Talk to your potential commercial customers. Find out what they need. Discuss their participation in Phase III (commercialization).

Establish long- and short-term goals for your business. The DOE found that it takes an average of 7 1/2 years for products that they fund under the Inventions & Innovation Program [formerly the Energy-Related Inventions Program (ERIP)] to reach the marketplace. It is a truism that it usually takes about 3 years for a company to fail and 7+ years for one to succeed.

Identify the field(s) or focus of your company and conceptualize a product line. It makes sense to "hook your wagon" to a long-term trend.

Decide on a form of business. An S corporation has many advantages but can cause "exit" problems.

Decide on a commercialization strategy--manufacture and sell? license? joint venture? Some program managers at NSF frown on commercialization using exclusive licensing. All types of licensing are acceptable to DOE.

Identify your business team--management, marketing, selling, bookkeeping, clerical, technical. With the SBIR program, the primary employer of the principal investigator (PI) must be the small business. This is not the case with the STTR program.

Set up a cost accounting system that complies with the Federal Acquisition Regulations (FAR). (If you get a Phase II grant, you will be audited and, if possible, they will ask for some money back--just like the IRS.)

Obtain other sources of financing for unallowable costs, e.g., interest, overruns, foreign patents, etc., and for cash flow--savings? relatives? second mortgage?

Set out and assign actions to achieve goals. Obtaining SBIR/STTR/I&IP money is one of many of those actions.

Identify exit strategy for you and your financial backers.

Assess your performance and update plan at least annually.

Download or order all of them; they're free. Obtain prepublication solicitation drafts and from agency WWW (World Wide Web) sites.

Read them cover to cover--look for subtle differences. Some agencies are practical; some are more research-oriented. Some are risk takers; some are not. Some have reasonably-sized programs and Phase II budgets; some do not.

Look for topic areas that fit into your business plan and help you achieve your corporate goals. Do not propose on topics just because you can win them. Propose only on those topics that will generate a new product line or improve an existing product.

Dismiss very specific topics unless you wrote the topic description. You can suggest topics through the Broad Agency Announcement process.

See if you can write one proposal, reformat it, and submit it to multiple agencies to beat the 1 in 8 success ratio.

Write a query letter (containing a nonenabling disclosure) to determine the agency's interest in your idea (and to connect with a technical, in-house champion) before the solicitation issues.

Obtain agency organization charts. Some are on the Web. Get to know your customer!

Target the agencies and divisions or programs within those agencies that you will be proposing to.

Call both SBIR/STTR people and technical (program) people with questions, preferably before the solicitation issues. At least ask them if your idea is of interest to the agency--you cannot always tell from the solicitation.

Determine their interests (SBIR/STTR people are usually not technical, but program people are) and their "hot buttons."

Meet face to face at least once per year with SBIR/STTR people, program people, and their bosses; touch base with them at the national SBIR conferences.

Keep them up to date regarding technical progress and progress toward commercialization.

Read that section of the solicitation at least three times. How many times?

Find out who the reviewers are, or at least understand the type of reviewer--e.g., NIH sometimes uses their normal review panels. Network, network, network. Do not expect all reviewers to be from your field or to know anything about it.

If you stay within the acceptable limits, total cost is unimportant (ask for what it will really cost). You must, however, be able to document all direct costs (with catalog cuts, etc.) and indirect costs (with historical data and/or a budget) during negotiations.

Evaluate the impact of alternative indirect cost bases, e.g., (1) direct wages, (2) direct wages plus fringe benefits and payroll taxes, or (3) total direct costs.

Understand Fast Track options. Line up matching funds. Establish measurable goals for Phase I.

There are three rules for proposal writing: get the job, get the job, and get the job. Tell the truth, but put it in the best light possible.

Propose a significantly innovative, commercializable solution to a real national problem. If "it's been done before," it is not appropriate for this program.

Conduct a complete literature review (and patent search) and a preliminary market assessment, including a sales forecast, before you start. Establish in the proposal that you know what is going on in the field.

Come to agreement within your team on written summarized versions of the project goals, hypothesis, objectives and tasks before you start writing the text. This will give you the bare bones on which to hang the details of your proposal.

Follow the outline (format) exactly, but anticipate reviewer questions.

Expect to spend $3,000 to $5,000 or so of bid and proposal costs [the allowability of B&P (bid and proposal) costs is limited by the FAR] to prepare each new proposal.

Most agencies appear to "bet on" the PI. Some agencies want the PI to have a Ph.D.

Watch out for "trick" format items, e.g., character and word limits, 25 page limit, lines per inch, and margin limits.

