A library should not be judged on how many books it owns, but the quality of the collection. The problem using statistics alone (like HAPLR does) is that it only counts volumes. Anyone who has taken a weeding class is aware of how reducing the collection increases its use, and growing the collection with unused materials reduces use. So I can make the determination of looking good on paper (making a grandiose statement that I have over 100,000 items in the library's collection), or I can actually create a leaner collection that is used more often (increasing circulation by double digits). Since circulation levels are such a moving target, most agencies that track success, only want to know how many items are in your collection. For instance, when you calculate impact fees they only look at the cost of replacing the entire collection. So if you have 100,000 items, what would be the cost of replacing that collection if you lost everything in the building. The more items you have, the more money you will have for future books, circulation is completely ignored. What is the point of being a successful library if you can't get credit or increased funding for it? HAPLR, impact fee studies, and the similar tools hurt libraries by looking at the wrong stats and not counting all of the stats.
This also goes for strategic planning. I can create a great strategic plan, but most of the circulation, computer usage, reference questions, database hits and program attendance goes to items the library has always performed successfully. Strategic planning works because it gets the community involved in planning for library services and demonstrates that value. However, going too far in sticking to the plan only makes you look good on paper, and does not really create value in the community. Is it better to have a program that the people want and have great attendance, or it is better to stick to the plan, create a program you think people want, and get poor attendance? Create the same scenario for collection development or database usage. I think the strategic planning model is better than just straight statistics, but both have trouble capturing true success. This is the balance I walk and most of the time everything falls into place. It is just very strange how difficult to demonstrate that impact easily.