EUSIDIC‐EUROLUG Survey of Document Delivery Services in Europe, 1996

Participants

In all, data from 167 organisations were analysed (22 organisations responding to the survey made no requests in the survey period), and these were spread across 16 countries. The country with the largest number of participants was Spain, with 29 compared with 24 from its nearest rival, Finland. It would be interesting to know which country generated the most requests, but (except for requests sent to BLDSC and INIST) this information is not provided (a grave loss, presumably, to document suppliers wishing to use the survey to target their services). In terms of activity, education/ academia was the largest at nearly half of the respondents. A breakdown of organisations by the annual number of document delivery requests made is provided. Thus, for instance, 48 of the responding organisations make more than 2,500 requests a year. What is not given is a table of the number of requests by type of organisation, country and type of request.

Requests

A total of 7,337 requests were analysed. By far the largest number of requests were for journal articles, with requests for the next largest, conference proceedings, coming 2,513 behind at only 260 requests. For each type of request, the number of requests made, the number not supplied, and the relevant percentages are provided. There is, however, no indication of the reasons for the failure of requests by their type, although there is a table providing reasons for non‐delivery by the total number of requests made. It may of course be that, in many cases, there was simply a delay in, rather than a failure to supply. The highest success rate by type of material requested is for patents with an impressive 96 per cent. For the record, 83.5 per cent of journal articles and 73.5 per cent of conference proceedings were supplied before the deadline.

Only 593 documents were delivered by electronic means, and then only if the 584 supplied by facsimile are included. (There were, however, an additional 353 documents delivered by “urgent” mode, which, one suspects, includes a number supplied by fax.) Postal delivery is far and away the most usual mode of delivering documents at 66.4 per cent. Next largest (14.3 per cent) was “supplier’s delivery service” which, in fact, largely applies to BLDSC’s mode of delivery. E‐mail/online delivery is negligible at only 0.1 per cent, which leads the reports writers to conclude that “‘the Internet’ is not a major delivery vehicle at this time”, and that “the information superhighway … is not yet a reality”. Again, it would have been useful to have been provided with data regarding the number of items supplied by which organisations to which countries and by what means, since one suspects that geography has at least something to do with the efficiency of particular modes of supply.

I am not entirely sure why one would wish to know the percentage of “documents received or not as a function of their ordering modes”, since I suspect that whether one requests a document by post or e‐mail has more relevance to supply times than to whether the item will be supplied at all. Nevertheless, assuming a causal connection, it should be noted that the highest satisfaction rate is achieved when orders are placed by “Other” means (94 per cent of items requested by such means were received before the cut‐off date, compared with a mere 85.5 per cent ordered by e‐mail; 91.4 per cent of items ordered by fax were supplied; 91.2 per cent of those ordered by telephone; and 63.7 per cent of requests made by post).

Requests not satisfied

A total of 1,309 requests placed during the survey period were not satisfied before the cut‐off date. It is of some concern that for 34.5 per cent of requests not satisfied information was not provided by suppliers. For a further 26 per cent of failed requests no information as to the reason for failure was provided by participants. Of the failed requests for which reasons were provided, the main reason for non‐supply was that the document was unavailable (17.5 per cent), which presumably should be taken to mean “not owned by the supplier”, since those items placed on a waiting list (i.e. items owned by the supplier but currently on loan, or items being purchased but which have not yet been received) are listed separately. Incorrect references accounted for 6.7 per cent and waiting lists for only 3.5 per cent of requests not supplied. For users of BLDSC, it is encouraging to note that, compared with the results of the 1995 Survey, the number of documents not supplied because of incorrect references fell by an astonishing 44.3 per cent from 57.7 per cent in 1995 to 13.4 per cent in 1996. Since BLDSC does not appear to be increasing the amount of checking carried out on requests, one wonders whether this may be accounted for by organisations performing more checks themselves prior to placing orders (a more necessary routine since the number of items not available at BLDSC increased from 20.3 per cent in 1995 to 28.4 per cent in 1996), or whether, simply, the quality of information provided by individual requesters to the requesting organisations has improved with the greater availability of citation information in electronic form. Certainly, we may infer that, where practicable, a little more pre‐order checking yields substantial rewards.

