Geekly Articles each Day
📜 ⬆️ ⬇️

Analytics in CRM: we go on devices

On the Internet, for many years, there is a story about how the Target online store got to the wrath of an angry father in the USA because of the distribution of advertising for pregnant women to an underage schoolgirl. A few days later it turned out that the girl was really pregnant. Target learned about everything earlier by analyzing big data. And due to the use of this data analysis system itself, it grew its revenue in the shortest possible time by more than 60%. The history is replicated because the example is enviable even for pioneers of using data - online retail. In general, if you can boast the same case - scroll Habr further, if not - let's talk under the cut about how to accustom yourself to analytics, learn to go on instruments in any crisis and finally grow up.



I want it so much!


Why grow up? It's simple. Think of a small child or a small child — how many people get tired of this notorious “I want” without any explanations or with unconvincing arguments like “for others”, “but need”, “I want to try”. Many business executives continue to make decisions like this - based on their desire (I want a beautiful price of 22,000), an example of a competitor (let's make a shopping cart, like Pyaticsoft), market fashion (everything, go to the cloud and develop a new direction, close the old one) etc. At the same time, rarely one of the workers is ready to ask whether there is any basis under such a decision. And if anyone asks, in response, you can at best count on "I decided so!"

Consider the roots of this problem and assess whether the root causes are relevant today.
')
  1. The complexity of analytics in the 2000s scared many, especially in big business. These were bulky and fabulously expensive BI-systems and OLAP-cubes, for the maintenance of which required an analyst and a SQL programmer working in conjunction. So take and pull the data in the desired section seemed like a matter beyond the standard skills. Slightly changed the situation 1C, but this was not enough.

  2. Small and medium business grew and had one measure of analytics - profit. The methods of crisis management were also (yes, to conceal, and there are) quite peculiar: dismissals, reductions, cuts in payments, etc. The analytics was simply forgotten, the data were not collected.

    Yes, I made a strategic decision. I decided to ignore the bad news.

  3. Managers and managers were not friendly with the figures, could not correctly read and interpret the data. The analyst was assigned to an accountant who was also incapable of cracking down on an array of numbers or on a programmer who owned numbers but did not understand the goals of the business.

And I will lie if I say that today things are different - the analyst is still in the pen, especially in small and medium businesses. Although there is a positive trend due to the development of Internet analytics systems and business attention to it. But this is not exactly what helps to solve. Anyway, with the advent of new, accessible and easy-to-use tools, business owners began to tinker with the data more. There is also good news - the first of the reasons described is overcome by the market. Today, a business can choose and buy a system with the depth of analytics it needs and find a solution that satisfies the price.


- What is the profitability of business intelligence?
- It is higher than the profitability of ignorance.

In this sense, a CRM system is a quality solution for analytics in small and medium businesses. It combines three essential functions.

  1. Collection and accumulation of data. The logic of the CRM system is designed in such a way that all data is collected in relation to the client, and you can customize input masks and receive information in a unified form. And since CRM is a tool for operational work, the data appears timely and always relevant.

  2. Interpretation of data. The CRM-system interprets the data and “spreads” them according to the necessary modules, where they are compared with other data and form analytical material. So, for example, if a customer bought a consignment of goods, then not only the customer’s profile, but also the warehouse and logistics analytics change. And if it is important to the salesman how many customers are willing to buy the same lot, then the storekeeper is important, for example, the speed of selling the lot, and to the logistician how such customers take out their orders.

  3. Construction of analytical and operational reports. Inside the CRM system (I’ll traditionally make a reservation that I’m telling on the basis of how everything is done in our RegionSoft CRM , options are possible on the market) the opportunity to build ready-made and user reports, to build samples using filters. Reports in CRM are flexible and quite informative, since they pull the necessary data using program logic from the database, which means they can form the sections needed by the manager.


The cycle of using data in a company with the participation of CRM-system

CRM battles


However, some time ago (2005-2010), an operating CRM system like our RegionSoft CRM could not even give a hint about analytics - representatives of analytical CRM immediately attacked it. However, nothing surprising - since there is a war Apple vs Samsung, Windows vs MacOS, then why not be a local CRM war. Although analytical CRM-systems gave up under the pressure of BI-systems. But first things first.

The first to become interested in analytics and analytic systems were industry representatives who already knew what really big data was: banks and telecom operators. Of course, it was not a question of introducing domestic solutions - SAP, Oracle Siebel, began an active expansion in Russia, and made Teradata unique integration projects. At about the same time, Russian vendors who already know what CRM is and write their systems split into two irreconcilable camps. And there were grounds for separation.

Operational CRM accumulated customer information, allowed to create document templates, actively fastened email and telephony clients, and went deeper into covering all management issues. At the same time, domestic analytical CRM, as it could, built OLAP cubes into its systems, drew dashboards and dashboards, and designed product matrices in a big way. All this was done with the use of expensive development tools and DBMS, some subjects managed (and still manage to) sharpen everything under Oracle. Naturally, such development costs could not but affect the cost. But it was not the only problem - the developers of analytical systems got a lot of performance problems, the code required deep refactoring, the weight of distributions grew to indecent, CRM started to slow down on client machines.

