Why do SED work poorly?

INTRODUCTION

In 1991, I was a student at one of the enterprises in the field of energy. The task that I had to solve was to develop a program to control the instructions issued by managers. The program was called “KIP - control of execution of instructions”. As it seemed to me then, I solved the problem. It was hard to imagine that I would have to return to this topic in 12 years and turn it into the main subject of my activity - the development and implementation of electronic document management systems. 20 years later, it turned out that this topic is just as relevant, and attempts to solve it do not always lead to the expected results. Why is this happening?
The reasons are probably many. Limitations on terms and budget, qualification and motivation of personnel, project methodology, political conjuncture - these and other factors contain prerequisites for design failures. I want to offer one more explanation of the problems arising during the implementation and operation of electronic document management systems.


')

TERMS AND DEFINITIONS

Denote the terms that will be found below.
Under the ORD (organizational and administrative workflow), we will understand the processing of documents that serve the company's management activities - incoming and outgoing letters, orders, instructions, instructions, memos.
The term “document flow of operating activities” will be understood as the processing of documents specific to the field of activity in which the company operates. For example, for insurance companies, this could be the processing of insurance policies, for banks - the processing of loan applications, and so on.
The concept of EDMS - electronic document management system - covers all types of document management companies. But in most cases, it is used to denote systems that automate ORDs, contractual activities, and sometimes accounting and personnel records. To a lesser extent, this term is used for the automation of workflow companies operating activities.
This division is very conditional, but it will help us to place accents when considering the problems associated with the processes of implementation and operation of systems that automate the workflow of companies.

PROBLEMS OF PROJECTS

I began to participate in projects of introduction of electronic document management systems since 2003. Gradually, my attention was attracted by the fact that all projects are executed according to the same scenario and in most cases lead to the same problems. This happened regardless of the customer’s company, budget size, or the composition of the project team members.
A typical project for introducing an EDS began by defining the types of documents to be automated and describing the business processes for processing each type of document. The main part of the terms of reference was document flow diagrams, and the core of the system was one or several business processes automating the described routes. Everything looked beautiful on the development booths, but as soon as the introduction of the system began, problems arose.

Problem 1. Typical routes are not followed.

The first significant problem faced by the project team was that the document routes — designed, agreed, and implemented — were not respected. It turned out that the company has a sufficient number of exceptions, without which it is impossible to implement the system. The implementation of these exceptions led not only to delaying the deadlines, increasing labor intensity and budget, but, what is most unpleasant, to a violation of the system architecture. If, before the start of the implementation, the processing of documents looked streamlined, then by the end of the implementation it could appear that there was no clear business process. Documents move from any stage to any stage, and employees, whose role was not provided for in the routes, may be involved in the processing. Making such changes to the solution architecture led to performance problems and complex errors.

Problem 2. The increase in workflow

The next trouble was that after the automation of the OSA, the volume of documents processed by employees increases. At the same time, an increase in the company's total document flow was not observed. This circumstance is often explained by the specifics of the company's work before and after the introduction of the SED. For example, a typical situation was when a manager gave documents to performers, through the heads of his deputies, and without registering with the office the fact of the transfer. Another example is when the head of a subdivision transferred employee service notes to him for processing to employees only after a second appeal to him with the same question. Service notes that were not reminded of him were safely sent to the trash. In either case, the introduction of business processes for processing documents excluded the withdrawal of documents from routes and forced all employees to process all documents intended for them. Which led to an increase in the processing of documents by employees.

Problem 3. Task Queue Overflow

The basis of the architecture of most workflow systems are business processes. The concept of employees in this case is based on the principles of working with the queue of tasks. Events that occur during the movement of documents are tasks that are placed in the inbox, which users must consistently handle. Employees are transformed into queuing systems, the purpose of which is clearing the queue. This principle of operation is not suitable for all user roles. Gradually, employees begin to ignore some of the events. In the future, this leads to an overflow of task queues, which is another reason for the decline in system performance.

Problem 4. Increased transaction costs

Once, while analyzing the results of workflow automation projects for several years of implementations, we made an unexpected generalization. Traditional EDMS, automating the organizational and administrative document flow of companies required more time, resources, budget; Customers placed higher demands on the user interface, while the level of customer satisfaction with the results was low. At the same time, projects that automate the business processes of the customer’s operations were faster, simpler, and, despite simpler user interfaces, customer satisfaction with the result was significantly higher. We concluded that one of the reasons for this state of affairs is the targeted orientation of the systems.
Automation of the company's operations is primarily aimed at reducing the complexity of work performed by employees, to accelerate the process and simplify the way to achieve results. In other words, to reduce the transaction costs of the operational process and increase labor efficiency. Automating the processes of ORD, as a rule, solves other tasks - informative, reporting, preserving the history of document processing, control over the actions of employees, etc. As a result of solving these problems, it turns out that the introduction of the SED does not simplify the work, but, on the contrary, leads to an increase in labor costs for performing the same operations. It became clear why the great honor of managers after the introduction of the SED transferred their functions in the system to the secretaries - the SED made their work less efficient.

