III Know What Your Customer Is Willing to Spend
Chapter 8—Find the upper and lower bid limits
General considerations
As a contractor, you must be seriously concerned with the question of what the customer is willing and able to spend on a project opportunity you think you want to pursue. The primary concern of course is being able to bid an amount that the customer can afford, but there are several side issues as well, for example:
Some of these issues relate to the often-asked question “Is the project real?” If there is a significant chance the project will not “go,” and especially if you perceive your win probability to be uncompetitive, you may decide you do not wish to mount a pursuit. Pursuits are not free, and every misguided pursuit you decide to take on may foreclose going after another, better opportunity.
Sometimes a customer will respond forthrightly to direct questions about these critical issues, and sometimes he or she may try to hide some of the answers, sometimes for valid reasons. In that case some digging may be necessary.
Government customers (at least in Western nations) usually operate on a fairly open basis, unless there is a specific national security or political problem with being too open. Several channels to good information may be available, including legislative sources, freedom of information and open meeting laws, trade publications, reference databases, lobbyists, and the press. Private-sector customers may be more reclusive for various reasons, but still a lot of their information finds its way into the public domain, especially for publicly owned companies.
An often-reliable guide to future behavior is past behavior. Like people, organizations develop habits. Study of past behavior may be worthwhile in estimating what a customer is likely to do in the future. For example, if a customer is known to have a $20 million budget for a project, historically you may know that 10% will be reserved for customer management functions and another 15% for contingencies.
Like poor poker players, customers often signal their intentions without realizing it. You need to stay in frequent contact with customers and prospects and closely observe their “tells,” that is, their language, body language, and actions.
All budgets are based on certain assumptions. Can you find out what they are? Are they realistic? Government budgets may have been the subject of legislative hearings, transcripts of which are available to the public. Even if exact numbers are concealed, knowledge of estimating assumptions made by the customer could lead to a good guess as to his budget.
Customer budgets are (hopefully) based on estimates. Are these estimates competently made? If so, are they later nullified at a high level of customer management for political reasons? Are they fiat estimates made by upper management that are otherwise unsupported? These practices are commonplace in some government agencies, and are not unheard of in private firms. A frequent result is under-funded and poorly performing projects.
The top end of the competitive range
Determination of the top of the competitive bidding range is normally approached in one (or both) of two ways. The first is examination of the customer’s available budget, assuming it can be determined or guessed to a close approximation, combined with consideration of historical patterns as to allocations for internal costs and management reserves.
A key issue in this process is whether the customer’s expectations fit within the amount available. For this reason, it is well to also pursue the second approach, namely to create a “ghost estimate” for the project.
A ghost estimate is an estimate made by or on behalf of the prospective contractor of what the project should cost, with a contingency added to account for obvious risks, with on the order of 75 or 80% confidence of having enough money to complete.
Such an estimate must make assumptions about the product design and programmatic solutions that will be adopted, and these should be based on the most expensive solutions the customer is likely willing to accept.
If the ghost estimate reveals that your customer is expecting too much for the available money, you should consider warning your customer away from this mistake. See Chapter 6 for further discussion of this.
As a practical matter, a ghost estimate cannot be made using the traditional bottom-up estimating technique. In this technique, the pursuit team members, based on personal and group experience, make wholly subjective estimates and sum the results. The information available in the early pursuit phase almost always is too little for a reliable bottom-up ghost estimate. The realistic estimating methods available are the analogy and parametric methods.
An analogy estimate uses judgmentally adjusted comparisons with past projects of a similar nature for which costs are known. You must be careful to take note of and account for all of the major differences in the projects. Examples of commonly encountered differences are year dollars (cost inflation), complexity of the product, inheritance from past projects, and skills of the project team. Production quantity differences must be dealt with carefully due to the complexities of the learning effect.
The best tool for ghost estimates is a credible parametric model, if such is available in your industry. A parametric model is a sophisticated, automated analogy estimator that asks the user for perhaps 10 to 50 bits of information about a product element, and then produces one or more cost estimates for it. Typical costs produced are for development and implementation. Its basis typically is a large pool of historical cost results and project characteristics that have been subjected to sophisticated statistical analyses.
Most contractors license or buy regularly updated parametric models from companies that specialize in this field. A few contractors have the resources to build their own, but this can be expensive, and the model can quickly become obsolete if not regularly updated. Failure to update is a common problem. Another common problem is that contractor models are often too specific to one product type and can’t estimate much else.
The name “parametric” refers to the fact that these models require information about the parameters of, or related to, the object being estimated. For example, to estimate cost of construction of streets in a new subdivision, a parametric model designed for this purpose might ask the user about dimensions, features, materials used, and so forth.
For development of software, a model might ask about number of lines of code, programming language to be used, experience of the software developers, complexity of the software’s functionality, etc. This is the superior approach for ghost estimating because one can usually supply the parametric information needed by the model, whereas good analogies may not be available, and judgmental adjustments of them are not always reliable.
One caution: Parametric models tend to give industry average answers. For a particular project, your team may be better than average, or worse. The best parametric models have a calibration feature that allows you to make adjustments (usually a few percentage points) to the model to bring its answers more in line with the way you do business.
The low end of the competitive range
What about the low end of the competitive range? Theoretically, a contractor with a good record of performance and a lot of cash could offer to do the project for nothing, for whatever benefit the experience might bring, but such an event is rare, and such an offer normally would be suspect to the customer. Almost everyone believes in the saying that “there ain’t no such thing as a free lunch,” and almost everyone becomes wary if one is offered.
A common problem facing customers is contractors who bid at their expected cost or below in the hope of “buying in.” Once the customer is committed to them, they hope the customer will ask for many changes for which they can charge exorbitant prices. Another problem facing customers is inexperienced contractors who don’t understand the difficulty of the project and bid too low out of ignorance. Such contractors almost always get into serious trouble, and may even abandon the project.
Most customers today are well aware of these problems. While they are attracted to real bargains, and will generally choose the lowest responsible bid, they will avoid a bid they think is too low. Their process for determining this cutoff number is typically to select a low cost, low risk approach and to do a ghost estimate of it, with assumptions of small or zero contingency funds and low or zero profit to the contractor. And that is exactly what you should do. After doing this you should ask yourself whether you are bidding this same low cost, low risk approach, and if not, why not? (See chapter 9 for further discussion of this vital issue.)7
You should set the competitive range early in the pursuit cycle so that you can estimate and work toward your competitive bid amount.
You should frequently update the assumptions that went into this initial estimate, and be willing to reset the competitive range as new information becomes available.
Chapter 8 Review Questions
1. On your most recent major proposal, how did you arrive at your bid amount? Did you win? Why?
2. Do you have or have you had a customer whose budget is politically derived as opposed to being estimated realistically? How did you come to know that? What problems has it caused?
3. Has your organization ever made a ghost estimate of either end of the competitive range? If so, how did you do it? Did you use the results to position your bid amount?
4. Does your organization use parametric cost models? Does it use them to help determine the bid amount?
5. Have you ever faced a competitor who literally tried to “buy” the project at a loss in order to assure a win? What were his motivations? What was the ultimate outcome?