Edward Lowe Foundation


Survey of Entrepreneurs'
Information Needs

 

Survey of Information Needs and Preferences of Entrepreneurs:

Prepared for the Edward Lowe Foundation by Q.Com Consulting
and Thompson Information Services

I. BACKGROUND & OBJECTIVES

The Edward Lowe Foundation (referred to in this report as "The Foundation") is a nonprofit foundation that has developed a mission of providing useful information, education, and representation to entrepreneurial firms. Under development is a massive electronic "library" of useful business information and data, as well as Internet links to other sources of information. The Foundation is targeting U.S. manufacturing firms with annual sales in the $0.5 to $25 million range, as well as consultants to this kind of company, such as CPAs, bankers, and small business specialists.

Currently The Foundation offers a fee-based bulletin board, a free Internet World Wide Web home page, and a limited fax back inquiry system. A magazine has been added to this list of services, and other programming is under consideration. The Foundation would like to make these services self-supporting, and is considering various fee structures or corporate sponsorships.

The Foundation has been developing the on-line library based on several assumptions:

With these assumptions in mind, The Foundation contracted our survey team to conduct a telephone survey to determine the market potential of information products and services offered by The Foundation. The major objectives of this survey, as specified by The Foundation, are to learn and understand:

II. METHODOLOGY

To accomplish these objectives, we conducted 434 telephone interviews with the owner/decision maker at entrepreneurial manufacturing companies ranging from $0.5 to $24.9 million in annual sales.

A. SAMPLING
This survey was conducted among a random sample of U.S. manufacturing companies in the $0.5 to $24.9 million annual sales range. The sample was selected from the Dun & Bradstreet (D&B) database using MarketPlace selection software to achieve a balanced random sample. The sample was drawn from SICs 20 to 39, the manufacturing classifications. To avoid the high costs of telephone screening, we accepted D&B sales estimates as accurate. We sampled only from single location companies and company headquarters, so that branch offices were not included.

In most cases, the D&B sample included the name of the key contact, usually an Owner, CEO, or President. This was the person we initially sought to interview. If this name was incorrect, or permanently unavailable (retired, former owner, etc.), we asked to speak with the person with the final word about the long term direction of the company.

As interviewing progressed, it became clear that the owners of larger companies were harder to reach than the owners of smaller companies. To compensate, we set quotas in each of several sales ranges to ensure that the final balance of respondents closely reflects the distribution in the total D&B database.

Initial plans also sought to interview service providers to manufacturing entrepreneurs. After completing 40 interviews, this plan was put on hold for assessment. Most of the providers felt they were incorrectly targeted or proved to be extremely difficult to reach. While they might work on specific issues with entrepreneurs, they are not routinely asked for help on the kinds of issues we were discussing. The manufacturing interviews confirm this, as only 5 of the 434 entrepreneurs interviewed (1%) choose an outside consultant for help when searching for this kind of information. If later analyses of these 40 interviews reveal useful insights, a report supplement will be issued.

To assure we included executives who were difficult to reach, we handled the sample with a call rule. The sample was divided into small replicates, each replicate a random sample of the total sample. Once a sample replicate was opened for interviewing, all working numbers in the replicate received a minimum of four attempts if busy or no answer, with these attempts occurring on varying days of the week and varying hours of the day. When no eligible respondent was currently available, we set up a callback appointment. Only as the call rule was completely worked were additional replicates released for interviewing. In this way, the survey more fairly represents those types of people who are difficult to find near their phone.

B. QUESTIONNAIRE DESIGN
The team drafted several questionnaire proposals for both the manufacturing and service provider versions of the survey. We then discussed and refined the questions until there was agreement that the survey instrument addressed the key issues within the limitations of telephone surveys, budgets, and timing deadlines.

C. INTERVIEWING
The questionnaire was programmed into the Computer Assisted Telephone Interviewing (CATI) system at the WATS interviewing center. The WATS center met our specifications by carefully screening interviewer applicants. They were guided through a training and evaluation program, with progress into more complex studies being carefully monitored, evaluated, and directed.

In October 1995, computers at each interviewing station controlled the telephone dialing, callback rescheduling, questioning sequence, list randomization, logic and consistency checks. Skip patterns and conditional branching were automatic, freeing interviewers to concentrate on respondents rather than questionnaire and sampling mechanics. Ineligible or out of range responses were automatically disallowed by the computers, almost eliminating the chance of human error.

Even the most experienced interviewers were supervised and monitored. Supervisors monitored at least 10% of each interviewer's calls. Individual interviewer performance was evaluated and discussed daily for continual skills development. The research team also monitored interviews on several occasions. We received daily phone or fax reports of interviewing progress along with any problems to be discussed and resolved.

These efforts, we believe, provide interviewing done to the highest standards, and results in which you can have confidence.

D. DATA REDUCTION
By using CATI, responses to closed end questions were automatically sent to our computer, without need for editing or manual keypunching. CATI greatly reduced opportunity for keypunch error, and substantially increased data accuracy.

Responses to open end questions were first carefully checked by a WATS supervisor. We then developed classification codes based on a minimum of 25% of completed questionnaires. These recommended codes were submitted and approved. As discussed in our proposal, a few additional open end codes were added by discretion as full coding commenced.

