Practical Strategies for Electronic Records Management

So, you’ve spent months developing, reviewing, revising, and approving a corporate retention schedule so that every conceivable record type is covered and users can easily find the policies they need. Now what? How do you turn that policy into consistent practice?
In this webinar, Zasio’s Warren Bean, CRM, offers practical tips on applying electronic records management retention rules to various types of repositories, even the slippery ones like email, cloud drives, and messaging platforms.
Key takeaways: Practical Strategies for Electronic Records Management
- Focus first on day-forward governance of incoming information, then go back and clean up legacy data, like stopping a leak before bailing the boat.
- For each system, ask whether its information is short-term only, whether it has retention capabilities, and whether it is a good place to manage records long-term.
- Short-term information doesn't need to be managed so much as forgotten in a timely manner, ideally in place in the system that created or received it.
- Long-term information belongs in a repository where you can search and retrieve, apply holds, run disposition and keep audit trails, moved there manually or automatically.
- For shared drives, divide folders into active and inactive, purge short-term content on a schedule, and save long-term documents where retention policies can be applied.
- Build retention into application onboarding, so every new system is checked for how long its information will be kept and whether it is short-term or long-term.
Audio-described version
Transcript: Practical Strategies for Electronic Records Management
Duration: 0:49:55 · Recorded: September 25, 2019
Speakers: Chelsie McGuinness, Business Development Representative, Zasio (host); Warren Bean, CRM, MCSE, Senior Sales Engineer, Zasio (presenter)
Lightly edited machine transcript. Slides are summarized where they carry information not spoken aloud.
Chelsie McGuinness: Hello everyone, and welcome to today's webinar, Practical Strategies for Managing Electronic Records.
My name is Chelsie McGuinness. I am the business development rep here at Zasio, and I'll also be your host for today's session. I'd also like to introduce our presenter, Warren Bean. He is our senior sales engineer and has been with Zasio for 19 years. Warren is also well known in the industry for being an information governance expert and trusted advisor.
I do want to mention, if anyone will be attending ARMA InfoCon in Nashville from October 21st to the 23rd, you'll have an opportunity to see Warren present live at this year's Z-Talks, a complimentary pre-conference event that Zasio is hosting on October 20th. Stay tuned for more details after the presentation. If you have any questions during the session, feel free to write them in the chat box, and if we have enough time, we'll do a Q&A at the end.
For those of you who are not familiar with Zasio, we are located in downtown Boise, Idaho. The company was founded by our current leader and CEO, Kevin Zasio, in 1987. Zasio started by offering a full life cycle record management application end-to-end for over 30 years, and has been an industry leader in helping businesses address their information governance challenges ever since. A fun fact about us is that half of the top 20 Fortune 500 companies trust Zasio for their record management and consulting needs. For legal research, we have a dedicated team of licensed attorneys and legal professionals whose primary focus is to research and help keep 120,000+ citations for 130 countries around the world.
Zasio's consulting team can help companies of any scale develop their information governance strategies as well. And with that, I'll go ahead and turn it over to you, Warren.
Warren Bean: Welcome, everyone, and either good morning or good afternoon, depending on where you're at. Today's presentation is basically a culmination of some ideas that I've been working through over the past few years, actually, and just coming from the experience that I have with customers where they're asking about solutions to different problems regarding electronic records and just the results of trying to put together some things that are practical and workable for them in order to encapsulate those issues that come up for most people. So it won't be solving every one of your problems, but hopefully it'll give you a good place to start and give you some practical tools to work with. So with that, let's go ahead and dive in.
I probably don't need to scare you with numbers, but just to give you perspective on the kinds of things that we're dealing with, I wanted to start off with a few quick stats. The first one is 2.5 quintillion bytes per day. That was a number presented by Forbes back in 2018, so just last year, and that is just the amount of information we are creating as a culture on a daily basis. And just to give you some context of how much that is, 1 byte is about 1 character of information that you're typing into, say, a Word document. So that's 2.5 quintillion characters, and quintillion, if you haven't used that number very often, it kind of goes billion, trillion, quadrillion, and then quintillion.
So that's 18 zeros. Lots of zeros, lots of numbers, something that most of us can't even imagine. So lots of data that we're creating each day. How they calculate that, I would have to go into more detail in the article, but that would be beyond my expertise, for sure. And with that, another study showed that a couple years ago, we are creating, or have created, more information in the last two years than in all of human history.
In fact, by now, that stat is probably out of date. We probably have created more in the last year, given the exponential rise in information gathering. And in fact, all of us probably have smartphones, and your smartphone is gathering information right now as we speak. So it is gathering things like statistics about where you're located, health information if you have those stats turned on, obviously emails and instant messages, maybe apps that are loading data in the background, all kinds of stats just as we speak. That is happening all day, every day by everyone who carries a cell phone, let alone all the other types of information that we're gathering.
