Informal Notes on the Ziff data The material includes information from the Computer Select disks for 1989/1990. It is copyrighted by Ziff-Davis Publishing Co. The data has been changed to standard SGML format, as shown below. ZF107-923-224 07 923 224.&M; PC Week Nov 27 1989 v6 n47 p126(1) * Full Text COPYRIGHT Ziff-Davis Publishing Co. 1989.&M; Retail outlets continue to dominate as source for PC peripheral purchases.&O; Rideout, Katherine.&M; According to surveys conducted by PC Week in 1988 and 1989, American corporations still prefer to purchase microcomputer peripheral equipment through over-the-counter retail outlets than any other distribution channel.&P; According to the surveys, almost six of 10 corporate and institutional sites use these outlets as a key source for new microcomputer peripherals.&P; Five purchase source categories were identified for respondents, including over-the-counter retail outlets, corporate sales forces fielded by major retailers, value-added resellers or systems houses, direct-from-manufacturer purchasing and mail order.&P; Thirty-one percent of respondents cited retail outlets as their primary source; 28 percent cited this channel as a second source.&P; When these findings are compared with data from the 1988 survey, small decreases in primary and secondary reliance on retail outlets, from 33 percent to 31 percent, respectively, are evident.&M; Topic: Peripherals Survey Computer peripherals Computer Retailing Computer stores Companies User Behavior Marketing Market Share.&O; Feature: illustration graph.&O; Caption: Relative stability seen in selection of purchase channels for PC products. (graph)&M; By Katherine Rideout&M; Little appears to have changed over the past year regarding corporate America's preferred purchase channels for new microcomputer peripheral products.&M; As was true last year, more sites purchase such peripherals through over-the-counter retail outlets than any other distribution channel.&O; Currently, nearly six out of 10 corporate and institutional sites use retail outlets as either a primary or a secondary source for new micro peripherals, a figure down only slightly from nearly two out of three sites a year ago.&M; These observations are based on comparative data from two surveys -- one conducted in May 1988 and the other in September 1989 -- of members of PC Week's Corporate Advisory Panel.&P; That panel comprises more than 500 individuals having responsibility for the purchase and support of microcomputer products at corporate and institutional installations across the United States.&M; Both of these surveys asked panelists about primary and secondary purchasing channels through which their sites buy various categories of micro-related products.&P; In addition to the primary source, panelists were asked to identify a single secondary distribution source.&M; For the purposes of these surveys, five categories of purchase source were identified for respondents: over-the-counter retail outlets; corporate sales forces fielded by major retailers; value-added resellers or systems houses; direct-from-manufacturer purchasing; and mail order.&M; When asked specifically about the purchase of micro peripherals, 31 percent of respondents to the September survey cited retail outlets as their primary source, with 28 percent citing this channel as a secondary source.&O; Comparison of these findings with data from the earlier survey indicates small decreases in both primary and secondary reliance on retail outlets, from 33 percent and 31 percent, respectively, in May 1988.&M; Corporate sales forces fielded by retail organizations proved equally popular among panel members as sources for micro peripherals.&P; Overall, 39 percent of panel sites currently employ retailers' corporate sales forces, up slightly from 36 percent in May 1988.&P; While primary usage of this channel increased from 24 percent to 31 percent, secondary usage declined from 12 percent to 8 percent.&M; Unlike either retail channel, mail order continues to be more widely used as a secondary distribution source rather than a primary one.&P; While 10 percent of these sites now purchase micro peripherals primarily via mail order (compared with 14 percent in 1988), 24 percent rely on mail order as a secondary source (down from 26 percent last year).&M; Distinctions Between Site Sizes&M; When comparing survey results in terms of size of installed base of microcomputers at panel sites, several notable distinctions emerge between large and small sites in purchase-channel preferences.&M; Retailers' corporate sales forces are the dominant primary purchase channel at sites with 100 or more PCs, accounting for 36 percent of these sites, up from 28 percent in 1988.&P; In contrast, only 6 percent of such sites purchase peripherals primarily via mail order, down substantially from 13 percent last year.&M; At sites with fewer than 100 installed micros, panel members identifying mail order as a primary purchase source increased from 14 percent in 1988 to 19 percent as of the later survey.&P; Significantly, most of this growth has come at the expense of retail outlets.&P; In 1988, 40 percent of panel sites used retail outlets as primary sources for peripheral purchases; 36 percent do so today.&O; Each document contains the DOC, DOCNO, and TEXT fields, with different possible additional fields between the DOCNO and the TEXT markers. The JOURNAL, TITLE, and AUTHOR fields contain respectively the journal, title and author of the material in the TEXT field. The SUMMARY field contains a summary of the full text within the TEXT field. Sometimes only an abstract is available instead of full-text, and in these cases there is no summary, and the abstract is contained in the TEXT field. The DESCRIPT field contains manually-indexed categories for the document. The DOCID field is the identifier used in the original data. Other fields in the data are: ABSTRACT, PRODUCT, ADDRESS, COMPANY, CATEGORY, SPECS, and NOTE. Some punctuation ("&", ">", and "<") has been put in SGML representation ("&", ">", "&less;"). Additionally SGML marks included in the data ("&M;", "@O;", "&P;", "]", "&lsqb"; "+", and "=") are left in the data.