Month: August 2020

In every system, we get to make a design choice – do we make it easier or better?

It’s the iphone question. When the first iphone was released, it was not easier to use than a nokia and it’s still not. Nokia – unlock, dial, call – simple. So who wants to go back to that easier model? Who even really thinks about a phone as a device for calling now? What people wanted was better.

When we always assume people want easier, we leave a massive opportunity on the table – the opportunity to make the system better.

Most people I talk to think it’s easier that users want. They talk about user complaints about clunky interfaces and how hard it is to get people to use it – because of the clunky interface. The simple truth is that if the interface doesn’t do anything they care about, it isn’t going to matter how easy we make it, and people don’t care about records.

The other problem is that the largest barriers to making a system “easy” are imposed by us. We don’t really want to make a system easy – if we did, the first thing we would do is abandon FAT classifications and let users organise their information for their own convenience. Records systems get much easier with no structure, no governance and no controls applied (even if it’s eventually a total mess), but I find few people who are willing to compromise like this.

I think this unwillingness to compromise has lead to where we are now – a place where we take really seriously the idea that users would just use a records system if it was easy enough. The truth is that it doesn’t matter how easy we make make records systems because records is our job, not theirs, and they only care about their job, if we don’t give them a way to do their job better in the records system, they’ll continue to not want to use it.

So there’s only one definition of a better records system that makes sense.

“A better records system helps a user do a better job at their job.”

Better is what makes people go to the gym, it’s what makes them go to university, it’s what makes them apply for more difficult jobs – and it’s what can get them to use records systems, no matter how many additional milliseconds each click takes.

There could easily be hundreds of questions in this post, here are 6 that I think are important.

1. Whose job does the system automate?

Most records systems only automate one job – ours. If it’s only making our life easier we’re losing a valuable opportunity to give business users another reason to use a system that will inherently be less friendly than the average file server.

2. How many people use the system to organise and drive their work?

People love systems that make their work form an orderly queue. If you can make your records system perform this function, people will pretty quickly feel like they can’t live without it. If you don’t provide this, people will invent it for themselves, somewhere else.

3. How much of a barrier to easy system usage have we put up?

In records, we LOVE metadata. How we capture it though, can create a significant barrier for users. A user who gets a project plan delivered to them with 37 documents in it, who then has to select from one of 25 record types and populate 11 pieces of mandatory metadata for each is unlikely to use the system for very long. This doesn’t mean that we shouldn’t do it, it does mean that we should weigh the value of each piece of metadata against the drop off in usage that capturing it will cause.

4. Does your business classification scheme match the way people organise their work and think about their jobs?

This is related to point 1. If people can’t see their job in the classification system, you’re fighting with how they want to organise their work, and you’ll lose. People will struggle with how to add information to the system, then they’ll struggle with how to find it. Eventually they’ll keep working data somewhere that makes sense to them, and eventually they’ll stop using the records system.

This does present a bit of a quandary for records managers. Function-Activity-Classification is an important part of standards compliance and an accepted part of records doctrine. I also think it’s partly responsible for many failed records programs.

5. Have you automated as many capture points as possible?

The best way to have people use a records system is as a place to go and get things out of, rather than a place to put things in. When we automate the capture of records, this becomes possible and completely obvious good things happen.

What’s less obvious is what automation does for our ability to report. If we can automate capture at a process initiation point, we know how many times a process was initiated. When we know how many process instances we have, we can work out how many records we SHOULD have overall and can report against that number. With good reports, we can give managers data to manage their teams – making records performance management a much simpler task. Which leads us to point 6.

6. Are you providing leading indicators to your executive that at some point in the future your organisation will fail critical compliance audits and enable corrupt?

Poor Recordkeeping has been found to be an enabler of corruption and abuse an uncountable number of times. An organisation that is investigated by a corruption commission, a royal commission or a regulator risks being wound up or heavily fined. This is C-level risk and we have to treat it that way. If we are not reporting organisation ending levels of risk to people senior enough to deal with it, we are missing the largest piece of leverage that we have. This level of risk cannot be ignored – and if it is, we should ensure that it has to be wilfully ignored.

Evidence that poor records performance represents this level of risk can be found in the recent AUSTRAC actions against the banks (resulting in over a Billion dollars in fines) and Corruption and Royal Commission reports – in particular the Royal Commission into Institutional Responses to Child Sexual Abuse in which section 8 is entirely dedicated to Records failures.

I think this is a critical but not obvious question for Records Management.

I think that a lot of the time, we get focused on the technicalities of our profession – so we end up storing for destruction, storage (permanent retention) or so we can give the record to someone else (archive).

So what about usage? And users?

Most problems I encounter are problems related to getting people to use the system.

Despite that, we aren’t having problems getting people to store records. 

They store them on file servers, in their inbox, on their laptop, in dropbox and in 20 other places.

The one thing these systems all having common, is that the users get to organise their content for usage.

It’s not function-activity-transation, it’s “emails from CLIENT”.

What if we started with the question “what are your future information needs likely to be, and how can we facilitate that?”

Would that get us better results with users than “how do we comply with all relevant legislation, ISO15489, ISO16175 and the local metadata standards?”

It’s the difference between a system that stores five records a long way apart – when they are often used in conjunction, and another that says “here are the five records you regularly use, conveniently stored together”.

The problem with this idea, is that to function it requires a different input and exit path, and that’s not something around which our heads are easily wrapped, and not something that’s easy to design for.