Obtain the best project team in the country by retaining consultants who are recognized leaders (and who have published or consulted recently) in the field. Use university consultants to supplement in-house expertise, if possible.

Prepare a draft sufficiently ahead of the due date for your consultant(s) to be able to review the proposal before it is submitted. (Some will be willing to put in time "betting on the come".) If you need one, retain a technical editor.

Emphasize the work plan. It should be the longest, most detailed section. Describe what you will do, how you will do it, and what results you expect.

If you have any preliminary data, present it (assuming it does not negate the need for the Phase I).

Propose either an elegant solution to the problem posed in the solicitation (my favorite) or a procedure for finding one. Use the scientific method--it is refreshing to reviewers.

Focus on testing the hypothesis that the idea is technically and financially feasible. The farther along you get, the easier it will be to obtain Phase III contingent commitment letters from strategic partners.

Do not mention patents that you have on the technology if doing so would confuse the reviewers concerning whether the technical feasibility of the idea is still at issue. Your proposal must "walk the thin line" between exhibiting too much risk and not enough.

The most common mistake is to promise to do too much. Reviewers recognize this mistake (if you are lucky) and such proposals rarely are successful. Focus Phase I on addressing the critical issues of technical feasibility.

Plan to build a working (breadboard) model or simple laboratory prototype. Test it in a well-designed experiment.

Prepare the proposal with the same care that you would prepare a technical paper for publication. Do not plan to be finished the day before the proposal is due. It takes longer than you think.

Always ask to be debriefed on proposals (unsuccessful and successful ones). Resubmit unsuccessful proposals if hope is offered.

Always ask for "profit" some of which will actually be used to cover unallowable costs, which typically comprise about five percent of total costs.

Always ask for advance payments or progress payments, if allowed.

Always specifically ask for facilities cost of capital, because it is unallowable if you do not.

Identify those parts of your proposal that are confidential in accordance with agency rules.

File your proposal as a Disclosure Document with the U.S. Patent and Trademark Office. (Actually, it is best to file a patent application with very broad claims, and to actually reduce your core technology to practice before the project starts, and to develop improvements during the project. In this way you maintain control of the technology even though the Government obtains a license to practice the improvements.)

Reformat your proposal and file it with another agency, if feasible. Do not use a "shoe horn" to fit your idea into a topic!

Remember that contract officials believe that you know exactly what you are going to do. Humor them.

At the beginning, remind the negotiator about your request for advance payments or progress payments.

The average fringe benefit and payroll tax rate for scientific organizations is about 38 percent of wages. Indirect costs will run 50-150 percent or more of total direct costs.

"Happiness is positive cash flow." So is survival. If recovery of indirect costs is critical, understand how less than full recovery will impact your cash flow. Expect 1-2 month delays in receiving all except wired advance payments.

If you haven't already, involve a potential customer in setting performance requirements for your technology. Commercialization takes 3-5 years or more; start now.

Be familiar with the quirks of the contract regulations of each agency that you work with as they are all different. Remember, you will be audited.

Use a computerized cost accounting system, e.g., Peachtree, Profitwise. Document all expenditures. Track them on a biweekly or monthly basis.

Start with an annotated outline of your final report. It will be your research "product."

Take lots of photos. Show what the product looks like--how it works. Videotape "action" experiments.

Do not expect to be given a time extension or any more money. Some agencies allow nine months but such extensions can eliminate the possibility of Phase II (i.e., it is a trap).

Be open to other applications of the technology. How can your core technology be leveraged into other product forms?

If a cost overrun appears appropriate (as is often the case when engineers and scientists are involved), figure out where you will get the money (including the indirect costs) before you spend it. (Expect a $5,000+ overrun on a Phase I, which is an unallowable cost.)

Document consultant responsibilities and financial arrangements. Remember that subcontractors own inventions conceived or first actually reduced to practice during SBIR projects. Try to obtain at least a nonexclusive, transferable, royalty-free license from any subcontractor who will be inventing during the project (but you cannot force a subcontractor to grant it). Intellectual property ownership issues must be settled beforehand in STTR projects.

Start thinking about how you can productively utilize (and pay) your project team during the 6 months to a year between Phase I and II.

If Phase I results are marginal, do not apply for a Phase II grant just because you have a 1 in 3 chance of obtaining one. Ask yourself, "Do I want to spend 7-10 years of my life pushing this technology over the hump to commercialization?"

The SBIR/STTR programs are an excellent source of supplemental R&D funding for small, technology-based businesses.

A long-term perspective, persistence and lots of other money are required.

Your goal should be to become ineligible for the program by becoming a large business.