Records management practice and the economics of alternatives.

One thing that is absolutely true about all people everywhere, is that people like doing things the easy and cheap way.

Over the long term, the competition between ways of achieving an outcome is always decided by the economics of the way it is achieved.

This is why we are seeing the power and prestige of the records management profession eroded now.

25 years ago when we lived in a paper world, the advantage gained by having records managers in your organisation was so large, that I don't think it needed justification - because every other way of dealing with large volumes of documents was harder and more expensive.

Consider the economics of some of our most basic tools - transactional aggregation, cataloguing and description.

If you have five hundred boxes of documents with no particular order to them, the only way to find what you need, or to be sure you don't have it, is to read five hundred boxes of documents.

The numbers on that are not friendly.

An average person reads about 240 words per minute - about a page.

Lets assume a document box holds 3000 pages.

Based on these assumptions, reading 500 boxes takes about 17 years (3000 minutes per box, 500 boxes, 420 working minutes in a day, 210 working days in a year)

Those are really bad economics.

You could not build any business around economics that bad.

Unless of course, the business you were building revolved around a set of tools and practices that could make the economics of that problem good enough that you could operate a business that relied on them.

Enter our tools - transactional aggregation, cataloguing and description.

Aggregate your transaction documents into a file (record), catalogue your records and describe them well, and the economics of that problem change dramatically.

17 years becomes a lot less - maybe a day if its a 500 page file, and you have to read every word.

What used to take 17 years (6205 elapsed days) now takes 1 DAY.

This means that in the same amount of time, we could achieve 6205 more things for the same cost that we used to achieve 1.

That is phenomenal. Lets make a comparison that shows just how big this really is.

How much faster does google make people?

Assuming I have a library within an hour of where I live (I actually have one less than 5 minutes away, but lets be generous to poor google), and it has what I want, Google probably reduces my time to find the information I need from about 70 minutes, to somewhere between 10 and 30 seconds.

Records management helps people find a transaction in about 1/6205th of the time (assuming the reading rate and corpus size above).

Assuming a 10 second find time, Google helps us find the facts we want in about 1/4200th of the time it takes to go to the library.

It's not a perfect comparison, but it certainly helps illustrate just how big a difference records management made in a paper world.

A bigger difference than google.

The obvious catch that you all saw coming, is that we aren't in a paper world anymore.

And as we said at the start, over the long term, the practices that win are the practices that have better economics - because over the long term, people will always default to the easy and cheap way of achieving the outcome they need.

This is important, because the economics of electronic documents and fileservers are very different, and we made some real mistakes when we tried to adapt our practices to an electronic world. Changes that catastrophically impacted the economics.

Lets start with file servers.

For starters, people are awake to the practicalities of storing things on file servers - so they generally organise, catalogue and describe them without having to be asked (it's not generally a great, but it gets the short term job done).

There is also search - title search at a minimum, content search often (which I know doesn't always help).

Let’s be clear up front, information management behaviour of non-professionals is not consistent over time. This means that fileservers are always optimised for the work being done now (what we might call the "active" phase of the lifecycle. This means that once routine operational needs are done, it is often hard to find transactional information on file servers.

With this in mind - lets assume for the moment, that for active lifecycle phase needs, file servers and records management practice based systems are equivalent in terms of economics (they aren’t, as we’ll see).

SO if we are economically equivalent for that phase, we need to focus on quantifying the longer term economics so that our organisations see us as having more favourable economics over the long term - which would mean (following our opening rule) that we win.

So lets quantify the economics.

The first thing we have to consider, is how much usage there is after the active phase of the lifecycle.

Generally speaking, about 95% of records don't see any use outside the first year.

As a baseline, this means that if we’re asking for effort on every record, 95% of any effort spent arranging, cataloguing and describing content for long term needs is wasted.

This also means that the payoff from any effort we expend needs to be worth at least 20x what it cost (only 5% of the effort has a payoff). In simple terms, this means that every minute or dollar of effort we ask for needs to save 20 minutes or $20 later on.

We might be able to make that case - if we had thought about it, and were thinking about it when we designed our records program.

