seems to be the most convenient one.
A tracking plan clarifies what events to track and why those events are necessary from a business perspective.
Don’t even think about skipping this step and going straight to tool configuration. This document will serve you for years and you will modify it every time you change your analytics strategy. Without it, you’ll easily forget about your main goals or even overlook some important features whose performance you definitely want to measure.
When creating a tracking plan, consistency is very important – if you don’t develop a unified system, then later you’ll get lost in duplicate and unintuitive names of events, website/products sections, and so on.
So, come up with one naming convention and stick to it when drafting your tracking plan.
The way you organize your documentation will obviously depend on what you want to measure.
Need a Help with That?
We think that for product analytics implementation one of the best solutions would be to divide it into the following groups:
- Name of the product section
- Name of the funnel
- Name of the event
- Conditions for firing an event
- Event properties and values
- Additional notes
Important tip! When it comes to product analytics, it’s convenient to separate what people are supposed to do in the onboarding phase from the rest of the user’s activities. That way, you’ll be able to distinguish the actions typical for new users and for users who have been with you for a while.
As for web analytics, apart from events you’ll also include every other analytics goal type you want to use to measure website performance, together with the properties and values attached to it.
5) Choose your weapon (tool)
Now when you finally know what you want to track using your analytics, it’s time to find a tool that will meet your expectations.
Of course, there are many more factors you should consider – including data privacy regulations (like GDPR or HIPAA), integration with other systems (like CRM, CDP or call center), and data ownership. We write more about this stuff in another article: 6 Features Every Enterprise Web Analytics Software Should Have. Be sure to check it out!
Important note! Many businesses also face another important issue – analyzing user behavior in the post-login areas of their websites or apps. This is especially tricky, because those places are filled with sensitive data about clients.
Using regular web analytics to track this kind of information might not be the best idea.
Fortunately, there are some alternatives, like Piwik PRO. You can read more about it here:
6) Implement your tool
After you choose the right tool, it’s time to implement it on your website or in your product. This step involves adding tracking code and modifications necessary to track additional things like e-commerce features, contact forms, sending and importing data from CRM or other external tools, and more.
7) Configure the UI of your analytics instance
In this step, you or the person responsible for analytics implementation configures the tool’s interface. This means that they implement all the funnels, goals, and events outlined in the tracking plan. At this stage, it’s also good to exclude internal IPs so you don’t track your employees’ behaviors and actions – this can sometimes skew your data.
8) Create different views with different filters
Next it’s time to configure what you will see when you enter your analytics instance. From our experience, it’s important to create the following views at a minimum:
- A “clean” view, without any filters, to always have access to the original data
- Users of mobile devices
- Desktop users
- Specific countries (if you run an international business)
- A view for analyzing paid campaigns
- All other segments important to your company
- A test view that allows you to check new ideas without losing your data
That way you’ll make sure that you have all the relevant data at your fingertips.
Important tip! It’s highly advisable to do steps 6 to 8 first in a test environment to avoid unexpected errors causing temporary website issues!
9) Grant proper permissions to new users
In this step you assign permissions to your stakeholders so they can view specific reports and dashboards.
The choice of who has access to particular information should be dictated by several factors, including security considerations and what data is needed in the day-to-day work of a given employee. It’s very likely that marketers will need a completely different data set than people in the UX department or customer success team, right?
10) Create dashboards, segments, and custom reports for each data view
Next it’s time to enrich your views with some additional information about user behavior. Here are some blog posts we’re sure will help you wrap your head around things:
- Customize Piwik PRO Dashboard in 7 Easy Steps
- Learn More About Custom Reports in Piwik PRO
- ABC of Segmentation in Piwik PRO
11) Validate reports described in point 10
If everything’s working as it should, that means it’s time to automate reports. This will provide every stakeholder with a dedicated report on a daily basis without needing to log in to the analytics system.
Here’s some advice on automating reports in Piwik PRO:
12) Constantly monitor changes to your website/product structure, goals, and KPIs
Your analytics strategy should never be a static document. Rather, it’s a constantly evolving process that changes with every new insight, business goal, a feature of your website or product, or new idea produced by experiments and tests. It is worth reviewing it every once and a while, refreshing goals, metrics, or reports if necessary. This way you will always be sure that your analytics system delivers only fresh and valuable data that is relevant to your business objectives.
A piece of advice: Consider using more than one web analytics tool. That way you’ll be able to spot data discrepancies, which helps you reach safer decisions and conclusions based on the collected data.
Analytics implementation – final remarks
We hope that the steps presented above will make the process of analytics implementation much easier for you.
However, we know that even with all these tips and tricks presented above, you might still have some unanswered questions. Don’t wait to contact us! Our team will be happy to help you with whatever we can.
Also, remember to download our tracking plan – we promise you’ll find it very useful!