I recently published a review by Carl Grant, President of CARE Affiliates, of one of the chapters in Christopher Cox’s federated search book. Beyond addressing the content of his review, Grant wrote that he didn’t think the RFP process worked well for either customer or vendor. He is referring to the process where a library issues a “Request for Proposal” (RFP) with difficult and detailed questions which each prospective vendor has to make a significant time commitment to respond to. The vendor in turn has to bury the cost of responding to RFPs in the purchase price of the systems it sells.
I was curious to learn more from Grant. What did he see as a better approach? I asked him and he responded via email and through his blog. He wrote:
Just this week Sol Lederman, author of the FederatedSearchBlog posted a review I wrote about a chapter in a new book that covers using RFP’s to purchase federated search systems. I’ve long been critical of the way this profession uses RFP’s. I’ve published an article on the topic before that while now in need of updating, still outlines substantial costs and problems with these tools. I’ve suggested an approach in another article I co-published. At least one company has been formed to try and address the problem for this profession. Yet, until it gets focus on a much wider level, they’ll continue to remain a costly and inefficient purchasing tool. Everyone agrees they’re painful. I guess not yet painful enough to really do anything about it.
I was very curious about the article Grant wrote for New Library World in 1999, while President of Ex-Libris (the first article mentioned in Grant’s blog post): Choose wisely: making the library’s money work for the library in the system procurement process. The article is available for a price but Mr. Grant kindly provided me with a review copy.
The article’s abstract hints at Grant’s major concern, that the familiar RFP model is a lose-lose one.
Procurement processes for automated library systems need examination to determine if libraries are choosing the best methods for getting the best return on their investment. Understanding the costs involved and how typical decisions in procurement procedures add to the costs of systems will allow libraries to judge if the process better serves the library than using the money in other ways. Alternative processes including standardized request-for-proposals, contracts and understanding how to utilize lawyers and consultants are suggested in an attempt to show that procurement processes could achieve the same goals using different methods while allowing libraries to use their money to purchase other goods and/or services. The benefits of the implementation of such processes are also reviewed.
Grant argues that libraries looking to procure an automated library system, closely related to federated search although he doesn’t use the term, drive up the cost of their own procurements by forcing vendors to jump through the RFP hoop. While the article was published nine years ago and the cost figures are out-of-date, the reasoning is still sound. Grant cites several frustrations with the RFP process:
- The cost to answer an RFP. The article states a 1999 figure of $5,000. That includes travel and staff time. I would guess that number is much higher now. Perhaps the travel can sometimes be replaced by web meetings but person-to-person relationship building is important as well.
- The expectation that 50-80 percent of the RFPs answered will not lead to a sale. This frustration is compounded by the painful reality that many libraries have already selected a vendor and issue RFPs because they are required to conduct a “fair” procurement. The pain all vendors feel is obvious from this statement from the article:
In effect the libraries transfer the cost of doing a sole source procurement from themselves to the vendors by submitting RFPs to firms whose products they have no intention of purchasing but to whom they need to submit a response in order to justify the procurement of the system they actually want to purchase.
- The cost to answer 100 RFPs in a year in order to sell 30 systems. At the $5,000 cost to answer an RFP, it costs half a million dollars to answer 100. That $500,000 is then divided among the 30 systems sold, adding $16,666 to the cost of each.
- Demonstrations and the cost to staff them. I would add the cost to produce them and to tailor them to the customer’s needs.
- The need to support complex computing and networking environments that were not purchased from the federated search vendor. When vendors sold hardware as well as software and services, there was profit in these sales which could be used to provide support. Now the vendor must bear that cost from software and service revenues. The customer doesn’t really save by purchasing their own equipment.
How can libraries lower their procurement costs? See Part II.
Tags: federated search