The functionality to enable this is available in most records systems – in TRIM it’s user labels, in Objective it’s the handy, in explorer it’s shortcuts and the quick access bar.

The problem with these functions is that to work well, they require a user to think like a taxonomist and design a taxonomy for their job ahead of time, and then implement a structure that will help them do their job – which they likely don’t yet fully understand.

As well as not understanding the job yet, ordinary people also aren’t used to designing classification schemes or taxonomies. They don’t immediately think about their job in ways that expose their well structured information needs. What they typically do is keep things on their desktop, organise it a few times and eventually settle on something they like – order is emergent.

To me, this is another opportunity to move the needle just a little for records. It’s an opportunity to get people to see us as being there for them and not just people who built a system that they like, and enforce behavior that gives them little benefit. 

It does require us to go 1:1 – or at least 1:activity, and it does require us to invest time in understanding the information needs of the people we support every day. Ultimately though, if we can understand how their job is structured, and bring them structures that provide easy access to the information they use most, the records system will get used, because it will be useful, and the shortest path to what they want.

All it requires us to do is think about how we store information for usage. What that really means though, is how we store information based on the needs of other people, and my belief is that when they know we are thinking about how to help them, they will think about how to help us.

I know that a lot of people already do this – some of my earliest good recordkeeping moments came from people helping me with exactly this. For everyone else, it will make a difference to how the organisation sees you, and it takes records from being a time sink to being a source of value.

Loss aversion is the idea that people fear losing things more than they will enjoy the gain from something new.

A Nobel prize was awarded for work in this field for a paper called “prospect theory – an analysis of decisions under risk”. What is says is that the gain we make from a purchase has to be “worth” 2.3x more than whatever we’re giving up to make that purchase (generally money) – or we won’t make the purchase.

This is problematic for records.

We are always proposing a loss of control.

We are almost never proposing a gain that will be felt by the person whose control we are taking.

I think that people respond exactly the way that the psychology research tells us they will – they hoard. They used to hoard in their office – paper under desks, in file cabinets, on desks etc. Now they hoard digitally.

While an argument could be made that the move from custodial approaches has made this less of a problem, I’m not sure that’s true.

We still see troves of records on people’s hard drives and practically anywhere else they can organise information for themselves and control placement. The volume is higher, and the challenge of figuring out what is redundant, outdated or trivial is just as low on the priority risk.

I think this behavior is a cue for us that we’re missing something. I think it’s an existential question for records – how can we present larger gains to people we need to do the record keeping?

Do you keep a near miss register for Records?

A near miss in records should be a time when policy wasn’t followed and the organisation was exposed to a specific risk – without that risk becoming an incident or a loss.

Near miss reporting is a common practice in safety management. It’s there to expose all of those times that the organisation avoids an incident because it is lucky rather than because it did things properly. I think it’s probably a useful tool that will help us deal with our creeping problems by identifying and giving us data on the leading indicators of major failure.

The idea is simple – every time we have a problem because policy wasn’t followed, we record it. This doesn’t mean that we record every file that’s on a file server somewhere – because no one cares about that until theres an actual problem. It means that we record the times when that kind of practice exposed the organisation to a specific risk. It’s a way of starting to understand the magnitude of the creeping problem of records non-compliance.

I don’t think anyone would disagree that records compliance is a creeping problem for lots of organisations. Every corruption report and every royal commission lists records being falsified, accessed improperly, not kept or otherwise ignored as a part of the problem that lead to the investigation.

I think a near miss register provides a tool that could turn these problems into hard data problems that executives can’t ignore (or have to specifically ignore).

I think that ultimately, we have two options – one is to report on the near misses, and hopefully find a solution. The other is to wait until there’s a catastrophic failure – as we’ve often seen. At that point, the records will be needed, and if they aren’t there, and we can’t show the data we provided over and over again to highlight the problem, we will have to accept a level of blame.

The idea of a near miss might be a bit nebulous, so here are a few examples of the kinds of things I mean –

1. The time a record required for an FOI (or royal commission) was in a repository where users have uncontrolled delete privileges.
2. That time we audited HR and found that they were keeping confidential information on a file server with no access control.
3. That time IT had to restore a file share because records were deleted to make way for a new project repository.

In each case, the organisation was exposed to a specific risk by a violation of policy.

I don’t think this is a magic bullet, but I think it’s one more tool that we can use to start turning the tide.

If this is something that you do, or an idea that you like – please email me or leave a comment, I’d like to see how well it works for you.

Records are important.

Data is important (sometimes).

So why is data getting all the investment, when at some level, all records are data and data of any importance is a record? I think it’s down to one thing.

People value what they rely on.

Data people are busy creating things that people will rely on.

Data people are focusing on urgent problems, and because of the nature of investment, they’re being hired for to organisational needs that are both urgent, and important.

Records is important, and always will be. But always is a very long time, and the truth is that most of what we keep simply isn’t important, and won’t have value – ever.

Data people have embraced the idea that what they’re doing has to be seen to have value now, or they just don’t get funded.

A lot of what we do is to manage information that only gets touched twice – once when it is stored, and again when it is batched for destruction. In between those events is a huge opportunity.

Within the records we keep is every decision the organisation has ever made, and the information they used to make it.

The next generation of great records managers are going to get funded by ensuring that people making decisions are relying on records to understand how they should make the decision.

The more people rely on our records of the past, the more value they will have into the future, and the more funding we will get.