Delivery delays

The overall average delivery delay for documents included in the survey was five‐and‐a‐half days, an average shared by journal articles, reports and patents. As one might expect, the mode of delivery has some influence on supply times: items requested by e‐mail arrive slightly quicker than the average at 5.1 days, while orders placed by post take an average eight days to arrive. The report notes that “Telephone is the most efficient means of ordering as reflected in the delivery delay”, the average supply time being a mere 2.3 days. So far as the UK is concerned, however, items requested by telephone from BLDSC (the supplier receiving the largest number of requests) are urgent action requests charged at a premium rate. It is surprising to note that, when the delivery delay is categorised by the method of delivery, urgent delivery has an average delay of 4.7, and fax delivery 1.7 days. Again, it is not possible from this survey to discover why this is so since there is no linkage in the information provided between the method of delivery, particular suppliers, average supply times, and the type of service (standard or urgent) requested.

Compared with the 1995 survey results, INIST’s supply time has remained roughly the same at 5.9 days in 1996 compared with 5.7 days in 1995, while BLDSC has improved from 6.1 to 4.1 days, probably, as the report notes, because of a move towards electronic media, although users of BLDSC will be aware that BLDSC has also made considerable efforts to improve its non‐electronic delivery times.

Overall, the survey shows that there has been an improvement from 6.6 to 5.5 days average delay between 1995 and 1996, which represents a substantial improvement. Interestingly, there is also an 11 per cent reduction in the use of the postal service for delivery.

Prices

Four categories of price are distinguished: free, standard, premium, and discount. No actual prices are given, however, and while the dedicated reader may decide to contact the named (but, evidently, not the unnamed) suppliers to compare prices, a table of comparative prices which one could compare with suppliers’ delivery times would have been useful. It should be noted that “in 1,205 cases, a mode of payment had been noted. Therefore we have deducted that these documents were not free.” Quite so.

A total of 8.5 per cent of documents requested were free of charge; of the remainder for which information was provided, 90 per cent were charged at standard rate, 2.5 per cent at premium rate, and 7.5 per cent at a discounted rate. For 22.6 per cent of documents requested, no price information was provided.

An analysis of prices paid by public and private organisations reveals that the latter are more likely to use premium services than are the former. Given that urgent requests are more usually charged at premium rates, however, it is odd that the percentage of urgent requests made by public and private organisations is roughly the same at 5.9 per cent and 6 per cent respectively. Rightly, this apparent discrepancy is noted in the report, and it is concluded that more information on pricing would be required before being able to determine with any accuracy the reasons for it.

There is, in my view, a serious error in the conclusions arising from the discussion on pricing. The report notes that “it is interesting to question whether users would really use the premium priced‘urgent’ mode if they realised that the delivery delay (4.7 ± 2.1 days) was mid‐way between that of the post and delivery by fax? (8.0 ± 4.0 and 1.7 ± 3.4 days)”. If this really were the case, and if requesters were to become aware of it, one can see that the use of urgent modes of requesting would indeed fall. However, since there is no combined table showing the delivery delay of requests according to type of material requested, the method of delivery and service (standard, urgent, etc.), it is not possible directly to compare, say, the delivery delay incurred when a journal article was requested using a standard service and delivered by post, and when a journal article was requested urgently and delivered by the same method. Even if items supplied by fax were not requested using an urgent or premium service (and, as noted in the report, one suspects that at least some were), it is difficult to believe that, for instance, a request ordered urgently from BLDSC by a UK library routinely takes 4.7 ±2.1 days to arrive. More usually, one would expect same‐day delivery for urgent requests satisfied by fax, and next‐day delivery for urgent requests satisfied by post.

Suppliers

An impressive 754 suppliers were recorded by participants, 18 of which received 50 or more requests from participants during the survey period. It comes as no surprise to learn that the British Library received by far the largest number, a massive 2,806 compared with its nearest European rival, INIST, which received 257. Unfortunately, this means that, in practice, a large proportion of the results of the survey can only really be meaningful in relation to BLDSC.

It tells us, perhaps, more about the type of material requested than about the quality of suppliers’ services that it was an American supplier, UMI, who received the second largest number of requests (293 in total, 289 of which originated from a single organisation, though this organisation’s type is not recorded). Although the 18 suppliers receiving 50 or more requests from participants during the survey period are listed, the remaining 736 are not. In part, this is for the very good reason that the report’s authors were not able to identify all the suppliers; but there remain those which were identified but which were not listed. This is a pity since it would have been interesting to know what types of suppliers are being approached by requesting institutions and for what type of material.

For information, and assuming that my calculations are correct, the total number of requests received by the 18 listed suppliers during the survey period was 4,543, representing 61.9 per cent of the 7,337 requests analysed. The average number of requests sent to each of the remaining 736 suppliers was, therefore, 3.8.