I mentioned a huge (catastrophic) mistake that we made earlier when we went electronic, which to me is one of the things that has really brought records practice undone, and made its economics uneconomical.

The mistake is this - we used to describe at the file level.

When we went electronic, we took this down a level, and while we kept classifying at the file level, we started asking people to describe at the level of the individual document.

What does this mean from an economic standpoint?

Let’s assume that there are ten documents per file. That means describing at the document level added 10x the work to the process of description over and above what used to be done (by us), and over and above what happens on a fileserver. This means that if description used to be a 1 minute task, it is now a ten minute task.

When we add these economic factors together, it means that the payoff from thinking long term, and employing records management practice over just having a file server needs to be about 200x the effort we ask for.

If you're an organisation adding 10000 new documents per day to your file server, and description would add 20 seconds to each document, that means that 1111 minutes (lets call it 20 hours) is being spent each day describing records (about 4000 hours a year).

If you've agreed with the maths so far, 95% of that time is spent on records that will never be needed again.

So out of the 4000+ hours consumed by description in a year, the value created by the increased description of about 500 documents (about 50 files/records) has to pay for the extra effort and pay for you, and your team, and your tools.

And the value minus the cost equation has have a better payoff than the alternatives do - alternatives which could just be a fileserver and search, or an ediscovery product and team, or an ever growing pool of cheap storage, or the risks that might be created by a default rule of just destroy everything after 25 years - etc.

If you've been paying attention, you're probably struggling with the maths, and feeling like this is very complicated.

The thing you should consider here, is that the executives who fund us feel like this is very complicated too.

They are trying to assess the value added by records practice, in the context of the cost it adds.

While doing the same thing with other alternatives.

If you're struggling with the maths, imagine how executives are going when they don't understand records management practice at all, and have very little professional incentive to deal with the complicated set of economics.

And very little incentive to deal with long term problems.

And many priorities that probably have a set of options and payoffs that are easier to understand.

Is it any wonder that they go with the fileserver, or the cloud storage, or the business application now, and deal with the long term problems when they occur?

This has been a long and complicated post, and I hope you've stuck with me to here.

There are three underlying points that I think we all need to consier.

The first consideration, is that we have competitors. Our practices are in competition with other ways of achieving the same goal, and the competition is ultimately an economic one. In the short term, ideology can win, politics can win, crises can win, regulators can win. Over the long term, the winner in any human game is the economical one because all human problems ultimately come down to affording food, and when we can’t afford food, the unnecessary effort or effort on long term problems gets cut in favour of eating.

Whatever goal we are employed to achieve, if our solution has worse economics than another way of achieving it, we are a temporary solution to that problem that can and will be replaced by the more economical one.

The second consideration, is that we have to consider and reconsider our practices in terms of how they affect the economics of the problem we solve.

It may have seemed like a great idea to describe at the document level when we got EDRMS and each document became a discrete thing that we could add metadata to. I'd bet archivists and historians were practically frothing at the mouth with excitement about it.

Then we went non-custodial though, and put 10x the work on rank and file employees to achieve the same task as the fileserver.

We also stuck a BCS or some kind of keywording system in front of them that cost them even more effort to decipher, and often made it harder for them to find things in the future when compared to the fileserver.

This is to say that we should look at the economics of every single core records practice, because they were all invented in the paper age when records managements saved people more time than google, and every other option was worse, now there are other viable options, and they are winning because their perceived economics are better.

The third consideration, is to remember that the economics are complicated. This is true for both the short and long term economics.

The only group that has an incentive to do the work on the economics of records management and the comparison to its alternatives, and to adapt new practices that solve the same problems even more economically, is us.

The bad news is that if we won't reconsider and reform our own practice economics, someone else will do it for us.

They won't do it by coming to us and saying "hey there, nice records people, your practices aren't economical in comparison to their alternatives anymore, lets all sit down together and work out how we can make them more economical so I can keep giving you this money instead of giving it to these nice people over here who seem like they're providing an easier and cheaper way to solve the same problem."

They'll just spend the money somewhere else.