Why you need comprehensive analytics
End-to-end analytics involves collecting, processing and analyzing customer acquisition performance (ROMI) and customer retention data. The period is counted from the moment the ad is displayed until repeated purchases during the customer's lifetime (LTV).
These metrics help us understand where we're effectively spending our marketing dollars and where we're wasting them. The task of comprehensive analytics is to combine application data from the website and sales data in the CRM system.
For example, we have a German language school with advertising campaigns in Google Ads and Facebook. Each system has several different advertising campaigns with different messages and addressed to different target groups. We need to determine which ad campaigns and audiences bring in not just apps, but real money.
Comprehensive user identification from the moment of first contact to sale comes in handy here. We brand campaigns and pass advertising data to the next stages of the funnel.
Who needs comprehensive analyses?
End-to-end analytics are mainly used in digital, where tools are available to track users' online activities: UTM tags, cookies, user account identification.
Suitable for online promotion of almost any business: education, finance, e-commerce, web services and applications.
How comprehensive analysis works
There are three stages of user contact with advertising:
Advertising - we see statistics in the advertising offices of Google and Facebook.
Website - with visit statistics (Google Analytics, Adobe Analytics).
CRM system (Bitrix24, Microsoft Dynamics 365).
Data from the tools is collected in one database. To calculate ROMI, we need to determine how much we spent on the advertising campaign and how much revenue we generated.
You can set up an autofunnel - a step-by-step scenario through which a person goes from the first visit to the website to making a purchase on their own, with transactions on the website without the involvement of sales managers. In this case, the costs of the advertising campaign are recorded in the advertising account. You can get revenue data in Google Analytics by setting up value goals or an e-commerce block. All you need to do is download the advertising account report and analytics, compare the data and calculate ROMI, e.g. in Excel.
The only downside to this method is Excel. Many tables require manual updating and compilation.
However, in 90% of cases, the problem can be easily solved by importing cost data into Google Analytics. Using the "data import" option, you can import revenue data, e.g. via OWOX Pipeline or other connectors.
Unfortunately, there are many such companies. We will usually still have contact with the manager, returns and other sales channels. We need a more flexible system that will be able to capture all stages and variants of the development of events.
For analytics to be truly comprehensive, we need to collect all data in one Database basket - a set of structured and interconnected tables containing data on all customer contact points. And then output the data to a report/dashboard.
To prevent everything from getting mixed up in the database, a marking is needed - a common data identifier. At the stage of user transition to the website, our main identifier will be the UTM tag. All online activity must be tagged, even if it is text messages, PR articles or QR codes. After entering the website, the second identifier appears - ClientID. This is a user ID that Google Analytics provides to you when you first visit the site.
Thanks to identifiers, we can combine data because the analytics will know the user's UTM tag and ClientID, and the CRM system will know the ClientID and the phone number or other user identifier.
How to process data?
To start with, you need to choose a basket that will be able to collect all the necessary data, compare and process it. There are several options:
- Excel or Google Sheets;
- Google Analytics;
- Comprehensive analytical systems in a separate tool or based on CRM;
- Custom solutions based on Google Big Query or MySQL.
Let's analyze the advantages and disadvantages of each solution:
Excel or Google Sheets
This solution is suitable if you need to aggregate a small amount of data or have technical difficulties with integration. For example, for one of our clients, we collect data in Google Sheets from advertising platforms and Google Analytics using App Scripts. For its part, it sends conversion data indicating the UTM tag through which the customer came. This, in turn, is received along with the lead data. Then we map the data and send it to the Google Data Studio panel.
The downside here is the limited volume of data processed and tracking repeat sales.
Google Analytics
Google Analytics has two features that make it a good data store.
The first of them is the "import of expenses" that we talked about at the beginning. The second one is the Measurement protocol, which you can use to send any data to analytics. We will need it to transfer sales data from the CRM system. For implementation, you need to configure several POST request scripts that will send data about who ordered from us, when and for how much.
If you're just getting started with end-to-end tracking, Google Analytics is a great and, best of all, free way to track your entire journey. Among the minuses - connecting the developer to submit requests from CRM and transfer ClientID to a separate field of the CRM system along with other data about the client.
Comprehensive analytical systems
There are many systems available on the market that allow you to collect all data in one place. Some allow you to configure attribution models (a mechanism for assessing the value of various marketing efforts – ed.). Just add a few scripts to the website and log in to your ad accounts.
When it comes to quickly and inexpensively solving the problem. But there are some disadvantages:
You do not have access to the data content to prepare custom reports.
All reports in the system interface. Sharing information in a large company or with contractors is difficult.
There will be difficulties in downloading sources not anticipated by the system (SMS mailings, PR articles, CPA networks).
Custom solution
Integration may be difficult and take two to six months, but the result is worth it. It is worth taking this path when we have already used one of the above methods, but it is objectively insufficient. If you have a lot of repeat sales from different sources and different user paths, it is unlikely that you can collect everything in a third-party system without additional attribution linking the individual sales sources.
Choosing the best way to set up comprehensive analytics will depend on your marketing efforts, sales funnel, repeat sales rate, and ease of use.
However, if your website has fewer than 50,000 users per month and advertising costs less than PLN 10,000, there is probably no point in spending money on complicated integrations.
Who does the comprehensive analysis?
For either option, you will need knowledge of web analytics: how data is collected and transmitted to various systems. You can solve this topic yourself, but in this case you risk filling many cones before everything works. If the amount of work on end-to-end analytics is large enough, you can safely hire an in-house web analyst or at least use the services of a consultant. But it is quite convenient to outsource this part of the work to advertising agencies or specialists in web and comprehensive analytics.
Where to start setting up comprehensive analytics
Comprehensive analysis is a general concept. Everyone gives it their own meaning and level of detail. Above all, analytics should be clean and useful, and data should be normalized.
Enrich your analyzes step by step. Start by summarizing the data in Excel, determine the error between orders and sales, and connect third-party services if necessary. Even these minimal steps can make working with reports and understanding the performance of your ads much easier.
Comments