It doesn't have to be one of life's great mysteries.

Although generating an estimate for datacom work may seem similar to estimating electrical work, the two differ in rather significant ways. First of all, with datacom estimating, you must pay more attention to the material supply channels than with conventional electrical estimating. Why? Simply put, the datacom purchasing process is more complicated than the typical electrical purchasing process. In fact, two or more links in the distribution chain are almost always missing.

You may or may not even have access to a full-service local supply house. Then there are tasks such as programming equipment, installing software, testing, and training, which can eat up a considerable amount of time. Datacom work requires that you not only test each individual conductor, but you must document the results and furnish copies of this documentation to the customer. Needless to say, testing datacom cabling takes a lot more time than testing power wiring. The secret to generating an accurate datacom estimate is to be well prepared and knowledgable of these key facts.

Delivery of goods. Considering the distribution channels for datacom materials are varied and sometimes unreliable, developing an accurate estimate is tough. How can you come up with realistic figures if you're not even sure you can get the materials required to complete the job?

To solve part of your problem, use estimating forms that have columns for listing vendors of each and every product. You also have to make allowances for unreliable supply channels. For example, let's say you get a price on network interface cards (NICs) of $85 each from a supplier you're not too familiar with. If you're not sure of the reliability of the supplier, you should increase the material quote as a safety measure. In this case, you may want to use a figure of $100, in case you have to go to another source at the time of installation. This is where your good judgement comes into play.

To prepare a proper estimate, you must also determine the expected labor hours and related costs before you can come up with solid figures. This means assigning labor units based on past jobs.

Bidding. There are many types of bids that you will be asked to furnish in the datacom market. Most of them will be lump-sum bids, or unit-priced bids. But there is another type of bid that you will encounter, the RFP. RFP stands for Request for Proposal. In the network world, RFP is almost equivalent to bid documents, except it isn't as detailed.

When these people want a bid from you, they will send you an RFP. This document will give you the general details of the project, and ask you to furnish a complete design, schedule, and price. Completing the RFP process is similar to estimating and bidding on a design/build project. Take care when you prepare RFPs, that you will not be doing the design work for the customer (for free), only to have another party do the installation according to your design.

Charging overhead. When it comes to assigning overhead to a project, we all seem to have our own opinions. But most electrical contractors would agree that when it comes to datacom projects, you should raise the figure somewhat.

When you contract to do datacom installations, you're agreeing to go through uncharted, or at least partially uncharted, waters. This involves greater risk. It's only sensible to cover these risks by charging a little more for overhead and/or profit.

Figuring labor units. Any good cost estimate must cover the following 10 essential elements of labor. The conventional method of estimating does this by assigning a single labor unit to each item below:

• Reading the plans;

• Ordering materials;

• Receiving and storing materials;

• Moving the materials;

• Getting the proper tools;

• Making measurements and laying out the work;

• Installing the materials;

• Testing;

• Cleaning up; and

• Lost time and breaks.

Although the conventional method of estimating is excellent where circumstances are consistent and predictable, datacom work introduces more room for error. For instance, we can specify a certain labor rate for installing a controller, but there may be other functions required to complete the installation, such as programming, software installation, testing, and modifications to programs.

Many datacom installations require you to teach the owners or their representatives to use the system. This is much worse for datacom work than for power wiring projects. After all, you don't have to spend time teaching building owners how to use conventional electrical items, such as light switches and receptacles.

On the other hand, you will certainly have to spend time teaching them how to use an FDDI network. You will probably have to supply operating instructions for the equipment you install, teach a number of different people how to use the equipment, and answer numerous questions over the phone, long after the project is complete.

Because of this situation, a single labor unit is not sufficient to assign labor for the item. What might be applicable for one installation may be double what is appropriate in the next -- even though you might have the exact same type of equipment. Your challenge with datacom estimating is finding a method of assigning labor in either situation. There are several solutions to this problem. Most estimators find it easiest to use two lines on the estimate form for each item: one for a base labor unit (which includes the material cost); the other for "connect" labor.

As shown in the 10 essential elements of labor above, there are basic functions that we bundle together to create a standard labor unit. Our solution to the high-tech labor problem associated with datacom projects is to isolate and remove the variable element from the other nine elements of labor, and charge for it separately. By removing the testing function from our labor unit, and charging for it separately, we better define the estimating process. All of our volatility is moved into one separate category, which demands special attention from an estimator when preparing the estimate. The basic labor unit covers the rest of the labor, which is much more consistent and predictable.

Keep in mind, you typically base standard labor units on the following conditions:

• An average worker;

• A maximum working height of 12 ft;

• A normal availability of workers;

• A reasonably accessible work area;

• Proper tools and equipment;

• A building not exceeding three stories; and

• Normal weather conditions.

Unusually poor working conditions typically require an increase of 20% to 30%. Some very difficult installations may require even more. Especially good working conditions, or exceptionally good workers may allow discounts to the labor units of 10% to 20%, and possibly more in some circumstances.

As you can see, estimating datacom work can be an adventure. If you're not totally prepared for the task, you could easily end up losing your shirt.

Next month, we'll continue our discussion by stepping through an example estimate.