A correctly posed question makes it more likely to get an objectively justified answer, and the procedure for solving a problem always matters. Analytical work is creativity using modern knowledge, experience and tools.
Mathematical statistics and programming make it possible to make sense of large amounts of data and justify a decision. Here, "manual" processing is not real, but there will be no result without the participation of a specialist.
Not every subject area has large amounts of data, but it always matters to supplement automatic analysis algorithms with a human factor. Some analytical reviews are only the work of a specialist, but mathematical and logical analysis will never be superfluous.
Science, practice and finance
Analytical review is a filtering of accumulated knowledge forrationale for specific ideas, actions or decisions. Even from a student's bench, a person collects and studies information for a term paper, a diploma, or a report on industrial practice. Participation in scientific conferences also requires preliminary research work, and a diploma or dissertation is the application of acquired knowledge and skills in the context of uniqueness, novelty and practical value.
Business practice, study of sales markets for products or organization of production requires a different analytics. Here the price of a mistake is the lack of profit, deprivation of wages or the collapse of the enterprise. The airbag is a company's chance to do analytical work, find mistakes and correct the situation.
Analytical review is a safe and stable business development, the basis for making informed and objective decisions. The widespread introduction of computer technology and software development have put the processes of processing and analyzing information on stream. As a rule, analytical reviews of markets, the state of the company, tax burdens (benefits), consumer demand, etc. fall on the desk of the head of the department (company) and enable him to quickly make the right decision.
Preparing filters and building a systematic picture of the situation for the problem being solved is a long process. The very first report on a new topic will be generated for a long time, and each next one will refine and supplement the once created algorithm in less time.
Finance isa separate category for analytical work. Information flows in a company are its circulatory system. Cash flows are the result and the instrument of control. A financial analysis review is a combination of several topics that complement each other and do not exist on their own.
Money in the company's account is:
- accounting - primary documents and information on accounting accounts;
- economics - cost, revenue analysis and forecasting;
- production - stable supply of materials and shipment of products;
- warehouses - movement of material assets, return, warranty, service, etc.
Business can always be evaluated by quantity and quality - these are its parameters. An analytical review is an assessment of the current state of affairs and options for appropriate dynamics, movement towards stability, profit, security.
Perfect result
Programming and mathematical apparatus based on probability theory and mathematical statistics, various methods of grouping data, determining correlations, relationships, dependencies, etc. - a practical set of tools that “makes” “raw” data out of gray streams real meaning, a system of events, connections or a picture of processes.
This is understandable in the field of climate research, trading on the stock exchange, selling consumer goods. This is incomprehensible in programming or in the production of unique piece devices.
Programming is already several decades old, but until now the concept of "rollback" exists only in office applications. For example, in MS Word or MS Excelthere are "undo" &"redo" functions: back (rollback) and forward (redo). Meanwhile, the content of an analytical review for creating a high-load web resource or a local program (for example, an information pre-processing server) will be an order of magnitude more promising if it reflects the history of data changes, the history of visits, or the dynamics of the visitor's interest in the functionality offered to him.
A program (a web resource is also a program) that takes into account time and history will be more promising, more objective and more functional than one that is focused only on the current moment: a visitor came, got the desired result and left.
Just the statistics of visits to a web resource is the basis for improving the functionality. The dynamics of visits for all visitors and the analysis of the logic of their work is a completely different aspect of the problem of improving the quality of a web resource and increasing its traffic.
The ideal result in the context of programming is to make the web resource adapt itself to the needs of visitors. In other words, what the programmer has implemented should develop independently as needed and in the direction that is in demand. The best analytical review will not provide for this.
Analytics performed by a programmer or manager is an option “before” putting a web resource into operation or before its next upgrade. The daily work of a web resource with visitors is an option "here and now". In the latter case, it develops "independently" and always reflectsdesires of visitors.
Search, filter and organize
When planning analytical work on natural phenomena, in cellular communication systems, in the field of managing the transportation of goods by land, sea or air, it is necessary to record the flows of "raw" information in databases, which causes an increase in their volume.
Fixing only the result of preprocessing, you can miss or lose something. In cases where work is being done with data sources that are characterized by persistence, it is advisable to focus on the development of search and filtering mechanisms.
