Successful EDM Implementation:

The Business Case

by Rainer Hoff, Ph.D., P.Eng.

May 1998

© 1998, Gateway Consulting Group, Inc.

Introduction

This is part 3 in a series about successful EDM implementation. Part 1 and the methodology diagram are given in the Jan/Feb 1998 edition of Document Management. An information systems project ("the Project’), either in the realm of electronic document management ("EDM") or product data management ("PDM"), is going to require substantial expenditures of manpower and financial resources. The Business Case phase demonstrates to management the financial viability of the project, and is often the key activity in obtaining funds for the project.

 

Do We Need A Business Case?

Yes.

In most cases, EDM/PDM projects are implemented to improve the efficiency of an existing business process, possibly in conjunction with a business process re-engineering effort. Both the costs and benefits are considered when making the financial analysis.

In other cases, the Project is part of a larger program, (e.g. ERP implementation, new plant construction, new product introduction) and the Project costs are simply part of the larger program budget. In this case, only the costs are determined, and then validated against program budget constraints.

Either way, the costs have to be determined, and usually the benefits, as well.

 

Why Do We Need It Now?

Referring to the methodology diagram, one can argue that the costs cannot be accurately determined until after the Vendor Selection phase is complete. That is a true but not a practical answer. Experience in information technology ("I.T.") projects has taught us that funding a major I.T. initiative is a process, rather than an event. As such, it is usually necessary to set cost expectations early, and then refine the cost information as the Project progresses.

An organization needs time to overcome the "sticker shock" that is usually associated with these Projects. If you wait until after Vendor Selection to determine and publish cost information, your Project will likely be on hold while the cost/benefit information is scrutinized and validated. Meanwhile, you’ve lost momentum and your team will become demoralized or be disbanded. That’s why we suggest doing the Business Case right after the scope is determined in the Initial Study.

Newcomers to this technology usually have a difficult time in determining the costs; vendors can help in setting product costs, while consultants can aid in determining overall project costs.

Of course, the cost estimates will evolve as the Project matures through the Requirements phase. But now you have a cost model which can be used to assess the impact of decisions made during the Requirements phase.

 

Benefit Determination

Benefit Identification

The benefits of an EDM or PDM implementation are numerous. There are many individual benefits, which I call "unit benefits". The unit benefits can be determined by brainstorming, interviewing, research in industry publications, engaging consultants, etc.

 

Benefit Classification

A business case presentation is an exercise undertaken to answer one question in the mind of the approver(s): "Convince me that this investment is financially sound." So, the validity of a given benefit is primarily a perception of the listener and only secondarily an inherent property of the benefit. Unit benefits therefore have to be classified into what are perceived as "strong" and "weak" arguments, in your organization. Obviously, the "strong" benefits are emphasized while the "weak" benefits are de-emphasized or eliminated.

Figure 1 shows benefits classified in terms of specificity and analysis technique. A "broad brush" analysis is usually couched in very general wording. For instance, "Engineers spend x% of their time looking for information. With an EDM system, we’ll save some portion, say 50%, of that time." This is much less convincing than an analysis of a specific situation. Specific situations may include shop floor drawing retrieval, ISO quality manual management and distribution, supplier drawing review and approval.

The analysis technique also impacts how the benefit argument is perceived. Here’s an actual example of such an argument based on conjecture. A Sales VP opines, "If we managed our documents better when we’re preparing bids for our customers, we’d win about 10% more of the work that we compete for." This argument, while it might be correct, is difficult to substantiate, so it’s only compelling to the extent that the VP is respected in the organization. It will only work when he is personally promoting it.

A statistical analysis is stronger than guessing. This was used in an analyzing how the frequency of scrap and rework events related to the timing of drawing availability, in a fast track design/construction environment. In this case, detailed information was available regarding the timing and frequency of events associated with drawing transmittal. Also, detailed information was available which related scrap events to drawing errors which could have been avoided—a necessity since not all scrap can be avoided by implementing EDM. Statistical arguments are difficult to convey because of the complexity of the mathematics involved—if you use them, you must pay close attention to how the data is presented. A prime benefit of EDM is cycle time reduction, often, unfortunately, demonstrated using a statistical analysis.

A deterministic analysis is the most compelling. This involves a detailed analysis of a specific business process. Consider: "A person in the plant requires about 30 minutes to walk to the drawing vault, identify the drawing needed, and get a print. This happens 7,300 times a year. This time would be reduced to, say, 3 minutes per transaction, with EDM on the shop floor." This type of analysis can be applied to any business process which is observable and can be measured in real time.

 

Cost Determination

There are two significant aspects of EDM costs: what the costs are and when they are accrued.

Unit Costs

The unit costs for an EDM system are identified in Figures 2, 3 and 4. Over the last decade, hardware costs have come down consistently. Software costs have come down, but are very volatile. In this regard, I would remind readers that software vendors have to sell software to make a profit. If you encounter situations that are "too good to be true", or software that is "free", be very skeptical. The services costs have actually risen slightly due to the higher demand and competition for limited resources.

Costs can be accrued at three different "times":

• Initial costs: when the system is first installed
• Incremental costs: whenever the system is extended in any way. See "Rollout" below.
• Annual costs: upgrades and annual maintenance costs.

We can see that EDM costs are, in fact, a time series—hardware, software and service cost events usually occur repeatedly.

The Analysis

Unit costs and unit benefits are necessary, but not sufficient information to determine the viability of a project. We have to know when these cost and benefit events occur, and this is determined in the Rollout Model.

In larger EDM systems, e.g. large, multi-discipline engineering organization, process plant, manufacturing company, the entire system is not installed on day 1. In fact, an "initial system" is installed, and then it is deployed or "rolled out" elsewhere. The dimensions of the rollout include adding:

• users
• concurrent users
• new applications
• new functions
• departments
• buildings
• remote sites

The Financial Model

Each rollout increment adds costs (determined from the unit costs) immediately, and adds benefits thereafter. We typically create a mathematical model of the organization, and parameterize the number of users, concurrent users, applications, departments, etc. This organizational model is then the basis for aggregating the costs and benefits in the financial model.

The financial model is constructed to determine the Net Present Value, "NPV", Internal Rate of Return, "IRR", and the Payback Period. IRR’s of 25% to 35% and Payback Periods of 1.5 to 2.5 years would be typical of EDM projects in the $1 million range.

 

Sensitivity Analysis

When the business case is presented, it’s likely that your assumptions will be questioned. In order to prepare for this eventuality, it’s possible to adjust the assumptions/parameters in the financial model to determine the "sensitivity" of the NPV and IRR to changes in the assumptions. Obviously, if the NPV or IRR is very sensitive to a particular assumption, a strong reason must be made for why the specific value was chosen. For example, one project for a major oil company showed an IRR of 0% if the system life was assumed to be 2 years, 44% if the system life was 5 years and 60% if the system life was 7 years.

 

Company-Specific Aspects

Companies differ in how certain aspects of a financial analysis are calculated. You should validate your calculation technique with corporate standards before presenting your results. Areas of greatest variation include the treatment of:

• taxes
• depreciation
• leasing
• cost of capital
• maintenance costs
• manpower costs
• capitalization of services
• infrastructure costs (e.g. networks)

This has been a brief treatment of a very vital subject. Feel free to e-mail me at rhoff@gatewaygrp.com if you’d like more information or elaboration.