Hello,

How do you know when an index is finished?

I think this is an important question, and also a question that is difficult to answer. “Good enough” can be subjective. Time and space constraints may be factors, as well as skill level or knowledge of indexing best practices. What an indexer might nitpick at may be perfectly acceptable to an author or publisher.

And yet sooner or later, if under contract, the index needs to be submitted. There is a point at which the index needs to be declared done. When is that point, and is an index ever really done?

It can be tempting to hold up perfection as the standard. I am certainly guilty of having perfectionist tendencies. I also know that if I don’t temper my perfectionist ways, I will never finish anything. I have learned to declare indexes—and other tasks—done and to move on.

So no, I do not write perfect indexes. I assume that they are not perfect. In the moment, when I submit an index, I can’t think of anything more to change, but I’m certain that if I give myself some distance and go back and review, that I will find something to correct, or at least to tweak.

Just a week or two ago, I flipped through the index in a complementary copy a publisher had kindly sent me and discovered that I had used a See also reference when I should have used a See reference. I think I know how that mistake happened, when I consolidated two arrays and must have overlooked changing the cross-reference. It is a minor mistake; readers are still redirected to the correct location. But still, a mistake.

When I served on the Award Committee of the Indexing Society of Canada/Société canadienne d’indexation (ISC/SCI), administering and judging the Ewart-Daveluy Award for excellence in indexing (I stepped down this summer after four years), the other committee members and I frequently agreed that we were not looking for perfection. An absolutely perfect index is an unreasonably standard. So if not perfection, what were we looking for?

We were looking for consistent excellence across the board. The odd error here or there wasn’t that big an issue, but it was a problem if we noticed a pattern, such as poor cross-references, an unwieldy structure, or over- or under-indexing, for example.

Despite most submissions not meeting that high bar, I can also honestly say that most of the indexes we judged were competently written. Despite having one, two, or three areas in which the judges deemed the indexes fell short, the indexes still exhibited many excellent qualities, and clearly showed the thought and work put into them. I think most authors and publishers would be very happy to publish those indexes.

So perfection is not the goal. Even being award-worthy is not really the goal. (Though I do encourage you to do your best, to strive for consistent excellence, and to submit your best work to indexing awards. And, coincidentally, the call for submissions for the Ewart-Daveluy Award just dropped.) It is good enough that the index be competently written.

But still, how do you know when the index is good enough?

I don’t know if I have a good answer. While best practices exist, the application of those best practices is so contextual. But I do have a few thoughts on signs that an index may be ready to send out into the world.

• Work until you feel like you’ve done your best. I realize this sounds trite. I still think it is true. While there may be ways in which you can grow your indexing skills (I still hope that I can grow my skills), the best you can do right now is the best you can do right now. When the index seems done and you can’t see anything more to do, then wrap it up and move on. As you learn, try to do better next time.
• Know and follow your process. Know and implement best practices. When I index, I have a sense for what I want the index to look like and I have a process for getting there. Once I complete that process, the index is done. You can do likewise. If it helps, write down your process, along with all of the aspects of an index that you need to check. If you have a good process in place, then you can trust the process and know that a good index should result.
• Acknowledge the law of diminishing returns. Given the opportunity, I could tinker endlessly. But I also have deadlines to meet and other indexes to write, and continuing to revise is unlikely to uncover major errors. At a certain point—when I finish my process—I need to trust that the index is good enough and let go.
• Let someone else review the index, if you are uncertain whether the index is ready. Fresh eyes can find a lot more. This could mean asking a peer to review. This can also mean sending the index to the author or publisher. I appreciate it when I know that the proofreader is also having a look and hopefully catching any final errors I missed.

Knowing when an index is good enough for publication is tough. Best practices exist (see, for example, the American Society for Indexing (ASI)’s best practices, and here for the criteria for ISC/SCI’s Ewart-Daveluy Award), and yet assessing our own work can be difficult. We want to produce good work, and clients want to receive good work. Deadlines need to be met, and we don’t often have the luxury to edit endlessly.

Not aiming for perfection can remove some of the pressure. As is remembering that skill-building is an iterative process, and that there is always another opportunity next time to do better. It is okay to declare an index done and to move on. It is good enough for the index to be good enough.

Yours in (hopefully not too error prone) indexing,

Stephen