Post by sumiseo558899 on Nov 7, 2024 6:34:35 GMT
Under what conditions will the end-to-end analytics system not work to its full potential — will it not help reduce budgets, increase sales, will it not help regulate business processes and control employees? In case of insufficient data!
In what cases they may be lacking, what is the reason for this and what to do about it, I, Roman Lomovskoy, head of the analytics department of the digital agency Original Works, will tell you.
End-to-end analytics has been discussed content writing service
for a long time, and many companies have already encountered it first-hand. However, first I would like to once again discuss what end-to-end analytics is and how it can be useful for business.
End-to-end analytics is a business tool for achieving commercial goals, as well as solving specific problems. More specifically, end-to-end analytics will help:
Calculate the profitability of different types of advertising: contextual, targeted advertising, SEO, etc. Based on the data obtained, you can redistribute the budget in favor of more profitable sources and not waste resources blindly.
Evaluate and compare the effectiveness of advertising campaigns and contractors.
Get advanced analytics for your website, online store or landing page to build conversion funnels. Which will also allow you to find bottlenecks and errors in the website interface.
Find bottlenecks in business processes by integrating with reporting from all business units (warehouse, HR, logistics).
Make smart business decisions based on data, not gut instinct.
I will also add that end-to-end analytics can provide answers to questions: how much money should be invested in processes to get X rubles of profit, and what prevents you from getting Y instead of X. And quite often the answer lies not in marketing or sales, but in management, logistics, warehouse or other links.
For example, a company produces machines for the woodworking industry. The production capacity allows for 10 machines per month, so no matter how hard the marketing department tries, it will not be able to sell 15 machines per month (if there is no warehouse stock). Of course, this is a very rough example, but it perfectly shows that it is not always worth looking only at marketing. With the help of analytics, you can dig deeper and see the picture of the business as a whole.
As you can see, end-to-end analytics is a very important, I would even say, necessary tool for every modern business. But for it to work as efficiently as possible, you need to remember the principle of “Garbage in, garbage out” and take a responsible approach to collecting data for analytics, both as a contractor and as a customer.
What does a complete analytics system consist of?
One of the most popular reasons why end-to-end analytics does not work to its full potential is working with superficial data. Such data cannot be used to build analytics that works as a business tool for achieving goals.
A full-fledged end-to-end analytics system should include, at a minimum, data on advertising costs, website traffic, applications, sales, and income received. Based on this, conversions, cost of lead and sales, and ROMI are calculated. These are the indicators that we always calculate, and this is the minimum for which it is worth building end-to-end reporting in principle.
news
However, you shouldn't limit yourself to these indicators. Each business has its own KPIs that need to be tracked. For example, for the auto business, these are visits to the showroom and GROI (margin / costs / average shelf life * 365), and for trading companies, the average sales period. We always take all these needs into account in the agency when developing reports.
From my experience, I can identify two main reasons why a system may lack data:
technical difficulties;
lack of coordination between the contractor and the client or within the client's company
Technical factor
First, let's look at cases where, for technical reasons, it is not possible to collect data in sufficient quantity:
Lack of a CRM system in principle. This can be encountered much less frequently now than before, so we will not consider this reason in detail. But here it is worth considering: do you need end-to-end analytics at all at this stage?
Data collection is incorrectly configured. This scenario is becoming more common. In this case, we analyze what data is missing, why it is not collected, and offer options for correction.
Home-made or little-known CRM systems. Let's consider this case in more detail.
There are many ready-made solutions for AmoCRM or Bitrix24 for integration with databases. The structures of these CRMs have already been studied, and in general it is clear what data needs to be downloaded, how they are related to each other, etc.
Self-written systems can cause a number of difficulties, namely:
Undescribed data structure. It often happens that a CRM system has an API*, but does not have clearly described methods. Actually, this cannot be called a problem, but unloading data from such a system takes several times more time than unloading from a conditional AmoCRM. We always take this into account at the calculation stage.
The CRM system does not have an API*, but the client can provide access to its database. In this case, we simply access the CRM system database directly. This option is not very reliable, as it creates an additional load on the CRM. The fact is that the server on which the database is located can be designed for a certain number of requests, and an attempt to send a request to the DB over this limit can provoke an error.
The CRM system does not have an API* and there is no access to the DB. This option is the most difficult, since in this case it is only possible to manually unload data from the CRM system to Excel or Google Docs, from where the BI system will take the data. This approach is not very good, since it assumes active human participation. Manual filling of any tables inevitably leads to errors. And the more of these tables, the larger they are, the more errors. In Google tables, the formatting can easily get lost or some data can be lost when copying. And of course, do not forget about the existing restrictions on the volume of data stored in such tables: for Excel files - 1,048,576 rows, for Google tables - 5 million cells per book.