While the vendors solved the problems that had fallen on them, the developers of operational CRM realized it and began to write the analytical part, taking into account the mistakes of their competitors. They did not focus on complex OLAP-cubes, did not delve into multidimensional databases, but purposefully created the reports needed by managers here and now: sales funnel, ABC analysis, someone even started playing chess. Something similar happened in RegionSoft CRM : we made it a rule to release not only operational features, but also new reports in new releases. As a result, there were about 100 of them, powerful flexi-filters, reports on customer interactions, plan-actual analysis, KPI, business process analytics and much more appeared. So a new type of corporate software appeared - universal CRM. These include the top Russian CRM, both in the cloud and on-premise.

And what happened to the analytical CRM? I am sure that there are those who cast stones at me, but Russian vendors either switched to versatility (most of them) or significantly reduced their market share. Let's see who transformed into what. In 2007, a CRM project manager from Sputnik Labs in his article proposed the following scheme for the evolution of analytical systems:


More precisely and not draw. The only thing I would add today is continuous interaction with the client based on their behavioral data. Then IoT, context and predictive technologies have not yet touched the business, so the scheme did not include the next step of evolution. However, today almost all steps (except the last) are implemented in the former operating, and now universal CRM-systems.

For example, we in RegionSoft have three large-scale implementations in which special attention was and is being paid to the analytical component.

  1. Implementation for a large network of household goods stores. For several years, RegionSoft CRM has been helping to analyze the assortment, display unpopular names in time, prevent theft and overstock. CRM integrated with cash register and loyalty program.

  2. The implementation of our RegionSoft CRM Media in television companies and teleradioholdings - CRM works with a huge array of data, ratings, airtime and allows you to link and organize the work of advertising departments and broadcast.

  3. In a large supermarket, the RegionSoft Retail system was introduced, which allows you to register sales of piece goods, self-produced goods, and weight goods using factory and internal bar codes (automatically), as well as short codes (manually). The system provides non-stop sales registration, designed for high-loaded self-service stores with a sales area of ​​up to 3,000 square meters. Retail development began a long time ago. Working with a large client with several warehouses and an extensive network of retail stores, we learned how to integrate our RegionSoft CRM with POS-terminals, scanners and other trading equipment. We implemented a large-scale implementation of a CRM system in a huge hypermarket, at the same time we needed a specialized program for POS-terminals. We took and developed. In the process of automating the store, directions for development and refinement became clear. Now with the help of RegionSoft Retail, small grocery stores can be automated, including weight goods, hardware stores, large supermarkets with their own production facilities, pharmacies, warehouses with retail function, points of issue of goods, as well as their retail networks.

However, in response to requests from specific industries, specialized systems continue to develop, which not only lead the market in their segment, but also set the framework for growth for all vendors.


Although, over the years of work on the CRM market, I made sure that each system will find its user. Therefore, a business must adequately address its needs for data analysis and select CRM “by measure”, since this is an important component of a successful implementation (we have already written about this).

What's going on with analytics in business?


As I said above, the first and the worst is almost none. Often we have to deal with companies that do not simply consider the percentage of completion of the plan, not to mention the profitability or return on investment. The second, no less terrible, is a false notion about normative values. For example, one had to hear how a business coach actively convinced that the normal profitability was 300-600%. There is simply no comment. Well, or the sales plan should be consistently exceeded by 20-30%. If your plan is consistently overfulfilled, revise it - your planning is lame in two legs.

In addition to these extremes, there are common basic errors related to business analytics.



My hobby is extrapolation.
“As you can see, in the next few months you will have about four dozen men.” Take care better about wholesale prices for wedding cakes.

What to do?


There are basic data groups that any company needs to collect. They reflect growth points, help manage the assortment, coordinate staff, conduct timely training, etc.

Economic data - a set of information that gives a good idea of ​​the company. These include, for example, sales, income, and expense reports. With their help, you can see which customer segments, product names, affiliates make you money, and which ones only draw money. After analyzing this information, a decision is made to develop effective channels and reduce or heal toxic ones.

Consumer data is a huge layer of information that can be collected in CRM and which is necessary for each of the departments of the commercial service. These are customer segmentation, clustering, sales funnel, grouping by region and order amount, and so on. Such data form the basis of the personification of offers to customers, which has recently been recognized as one of the key growth drivers.

Sales studies are aimed at evaluating various sales channels, implementing a plan, and developing new strategies for communicating with customers.

Product research is aimed at improving the product to meet the needs of consumers. By the way, it is worth mentioning here the measurement and analysis of the quality of service (yes, CRM collects data and for this — call duration, call recording, number of sets of numbers, call conversion into purchase, etc.).