So it's an enormous problem, just more information than we can wrap our heads around, let alone try to think about how we're going to actually manage that from, say, a business perspective. The data is simply growing faster than we can develop solutions to manage it. And so this is a dilemma I've been wrestling with for a few years. In fact, I remember two or three years ago at one of the MER conferences discussing this with a colleague, and just how we actually try to corral this information and do something with it.
In fact, it begs the question, is complete information governance actually realistic? Can we actually say that we have tools and capabilities to not only wrap our heads around it, but actually wrap our tools around it, so to speak? In other words, to corral it, to know what it is, to know where it's at, to say who needs it, when we can destroy it, and so forth. Well, in reality, probably not. There's just too much there to manage, too much happening too fast, so we'll likely lose more data than we'll ever manage, or even remember, for that matter.
So maybe we should just give up now. Maybe it's too far gone. Maybe there's just no way to actually tackle this. Of course, we wouldn't be here talking in a seminar about it if we didn't have at least some solution. And in fact, in trying to grapple with this, what I've realized is there are tools that we use every day, each of us use every day, that can sort of be a corollary to this problem, and it gives us at least a good framework for working with this kind of information and at least getting some ways to put our heads around it.
So if I were doing this presentation live in an audience, I might ask you what idea I might be talking about and get some feedback from the audience, but really what I'm talking about here is simply human memory, something that we all have. Of course, obviously, if you're impaired in memory, that's another story. I certainly seem to be impaired some days when I can't remember certain basic things, but we all use memory every day, and our brains are pretty good at doing that. And actually, as a test of that, let me bring up our first poll here with the question being, "Do you remember how many bytes of data we are creating each day?"
That was one of those first slides that we just had a moment ago. And we have some options here. Basically, we've given you some multiple-choice questions, and so hopefully you're seeing that poll screen up there now, and you can answer which of those options were presented on that first slide. [poll] It looks like about 70-some percent got the right answer of 2.5 quintillion. Very good.
Excellent. We've got a lot of people with good short-term memories. If I hadn't put up the options there, maybe that would be a little bit more difficult. But the question is, why, if you did remember it, why did you remember it? Or if you didn't, why not?
And the answer is simply that that's information that may be useful from a short-term basis. Maybe you're just so enthralled with the presentation that you're hanging on to every word. But most likely, it's just not a number that's going to be interesting beyond just this discussion. So it may have been something that raised an eyebrow for a moment, but it's not something that you're going to remember perhaps tomorrow or next week, unless you go back and review the slide deck, of course. It's just kind of interesting information, but not critical.
So just to kind of give some additional context around what our brains are doing, if we take a look at some of the astounding things that our brains are doing, we'll see some interesting stuff. So for example, throwing out some more quadrillion numbers here, 10 quadrillion calculations per second, that's how many calculations our brains can manage every second. So as you recall, quintillion is one above quadrillion, so it's not as many as that. But it's 10 of those per second. So all of the information that we can take in about our sensory environment, whether it is something that we're reading, light, dark, things that we touch, things that we smell, all that's being fed in, we can calculate and process that much information every single second.
It's pretty astounding. And in fact, in addition to that, when we take things in, we store it. And scientists don't have this nailed down completely, but estimates currently are somewhere between 1 and 1,000 terabytes of storage can be managed in every single brain. Now that's huge. Just to give you some context there, the Library of Congress is about 14 terabytes.
So if we have somewhere even middle of that range, we can all store the Library of Congress and then some in each of our brains. It's pretty wild when you think about it. So all of this is done by the hippocampus, the frontal cortex of our brains. It takes in all that sensory input, all that data, it sorts it out, figures out where it needs to go, what's important, what's not important, and it kind of files it away as either stuff we can forget immediately, maybe filtering out most of it, in fact, that we're bringing in. Others it says, well, I might need this for a short time.
Others it stores for long term. In fact, as one example, when I travel, I can remember the parking space number at the airport for that week or two that I'm gone, as well as my hotel room number, usually. And I can remember it just for that duration, but a few days later, that's gone, maybe even that very night, after I get back, that number is gone and my brain just kind of discards it. It's not something I consciously do, it's just sort of something that my brain has grown accustomed to because I need that information for that time.
So it sort of begs the question, then: how much information does our brain need to keep? Do we need to keep every quadrillion calculations per second, or 10 quadrillion per second? Obviously not; that would just overload us. So if you think about it, how do we sort out what's important, what's not important? Well, it basically boils down to what do we need every day to thrive and to survive?
That's rather simplistic, but those are the two key elements that I thought are useful for anything that we need to keep. So by thrive and survive, I mean, of course, surviving means we need to know how to eat and where to get food. We need to know about getting shelter, how to survive in the elements, but also to thrive, we need to recognize our family members, remember our name, our address, where we live, how we work, what we do, phone numbers, et cetera, all kinds of information that we store because we use it all the time. Now, there's lots of stuff that we filter out. As you watch television, most of that information is flowing past your brain.
You don't need to keep it because it's not going to affect you, but oh, hey, if they have an advertisement about that car that you're thinking about, you might actually store that for a bit because you're comparing that to some other vehicle and so you might keep that at least for a short-term period. So we just do that automatically, sorting and filtering that information.
