Stephen Ullstrom

The words I dread the most from a client are, “You only have x number of pages for the index.”

Those words put me on notice that I will probably have to censor myself as I write the index. Instead of thinking, “How can I make this information most accessible?” I start to think, “Of the access points available, which should I use and which can I do without?” I start to triage potential entries and subheadings, labeling some for potential deletion, while others may not even get included in the rough draft. This approach is the opposite of how I want to index, which is to be comprehensive and inclusive. I want the information in the book to be fully accessible. Not surprisingly, my indexes tend to be on the thick side. 

But, sometimes there are limits. Maybe the client can’t afford to add an extra signature or half signature. (Books are printed in blocks of pages called signatures. See here for more information.) Or maybe the full index will only fill part of an extra signature, and again, it just isn’t worth the extra cost to only fill an extra page or two. Publishers generally don’t like empty pages at the end of a book. In some cases, the client may have a pre-established notion for how long the index should be, perhaps based on a formula to calculate the length of the index. But I would suggest that every book is different. A formula can give you a rough estimate, but it doesn’t take into account the content of an individual book. For these clients, a longer index might be acceptable if you can justify it. 

So, step one, if told that there is a space limitation and if you have examined the proofs and are leery of the page limit being enough, is to ask if the limit is negotiable. Because it might be. Maybe you won’t be able to get the full number of pages you want, but you might get a few more.

If you find with a particular client that you often have trouble with the space they allot, you might also want to contact them to discuss the issue. They may be unaware of the problem, there may be factors for the length of the index that you are unaware of, and having a frank discussion about the index length may lead to happier times on future projects. At the very least, raising the issue shows that you are engaged in the client’s books and in doing the best job you can. If nothing changes, that might also be an indication that you should fire the client and find another client or two that share your indexing vision.

But even if you do decide to eventually move on from this client, this particular index still needs to be finished. You know it is too long. How are you going to shoehorn it into the allotted space?

I always start by prioritizing what I want to keep. This may seem counterintuitive, because the immediate goal is to make the index shorter. But it is still important to keep the ultimate purpose of the index in mind, and that is to make the information in the book accessible. So what, out of all of the entries, absolutely needs to be in the index?

The metatopic is a good place to start. What is the book about? Are there specific people or case studies that the author discusses at length? Concepts that carry the argument forward? Let’s call these the core entries, the entries that lead the reader to the meat of the book. These entries get the comprehensive treatment, and I will do my best to leave them unscathed. 

The rest of the information I prioritize in decreasing importance. Think of it as concentric rings around the core entries. When I start cutting, I start with the outermost ring–the least important entries–and work my way towards the core. The goal is to get the index to fit while inflicting the least damage. 

How do I do this? There are three aspects of the index I focus on.

As I mentioned, I sometimes label entries as I index. One type of label is for entries that seem minor or tangential to the main subject of the book. Some of these might be considered borderline passing mentions, and I pick them up in my first pass through the text in case they are discussed more fully later on. In a comprehensive index, I would probably include most of these entries. If the index needs trimming, these can be a good place to start. Since they are labelled, it is pretty fast to find and cut. 

Another easy cut is to get rid of subheadings. My usual rule of thumb is to add subheadings if there are more than 6 locators per entry, but if space is tight, I might increase this limit to 8-10 locators. Another option is to combine subheadings. This way the subheads are not as precise, but there will be fewer of them. The importance of the entry is also a factor in deciding how to deal with the subheadings. Where on those concentric rings does the entry lie? If it is close to the core, I am less inclined to heavily modify it. If it is more of an outlier, I’ll cut deeper. 

I also earlier mentioned reducing access points to the information. One common way to provide multiple access points is to double-post, say under both an acronym and the full term or name spelled out. Another is to have the information as a stand-alone heading and as a subheading (or two) under the relevant umbrella terms. You can also think of multiple access points in terms of an event, for example, the bombing of Pearl Harbour, and then the different people, organizations, and concepts involved in that event. By providing these multiple access points, the reader can search using whatever term they want and will still, ideally, find the relevant information.

But maybe there isn’t enough space for all of these access points. So you have to think, what are the most important access points? Instead of double-posting, what is the reader most likely to look for? Is it better to have stand-alone entries or to consolidate entries under umbrella terms? Can nested entries be used to concisely delineate the metatopic, instead of scattered entries? Can cross-references be used judiciously to steer readers to the one and only access point? The goal is to still make the information accessible. The access points can’t disappear entirely. You just have to be a little more creative on how to make those access points pop.

You may be wondering, how do I decide which are the most important access points? How do I prioritize the information and entries? I admit that it is not an exact science. You will have to make a judgment call while knowing that someone else might make that call differently. You might also find that you need to change your mind as you cut. Those subheadings that you were holding onto? The index is still too long, so bye-bye subheadings. Those people who played a relatively important role in that event? Well, they now need to be reassessed as relatively minor, and cut. Cutting the index like this feels like intentionally dulling a sharp knife. Sometimes that is what you have to do.

Throughout all of this, try to be consistent in what you cut. If you decide that acronyms need to go, then take out all acronyms, unless there is a very compelling reason to keep one or two. When I cut, I still want the final index to look thoughtfully structured. I don’t want the cutting to be random, or for a reader to wonder why one entry is present and another term, identical in type, isn’t.

It isn’t fun trimming an index that you already consider to be print-worthy. Especially if you have to cut deep. That hurts, seeing your hard work reduced to something less than what it was. But if you do have to cut, have a plan. Keep the ultimate purpose of the index in mind–making the information in the book accessible. Understand the structure of your index, and understand how that structure might have to change as you cut. Figure out how to prioritize the content of the index. This will make it easier to identify, out of the hundreds or thousands of entries, what can be cut. Do all of this–a controlled cutting process–and the end result should still be a usable index.

What are some of your favourite ways to trim an index? What do you cut first?