In addition to UMI, 32 suppliers were based in the US and these received 89 requests; there were 12 other non‐European suppliers, receiving a total of 12 requests. In total, there were 45 non‐European organisations receiving a total of 394 requests (293 of which were received by UMI). As might be expected, then, the majority of requests made by European organisations were to suppliers within Europe (709 suppliers receiving a total of 6,943 requests). The highest concentration of suppliers was within Spain (118), though it should be noted that Spain also had the largest number of organisations responding to the survey. Other countries with a large number of suppliers include the Nordic countries (Norway, Sweden and Finland boast a total of 242 suppliers), Switzerland (54), Germany (41) and the UK (80).

Suppliers were not categorised by type, so that it is not possible to say with any certainty how many of the 754 offer commercial services, nor what proportion of the requests made were sent to commercial suppliers. Of interest, however, is that 103 (13.6 per cent) of the suppliers accepted payment by credit card, and that 340 (45.1 per cent) of the suppliers offered (or perhaps, in some cases, required the use of) deposit accounts. The report notes that many suppliers accept more than one form of payment, but how many of them do so, which forms of payment these are, and which of the suppliers, if any, accept only one form of payment is not determined. It should be noted that although 187 suppliers provided documents free of charge in all cases, this only applies to the documents supplied during the survey period: it may be (since this information was not requested in the questionnaire, nor obtained by EUSIDIC) that these suppliers do sometimes charge for documents. The suppliers of these “free” documents are not identified, but it is interesting to speculate whether any of the documents are entirely free: it may well be that hidden charges in the form of reciprocal arrangements or annual subscriptions which include the provision of “free” documents explains the apparent lack of charge.

The two main European suppliers, BLDSC and INIST, are analysed separately. BLDSC received the largest number of requests (1,475 of the 2,806 it received) from 17 UK organisations, while INIST, by way of contrast, received 96.1 per cent of its 257 requests from French organisations. In both instances, over 90 per cent of the requests received were for journal articles. The failure rate (bearing in mind that some documents are likely to have been supplied after the survey cut‐off date) is remarkably similar for the two suppliers, at 10.6 per cent for BLDSC and 9.3 per cent for INIST. Supply times for the surveyed documents were an average 3.7 days for items supplied to the UK and 4.6 days for items supplied elsewhere by BLDSC, and 5.9 days for items supplied by INIST. The discrepancy may in part be accounted for by the fact that BLDSC uses its own supply service in 31.3 per cent of cases, fax in 16.1 per cent and post in 50.8 per cent of instances, whilst INIST supplies all its documents by post.

I was, I confess, rather puzzled on reading that “post is the main mode [of delivery] … for INIST with … 96.6 per cent of documents. It is the only mode [of delivery] for INIST …”. There may well be a sound statistical reason why 100 per cent should be noted as 96.6 per cent, but I confess that, for my part, I prefer the more traditional x = x to x = (x ‐3.4).

Conclusion

There are undoubtedly difficulties in organising and presenting the data accumulated in such a survey. Nevertheless, at over £100 per copy, one would hope to glean some useful information and, at the very least, for the information given to be presented in a manner accessible to those to whom such information should be of most use. This, however, is a difficult report to read, not only because one must frequently seek to identify information in other parts of the report to make sense of the section one is reading (a task which I would expect to have been done for me by the report’s writers), but because, as I have indicated, a significant amount of information available from the raw data is not supplied. In respect of the main objective of the survey, “to evaluate the type and quality of the document supply services offered in Europe,” I am afraid that little has been achieved. In part, this is because so few organisations received sufficient requests from participants during the survey period for the data to be of much significance, and this can hardly be counted a fault attributable to EUSIDIC. As I have indicated, however, there are things that could have been done.

One of the main difficulties in compiling this report, it seems, was that the information supplied by participants was not always complete. A further difficulty is that the report’s authors do not, and probably for very good reason, seem to place a great deal of reliance on some of the information even when it is provided. In the discussion regarding the method of payment, therefore, it is noted that, since some participants do not know the actual prices of documents, the figures concerning the means of payment “should be treated with caution”, even though this information was actually provided for 92.6 per cent of documents.

The survey concludes that “document delivery ‘Europe’ in the sense of a common market is far from a reality”. It is a pity that this report could not have done more to remedy this by providing the sort of details on which organisations seeking to improve their document supply operations could have acted.