Open end responses were then keypunched into the entrepreneur respondent database from the closed end CATI interviews. Key accuracy was verified by re-keying.

Data was computer cross-tabulated on our full-featured "MicroTab, Platinum Edition" survey analysis software. Though we bid based on one banner (about 17 vertical columns per question), we have provided two banners at no additional budget. We are also reviewing both banners in this report at no additional cost. Additional banners can be cross-tabulated on request. Additional banners, analytical and reporting costs would be estimated separately.

III. EXECUTIVE SUMMARY OF KEY RESULTS

Though the survey revealed many insights and relationships, and we urge review of the full report, the following presents some of the key highlights.

  1. Most manufacturing executives are computer users. More than half (57%) rate their frequency of computer use a "five" on a one to five scale. However, 20% rate their computer usage only "one," suggesting that about one in five seldom or never use computers.
  2. Though most (85%) of these entrepreneurs are male, computer usage is higher among females, 77% of whom rate their frequency of computer use "five." Computer usage is also higher (66%) among the younger executives.
  3. Though most use computers, only 30% use on-line services, and among them 48% go on-line less than five hours per month. Overall, only 15% of these entrepreneurs are on-line 5 hours or more a month. Clearly, the on-line library concept will have to deal with people currently unfamiliar with and perhaps intimidated by the on-line world.
  4. None of the suggested topics and types of information suggested was widely rated as highly important, but all received a "five" rating from some respondents. This suggests that broad appeal will depend on a wide diversity of information types, easily found and accessed.
  5. The single most important type of information is "Articles and data about your specific industry" (3.44 on a scale from 1 to 5). From the tone of some answers to an open end question, and from listening to some interviews, the team believes that "specific industry" is taken to be much more specific than "manufacturing," but this quantitative survey doesn't provide data to confirm or deny this impression.
  6. Other topics ranking fairly high in importance include "Health care and Workers' Compensation insurance regulations" (3.00); "Latest OSHA and employee safety issues" (2.97); "Financial strategies, techniques, and resources" (2.90); "Pending taxes and legislation affecting entrepreneurial businesses" (2.88); "Personnel management and supervision" (2.87); "Technical, engineering, and scientific information specific to manufacturers" (2.87); and "Computerization and automation for your manufacturing sector" (2.86).
  7. Respondents were not highly creative when asked for other topics. The most requested information was "Information on my industry"(2%); "Information on customers/prospects/prospect databases" (2%); "Internet usage/publishing/selling" (2%); "Scientific/technical information on our industry" (2%); "SEC reports/SEC filings/D&B reports/Credit reports" (1%); and "Capital sources/obtaining capital" (1%).
  8. The two preferred methods of access are telephone request and fax reply (3.18), and computer request and fax reply (3.16). As might be expected, various forms of computer access are significantly more popular among those currently on-line than among those not on-line. Likewise, various access methods using computers are more popular among the "heavy" users of computers than among "light" users. Younger entrepreneurs are also more receptive to computer access than older people.
  9. Among those expressing interest in subscribing to the on-line service, 95% would access the library from within their company, with 57% doing it themselves. Only 2% of those interested in the on-line library would look to an outside consultant like a CPA or lawyer. Of course, while not searching out consultants, it is possible they might be receptive if a consultant approached them about available information.
  10. On a one to five scale, subscription likelihood is rated 2.45. About 8% rate their interest "five" and 20% rate their interest "four" or "five." More than half (52%) rate their interest as only "one" or "two." At this time, it appears the on-line library will be of specialized interest rather than broad scale appeal. That could change as the on-line explosion continues, and if word should spread that the information genuinely gives a company an advantage.
  11. Not surprisingly, subscription interest is higher among those already on-line (2.98) and among "heavy" users of computers (2.65).
  12. At this time in the introduction of a new and unknown concept, there is definite price sensitivity. At $10 per month plus 25¢ per page, 37% rate subscription likelihood "four" or "five," vs. only 20% when no price is mentioned. But at $20 per month plus 25¢ per page, subscription likelihood drops back to 20%. At $40 per month plus 25¢ per page, subscription likelihood is only 7%. This could change dramatically once the value of the on-line information is proven.
  13. Sponsors are a good idea. Seventy-two percent say sponsors would make no difference in their interest, and 18% said sponsors would increase their interest. Among those with interest in subscribing (3, 4, 5) sponsors are even more positive with 23% saying sponsors would increase their interest.
  14. Respondents are divided on preference for computer vs. telephone access, with computer access having a slight edge (47% vs. 40%). Preference is clearly for computer access among those already on line (60%), among those interested in subscribing (56%), and among "heavy" users of computers (53%).
  15. Those entrepreneurs preferring computer access like doing it themselves (27%), and feel confident they can do a better job in less time (21%). Those preferring telephone access feel it would save them time (32%), and is more personal (20%).
  16. Interest in subscribing to the telephone access is very low at all price points, especially at $75 per hour and 25¢ per page (17%). It seems those not interested in computer access are not interested in telephone access either. Again, this could change once the information value is proven.