In confirmation of the thesis about the importance of analytics for the listed groups of indicators, I will tell three absolutely real stories.
Simple history 1. The company sold plants and goods for the garden. Deciding to act rationally, she opened one salon at the intersection of several summer cottages, and the second - on the outskirts of the city, where there are no gardens. Revenue in the stores was about the same, everything was fine until the crisis struck and people did not begin to cuddle up in their spending. Besides, it was almost autumn, and the decline was both. When spring came, the revenue in the store at the gardens decreased significantly, and the situation was better on the way out. Closer to the summer on the outskirts of the city it became even better, and the first store confidently went to the closing. Through a long analysis and survey of employees, the managers identified the reason.
Those who attended gardens, as a rule, did not live in them, so they bought many plants that can grow without a host such as seedlings, saplings of bushes, etc. The reduction in demand coincided with the saturation of a narrow market - people abandoned goods bought out of curiosity. And on the road the audience was wider - and those who lived in the summer at the big country cottage bought more expensive seedlings and, moreover, there were different flower vases, lanterns and so on. The trouble was that the owners were only concerned with the check, and the assortment was delivered thoughtlessly, almost randomly. If the accounting was carried out according to nomenclature positions, then the owners could rebuild the assortment policy and sell, for example, gardens, fertilizers, tools, etc. But this did not happen, the company no longer exists.

A simple story 2. Approximately the same thing happened with the city store of inexpensive clothes. But the guys quickly discovered problems with the assortment and got rid of stocks, avoiding the glut. But the fact was that the main visitors to the “family” stores were women with children, and almost all men's clothing was a heavy burden. But adding to the assortment of toys and children's jewelry has increased the average bill. By the way, unlike the heroes of simple history-1, the accounting system was in this network, and the marketing director was not idly interested in what was taken into account in CRM.

The reverse story. A small mobile operator was chasing sales. Point A made a plan and about 1000 SIM sales per month, point B did not make a plan, there were about 300 sales. As a result, the sellers of B were left without premiums, and they also received from the management a scolding for lagging behind the plan. This lasted for almost a year, until the new sales team decided for its own purposes to look at the charges of subscribers in the context of points, including these two own offices. So, from the A volume, 85-90% reprimanded 100 rubles on the account and threw out the SIM, and from B about 70% used the number for more than three months with an average of 95 rubles per month, and 30% for more than six months. When they began to delve into it, they found a huge fraud and even offenses at point A. As a result, KPIs were reworked and sales plans were revised.
Collect data is not enough, it is important to be able to use them for the benefit of the business. There are several basic ways to do this.

Implement CRM if it is not already there. A CRM system can become a source of data not only for top managers, but also for employees themselves. Your task is to teach them to work with data, use them for making operational decisions. From the moment employees are immersed in analytics, miracles begin to occur:



“I have good news for you.” (In fact, CRM opens up tremendous opportunities to expand knowledge about your business and to find growth areas).

In principle, modern business risks nothing - almost in all CRM systems, and in RegionSoft CRM including, a procedure is provided for differentiating access rights to various data.

Work with segments. If you learn how to make segments and use the created groupings, you will solve a huge layer of business problems. Based on the selected groups, you will create point offers, mailings, divide groups between managers and get a customer base with greater loyalty.

Visualize the data and show it at the meetings and reporting meetings. It’s one thing if, for the month, you monotonously mumble in front of employees about the implementation of the plan and further development, and the other is showing graphs and clearly explaining what is and what you should strive for. Visualization greatly facilitates understanding and memorization - feel free to take it into service.

There is another interesting way to use the collected data - the creation of analytical newsletters and leaflets and their distribution to a wide audience for the purpose of public relations. Not every company will dare to do this, but if you have something to share with customers, consumers or the media, feel free to do it in a readable and understandable form. Such actions usually pay off by increasing fame and trust in the company.

What can you do with the data in CRM?


CRM - a powerful source of data in the company. And you can use a ready-made set of reports, or you can upload information and work with it yourself - the main thing is that you already have it in a unified form. And the main rule is to collect data continuously and iteratively, compare the results from period to period, look at the development of indicators over time.


In principle, based on the data obtained in CRM, you can build a fairly deep business analytics. However, there is a set of indicators and actions with them, which are the minimum and should always be in operation.


CRM-system is an excellent source of data, you can process them yourself and they are processed in the CRM itself. There must necessarily be measurable metrics that, in addition, you will be ready to measure - that is, use for some purposes. If for some reason the indicator is not needed - do not waste time collecting data to calculate it. For the purposes of business and decision-making, it is not necessary to cover the entire array of numbers; it is important to be able to single out the main thing, to approach the question not formally. At the same time, a manager should not rape his subordinates with numbers and reports - his task is to learn how to work with data and show his employees how cool it is. And finally, the main thing is not to collect data haphazardly and not to use information as the entourage of a business. Analytics should work - at all levels.

Throughout December, we offer discounts on RegionSoft CRM and all our proprietary software. From December 1 to 15 - 15% and steep terms of installment and rental. We do not have -70% and -90%, because we keep the economically reasonable price for the license, and do not take it from the ceiling.

And we continue the survey, the results of which will be sure to tell. We ask you to answer the questions in a simple form - there are only 10 plus 3 of them for those who are interested in learning more about us. Please reach the end of the survey, just skip any unnecessary items.

Take a poll here

Source: https://habr.com/ru/post/312406/

More articles:


All Articles