In the first case, the center of gravity is transferred to the processing of the flow of information, in the second case, the work is divided into two components:
- clarification of the list of information sources;
- selecting the right information at the time it is needed.
Optimizing search and sampling algorithms is a related task. A characteristic feature of the second option is that duplicates can be added to the database. To avoid this, it is advisable to assign a unique code to each new data, which is determined by its "body", for example, the MD5() function in PHP gives a unique 32-byte code for any data sequence.
The string of information put into the database may be different. For example, a product line: name, cost and characteristics. These three positions are always unique, but if we also take the time of receipt of the goods or the change in its quantity -you can add unnecessary duplicates to the database.
Organization of information is always a problem. The program is not a person. What is obvious to a human is not "obvious" to a program. Usually, the problem of organizing information is solved at the stage of the technical task, and after the web resource is put into operation, the problem arises of how to change the system, for example:
- product categories;
- types of goods;
- models or types;
- supplier names, etc.
An original and practical solution is to build a search filter and an algorithm for organizing data based on information. For example: a sample is received - an algorithm is executed - there is a table of results. The work of searching, filtering and organizing goes in a cycle.
As soon as the first table of results has appeared, the next algorithm is launched on it, which looks for analogies, performs generalization, and forms categories. The first result table is transformed into several systematized tables, and ultimately the primary processing cycle is supplemented by a cycle of continuous work to summarize the information already received.
Smart pharmacy, medicines and he alth
Analytics, which lay on the manager's desk, is the result of the time spent by a company employee, for the sake of an instant, but informed decision. An analytical review of the company's economy is, for example, a weekly adjustment of the company's strategy (development concept) in the following areas:
- accounting for we alth (accounting);
- assessment of the dynamics of expenditures and incomes (finance orplanning);
- management and control of production (deliveries, shipments, warehouses).
In reality, goals can be very different. But the classic analytical review is the amount of work performed by an employee or an entire department of the company in order to make a specific decision (range of decisions). The first is tangible time, the second is a quick decision by the head, board, board of directors or other governing bodies of the company.
The pharmacy sells drugs, but the pharmacist does not write prescriptions, and is not authorized to treat. The doctor treats the patient. Among the many drugs, there are those that are available only by prescription, but many more can be used independently. Knowing the purpose of the drug, evaluating the symptoms that have arisen, a person can make a decision on his own and improve his well-being without waiting for an appointment with a doctor to make an accurate diagnosis.
For example, there is a task: "Smart Pharmacy" is a web resource that advises visitors on what can and cannot be used, if there are certain signs of a disease. The symptoms that the visitor reports are dynamics, the properties of the preparations are statics. Here, an analytical overview is a dynamic summary of the available data (and accumulated in the process for all visitors) to make the right decision for a specific visitor.
The motto "consult your doctor" is declared to every visitor. In this context, this is an essential rule. However, analytics is formed dynamically according to what wasapplied, by whom, on what basis, what was the result. The right of the visitor is to accept or not to accept information, but, of course, he will benefit from information about the properties of drugs and the prescription of drugs, formed dynamically, concisely, accurately and to the point.
You can read the inserts for medicines when they are bought, or in the process of communicating with a pharmacist. This is the time, as well as the requirement for understanding what is written. A web resource that will instantly form a subject sample based on symptoms is a dynamic that, if it does not help a person, will definitely guide him in subsequent actions.
Scope and structure of the policy brief
The ideal result is three paragraphs: the first - what is at stake, the second - what filters (methods, programs, functions) are used, the third - what is proposed. The maximum volume is one page.
Analytics is not a work of art, not an article on philosophy or mathematical statistics. In the first paragraph, the goals and the range of its constituent tasks are precisely indicated. Description of research methods, tools and programs used at the request of the manager (customer). The final part is the proposed solution.
Even if the review is a hundred pages, you can always state everything on one first page in the format of three, presented in an accessible and understandable way, paragraphs. Everything else should be called an explanatory note, and everything that the employee (or the work team deems necessary) should be written there.
Perfect question and real answer
The ability to set the task correctly, to ask the question accurately -half the success when writing an analytical review. This is a description of the desired solution. The task (question) is formulated by the manager (customer). The result should be clear not only to the performer, but also to the one who ordered the work.
Correct understanding of the task by the specialist allows him to correctly present the solution. The question and the answer must match each other - this is a prerequisite that ensures the success of the work done.