Do you need SED?

The experience of implementations, as a result of which we received systems that complicate the work of employees who have performance and fault tolerance problems, suggested that SED is evil. Business does not get returns, it does not become more efficient. Does it make sense to introduce them at all? It remained unclear why in the conditions of a large number of controversial implementations, however, the demand for workflow automation increases every year. Obviously, there is a real need for solving the tasks of working with documents, but it is necessary to find a way to properly satisfy this need. The solution to this issue came gradually. The key to understanding was the experience of several successful implementations of electronic document management.

WORKING ELECTRONIC DOCUMENT

Coordination of project documentation

One of the first examples of successful use of electronic document management systems was the implementation of the project documentation approval process.
The project team was faced with the task of automating the workflow of the operating activities of one of the companies on the EMC Documentum platform. The tight deadlines of the project did not allow replacing the traditional, not very friendly interface of the platform with their own. Understanding that when introducing the system, we will not be able to avoid the negative reaction of users to the interface, we needed a tool to reduce customer expectations. The solution to this situation was the decision to accustom key users to the interface of the platform even before the system was launched - during the course of the project. To this end, a project archive was created on the basis of the Documentum platform, in which all the work with documents was organized, including the process of coordinating project documentation. The goal was achieved. As a result, we received:

• Structured repository of project documentation. Each document was in its place, which reduced the search time.
• The project archive always contained the latest relevant documents.
• All versions of documents were saved, according to which at any time it was possible to restore the history of changes in documents.
• The coordination of documents was carried out only in the project archive, the documents were not printed out and were not sent by e-mail.
• There were results of the coordination of documents by all members of the project team, with their copies and revisions.
• There was up-to-date information on the execution of the deadlines for the coordination of documents.

At the same time, for the organization of this work only standard functionality of the EMC Documentum platform was used, without additional customization. It was also curious that business processes were not used to organize the coordination process. The effect, which was achieved with minimal effort, called for a rethinking of the experience of automating the procedures for working with documents and the processes of their harmonization.

Accounting Outsourcing

Another example of the successful use of electronic document circulation was the solution to the problem of reducing the costs of accounting. The task was to transfer accounting to outsourcing. The problem was the remoteness of accounting from the location of the primary accounting documents. It was necessary to find a way to quickly exchange documents between the company’s office and the accounting department in an environment where the office and accounting department are located in different cities. The solution was the organization of a structured document repository based on the Alfresco platform with the implementation of business rules for processing documents. The result was a process:

• Upon receipt of the primary document, the employee scanned it and placed in accordance with the nomenclature of cases.
• When the document was saved, a business rule was launched, which generated an email notification to the accountant with a link to the document in the vault.
• The results of document processing and accounting in the accounting system were reflected in the document card. If necessary, the accountant posted a response document, also in accordance with the nomenclature, which was processed in a similar way.

The results exceeded all expectations. The company received a structured repository of all personnel documentation in electronic form. Document exchange took minimal time. Subsequent search and access to documents accelerated. Again, to achieve this result, mainly standard platform functionality was used. As in the first case, for organizing the process of exchanging and processing documents, the business process tool was not involved.
The experience gained required answers to the questions:
Why in some cases, large efforts expended on the automation of workflow, lead to modest results and user dissatisfaction, while in others, the minimum costs give a tangible effect, including economic?
How should the workflow system be built so that its implementation leads to unconditional results for the company?

ASSESSMENT OF ECONOMIC EFFICIENCY

The answer to the first question can be found in the works of Paul Strassman - one of the leading experts in the field of evaluating the effectiveness of investments in information technology.



Paul Strassman - honorary professor of computer science at the George Mason School of Information Technology and Engineering; Fellow of the United States Information Security Center; author of 8 books, more than 250 articles on information technology and the effectiveness of investment in information technology.
Since 1956 - CIO in companies: General Foods Corporation, Kraft Corporation, Xerox. In 1991-1993 - Director of Information Security at the US Department of Defense. In 2002 - CIO at NASA.
P.Strassman was an information systems consultant for companies: AT & T, Citicorp, Digital Equipment, General Electric, General Motors, IBM, ING, SAIC, Shell Oil, Sun Microsystems, Texas Instruments.