In what cases they may be lacking, what is the reason for this and what to do about it, I, Roman Lomovskoy, head of the analytics department of the digital agency Original Works, will tell you.
End-to-end analytics has been discussed content writing service
for a long time, and many companies have already encountered it first-hand. However, first I would like to once again discuss what end-to-end analytics is and how it can be useful for business.
End-to-end analytics is a business tool for achieving commercial goals, as well as solving specific problems. More specifically, end-to-end analytics will help:
Calculate the profitability of different types of advertising: contextual, targeted advertising, SEO, etc. Based on the data obtained, you can redistribute the budget in favor of more profitable sources and not waste resources blindly.
Evaluate and compare the effectiveness of advertising campaigns and contractors.
Get advanced analytics for your website, online store or landing page to build conversion funnels. Which will also allow you to find bottlenecks and errors in the website interface.
Find bottlenecks in business processes by integrating with reporting from all business units (warehouse, HR, logistics).
Make smart business decisions based on data, not gut instinct.
I will also add that end-to-end analytics can provide answers to questions: how much money should be invested in processes to get X rubles of profit, and what prevents you from getting Y instead of X. And quite often the answer lies not in marketing or sales, but in management, logistics, warehouse or other links.
For example, a company produces machines for the woodworking industry. The production capacity allows for 10 machines per month, so no matter how hard the marketing department tries, it will not be able to sell 15 machines per month (if there is no warehouse stock). Of course, this is a very rough example, but it perfectly shows that it is not always worth looking only at marketing. With the help of analytics, you can dig deeper and see the picture of the business as a whole.
As you can see, end-to-end analytics is a very important, I would even say, necessary tool for every modern business. But for it to work as efficiently as possible, you need to remember the principle of “Garbage in, garbage out” and take a responsible approach to collecting data for analytics, both as a contractor and as a customer.
What does a complete analytics system consist of?
One of the most popular reasons why end-to-end analytics does not work to its full potential is working with superficial data. Such data cannot be used to build analytics that works as a business tool for achieving goals.
A full-fledged end-to-end analytics system should include, at a minimum, data on advertising costs, website traffic, applications, sales, and income received. Based on this, conversions, cost of lead and sales, and ROMI are calculated. These are the indicators that we always calculate, and this is the minimum for which it is worth building end-to-end reporting in principle.
news
However, you shouldn't limit yourself to these indicators. Each business has its own KPIs that need to be tracked. For example, for the auto business, these are visits to the showroom and GROI (margin / costs / average shelf life * 365), and for trading companies, the average sales period. We always take all these needs into account in the agency when developing reports.
From my experience, I can identify two main reasons why a system may lack data:
technical difficulties;
lack of coordination between the contractor and the client or within the client's company
Technical factor
First, let's look at cases where, for technical reasons, it is not possible to collect data in sufficient quantity:
Lack of a CRM system in principle. This can be encountered much less frequently now than before, so we will not consider this reason in detail. But here it is worth considering: do you need end-to-end analytics at all at this stage?
Data collection is incorrectly configured. This scenario is becoming more common. In this case, we analyze what data is missing, why it is not collected, and offer options for correction.
Home-made or little-known CRM systems. Let's consider this case in more detail.
There are many ready-made solutions for AmoCRM or Bitrix24 for integration with databases. The structures of these CRMs have already been studied, and in general it is clear what data needs to be downloaded, how they are related to each other, etc.
Self-written systems can cause a number of difficulties, namely:
Undescribed data structure. It often happens that a CRM system has an API*, but does not have clearly described methods. Actually, this cannot be called a problem, but unloading data from such a system takes several times more time than unloading from a conditional AmoCRM. We always take this into account at the calculation stage.
The CRM system does not have an API*, but the client can provide access to its database. In this case, we simply access the CRM system database directly. This option is not very reliable, as it creates an additional load on the CRM. The fact is that the server on which the database is located can be designed for a certain number of requests, and an attempt to send a request to the DB over this limit can provoke an error.
The CRM system does not have an API* and there is no access to the DB. This option is the most difficult, since in this case it is only possible to manually unload data from the CRM system to Excel or Google Docs, from where the BI system will take the data. This approach is not very good, since it assumes active human participation. Manual filling of any tables inevitably leads to errors. And the more of these tables, the larger they are, the more errors. In Google tables, the formatting can easily get lost or some data can be lost when copying. And of course, do not forget about the existing restrictions on the volume of data stored in such tables: for Excel files - 1,048,576 rows, for Google tables - 5 million cells per book.