Paul Strassman investigated the open financial statements of more than 500 US companies, trying to find a relationship between the volume of costs in information technology and the economic performance of companies. The results of P.Strassman's research showed that there is no direct correlation between the IT budgets of companies and target indicators of business economic efficiency. The figure shows a typical picture of the statistical relationship between the level of IT costs (the ratio of the IT budget to the company's revenue) and the level of return on business (Return on Assets). A similar picture was observed in relation to other economic indicators.


At the same time, Paul Strassman found an indicator that had a positive correlation with IT costs. It turned out to be a group of costs for managing, administering, and supporting sales — Sales, General & Administrative (SG & A). SG & A includes salaries of management, administrative staff and salespeople, marketing and sales costs, as well as third-party management services. It turned out that an increase in management and sales costs is accompanied (directly proportional!) By an increase in information technology costs.


From the research of Paul Strassman it follows that information technologies are primarily in demand in the areas of management and marketing. And this means that IT should be aimed at improving the efficiency of management and sales. Accordingly, the electronic document management system should solve the problem of reducing costs in the company's activities and improving the efficiency of the management system. Obviously, in this case, we should observe a positive effect from the introduction of the SED.

FORMS OF MANAGEMENT

Studies by Paul Strassman explain the tasks that should be addressed by the introduction of electronic document management systems. But do not answer the second question - what should be the SED to solve the tasks. To answer this question, we turn to the work of one of the leading specialists in the field of management, Henry Mintzberg.



Henry Mintzberg is a professor of management at McGill University in Montreal. Professor at Carnegie Mellon University, University d'Aix-Mardeill, Ecole des hautes etudes commerciales, Montreal, London Business School and INSEAD. Winner of 15 honorary degrees from universities around the world.
G. Mintsberg - author of 15 books; more than 150 articles, two of which received the MacKinsey Harvard Business Review.
According to The Wall Street Journal, G. Mintsberg is among the 10 most influential thinkers in the field of management.

In the 1970s, Henry Mintzberg set a goal to streamline the findings of researchers in the field of management, combining them into a general picture of the structuring of organizations. The work took more than 3 years. The research results resulted in 512 pages of text in small print, containing conclusions about what makes an organization effective. The findings were supported by quotations from the works of other researchers.
G. Mintzberg discovered that despite all the variety of types of organizations, differing in size, scope, geography, capital, etc., management in all of them is based on only 5 basic forms, four of which are basic. These forms of management are:

• Direct control
• Mutual agreement
• Standardization of operations
• Standardization of skills

Management form "Direct control"

With direct control, the responsibility for the work of all employees lies with one person. The head coordinates the actions of his subordinates, issuing instructions and controlling their execution. In small organizations, all employees report directly to the manager. In larger companies, a hierarchical structure is created, with the basic principle being preserved - management is built on top teams.

Management form "Mutual agreement"

With mutual coordination, control over the labor process is carried out by the employees themselves. This form of control is used in situations with a high degree of uncertainty. Responsibility for making decisions can take on each of the employees, coordinating their actions with the rest.

Management form "Standardization of Operations"

In the standardization of operations, the activities of employees are divided into the simplest operations that do not require a complex thinking process. The process of activity is strictly regulated. Employees are subject to instructions that new employees easily learn.

Management Skills Standardization Form

For areas of activity in which the result is unpredictable and requires highly skilled staff, mechanisms for standardization of skills and knowledge are used. Achievement of the result is ensured by professional actions of employees specializing in certain areas. The task of the company - search, training, certification of specialists.

Organization Configurations

In organizations, as a rule, several forms of management are combined. At the same time, G. Mintsberg showed that each organization is dominated by one of the forms on which the company's operating activities are built. In accordance with the determining form of management, an organization can be assigned to one of the types:

• Simple configuration
• Adhocracy
• Mechanistic bureaucracy
• Professional bureaucracy

Examples of companies corresponding to the types of configurations:
Simple configuration (direct control) - small companies with access to local resources and distribution channels - small shops, farms, handicrafts, small travel companies; private companies enjoying administrative support; business based on personal connections; management companies.
Adhocracy (mutual agreement) - start-ups, design bureaus, innovative divisions within large companies, medical research centers, engineering companies, individual and pilot production, sale of original works.
Mechanistic bureaucracy (standardization of operations) - large companies producing consumer goods, mass simple services; large retail chains; fast food
Professional bureaucracy (standardization of skills) - auditing companies, consulting firms, specialized medical centers, antique sales.

TYPES OF DOCUMENT

The structuring of organizations done by Henry Mintzberg makes it possible to single out the characteristics of each type of organization. Contract activity, sales structure, work with personnel, reward system, working hours, communication principle, etc. - each configuration has its own characteristics. Workflow for each type of organization also has its own specifics.
Simple configuration (direct control) - the basis of the workflow - the instructions of managers. With the growth of hierarchy - the increase in orders issued by the documents. A large number of memos that are a form of transfer of responsibility. The coordination of documents is distributed on levels of hierarchies.
Adhocracy (mutual agreement) - the basis of the document flow - documents of the subject activity. A large number of unstructured communications - forums, chat rooms. Almost complete lack of instructions. Unstructured (collective) coordination of documents.
Mechanistic bureaucracy (standardization of operations) - the basis of document flow - documents accompanying the company's operating activities. Forms of documents are clearly typed. The movement of documents is streamlined. Coordination is structured by type of documents and by their sections.
Professional bureaucracy (standardization of skills) - the basis of workflow - documents of professional activity. Documents are structured by roles / professional areas of employees. To create new documents, employees use material gained from their previous experience.



We note an important conclusion:
Each business configuration has its own type of workflow, supporting the management structure and operations of the company.

ARMS OF SED

Understanding the specifics of the workflow for each type of configuration, one can explain the problems encountered in projects. In most cases, the development of the system was based on the mechanism of business processes. At the implementation stage, a system with typed routes was issued. At the same time, the introduction of systems often took place in structures close to a simple configuration. The management form of a simple structure is contrary to the principles that are embedded in the business process architecture. As a result, this leads to a violation of the stages of the process, a change of executors and non-compliance with the procedure for processing documents.
The situation when the ECM platform is taken as the basis for the SED in such projects leads to even more disastrous results. In this case, developers first have to adapt ECM to the system architecture based on business processes, and only then destroy the newly created architecture to the requirements of unstructured processes of simple configuration.
We approached the answer to the question posed, what should be the system of electronic document management of companies:
In order to avoid systemic problems when introducing electronic document management systems, it is necessary to choose an ERMS architecture corresponding to the type of document management and the form of company management.
The figure shows typical system architectures that most closely match the types of company configurations.



For a simple structure, the main element of the ERMS architecture is the subsystem for issuing and controlling orders. The most adapted architecture for such a system is a platform based on the principles of electronic message transmission. The manager must be able to issue an order anytime and anywhere. The reporting system should provide complete information on the status of executed and outstanding orders.
For Adhocracy, the main element of the SED architecture should be a content management module (Enterprise Content Management, ECM) with support for collaboration tools. The system should allow to aggregate a large amount of information, and provide functional search tools. The system should be implemented tools for teamwork with the support of forums, chat rooms, news feeds.
For the Mechanistic bureaucracy, the main element of the ERMS architecture is the Business Process Management (BPM) module, which allows to automate the business processes of the company's document processing. The workflow business process must strictly fit into the existing technological process of the company, with pre-defined types of documents and routes of their movement.
For Professional bureaucracy, the main element of the SED architecture, as well as for adhocracy, is the content management module (ECM) with the support of individual work tools. The system should provide a means of structuring information, giving the user the opportunity to categorize documents and refer to their previous experience. The system should include the creation of document templates and contain sample templates for all user roles. The system should support non-formalized document reconciliation tools based on the document life cycle, as well as collaboration tools that allow multiple users to work as a team on a single set of documents.

CONCLUSION

The proposed solution seems simple and applicable for ideal cases. In reality, organizations combine various forms of workflow. It is not by chance that the fifth type of configuration proposed by G. Mintsberg is the divisional structure, which can include all four previous types. To meet the needs of such organizations, an EDS is required, which will combine several types of architectures, which will allow implementing different types of workflow in one organization.

LITERATURE

1. G. Mintzberg. Structure in the fist: the creation of an effective organization / Trans. from English by ed. Yu. N. Kapturevskogo. - SPb .: Peter, 2004. - 512 pp., Ill. - (Series "Business bestseller"). 18 5-469-00256-
2. “Is it possible to measure the economic effect of IT?” PC Week / RE # 4 (789) February 21, 2012. Author: Olga Pavlova
3. “In search of efficiency - part 1”, Intelligent Enterprise № 7 (201), 2009, Author: Ananyin V.I.
4. "Formation of the corporate information system architecture by natural selection . " Intelligent Enterprise № 17 (149), September 26, 2006. Author: Ananyin V.I.