So how does that apply to business?
Well, in the same way that we think about human memory, let's talk about institutional memory, sort of making the brain mimic how, I should say, making business mimic how the brain works. And I'm going to illustrate this by a data map that we created here. This is one of our consulting projects that we did recently for a company where we basically did a full data inventory and we tracked four different elements that you can actually see on screen there. Let me actually get a highlighter so you can see what I'm talking about. And basically, here on the top row, what you see are the inputs, the generators of information.
And some of these have been sort of coalesced into groups like this one here at the top is just general employees. These might be different systems or different types of inputs from external sources. But essentially, these are all the inputs coming in. Just like we have sensory inputs in our brain, these are the inputs from a business. In the second row, these red dots here, those indicate applications that are processing the information.
So just like the frontal cortex and the hippocampus process information, we have applications that are taking that data and doing something with it, whether it is reformatting it, calculating it, et cetera, all kinds of different applications that do that. And of course, once we process it, we have to store it. And that's what's represented by this third row of information here, these different yellow diagrams or triangles. Those are the different repositories where information might be stored. So those could be shared drives, or those could be databases, et cetera.
And then lastly, we've got the really long-term storage. These are backups here, these purple ones at the bottom. And so this is a good way to visualize the inputs, the processing, the storage, and the long-term keeping of information that we process as businesses. And even without knowing what each of these items are, you could probably look at this data map and tell me what information is probably likely most important. And I'm going to emphasize this by just highlighting this quote down here about how the brain works.
The more signals sent between two neurons, the stronger the connection grows. That's how we build memory. So our brains, as we walk those paths more frequently, that will mean that we store them longer term in terms of our memory. So just by looking at where we've got clusters of information here, or maybe here in the middle where we've got lots of intersecting lines, lots of things grouped together, you could pretty well summarize the information here by saying we've got lots of important information, especially here in the middle of this chart. Obviously, we've got important stuff here where it's being backed up.
And that's without knowing the specifics about each data element. Now, that's not to say that something way out here at the edge is not important. Maybe that's really vital information. We just have fewer systems involved in processing it, and that's where you'd have to get into the specific details. But generally, we can see how that tells us where the important information is and what we're doing with it.
So that's a good visualization, but obviously we've got to get more into the details to kind of figure out how we actually do that in a business or a government environment.
So let's talk about some strategies that can help us to actually do this and to use that same idea of memory applied at an institutional level. So before we dive into that, a couple of things that I'm going to just emphasize in terms of today's presentation. One is that in terms of the strategies I'm talking about, I'll be focusing the most on day-forward strategies. In other words, if your ship is leaking, you've got water, you need to stop the leak first, and then you can bail the boat. So we want to make sure that we are putting some governance around the information that we're bringing in, putting some rules around it, some structure, and then we'll figure out how to go back and actually clean up the stuff that was there before.
There's been some other really good webinars here at MER about things like how to do data share or file share cleanups, things like that. So I won't be tackling that in detail here in this particular webinar. And then secondly, I want to talk about terms of long and short-term memory. So in terms of short-term, I'm talking about that's basically where the information that is being generated is going to live, and it's usually in the system that created it or where it was received. In other words, we've got these inputs, we've got these processors, and we're saying that short-term is generally going to stay in those systems, and it doesn't need to be so much managed as it just needs to be forgotten in a timely manner.
So when I talk about short-term, that's the context in which I'm referring to. And then long-term is where you're actually going to purposely put it somewhere or keep it to manage it over a long-term retention schedule, hopefully. If you don't have a retention schedule in place, that's always a good place to start, and I'll be talking about some of those starting points in just a moment. So that's what we're talking about in terms of some of the foundational concepts going forward.
So let's actually now get into the nuts and bolts of what type of inputs we're talking about and some ways that we can actually tackle them. So to begin with, let's think about in your own organizations the types of inputs that you have, our operational inputs, the different types of, in this case, electronic records. So to do that, I'm going to list a few of them here, but I want you to be thinking too about some additional ones, and you'll have an opportunity here in just a moment to type those into the chat window if you want to add to that because all of us deal in different types of information. So top one that always comes to mind is email because that's something that you deal with on a daily basis, actually on probably a minute-by-minute basis. I know my inbox is already filling up as I speak, yours probably as well.
And of course, unstructured documents, and I'm referring to things on your shared drives, maybe things on your desktop, Word documents, PDFs, spreadsheets, things that are not organized into an application. They're just different documents that have data structured in various ways, each kind of encapsulated by itself. Those are probably a big issue for everyone. Instant messaging, you probably have something in your own organization, some sort of platform where you can send messages back and forth, and you need to, in some cases, capture that. Collaboration platforms, that can be SharePoint, it can be anything where you might be sharing information that's not necessarily so much business-related, but maybe it's kind of just sharing ideas, concepts, and so forth.
It can even be social media platforms, that can be another collaboration point. And of course, structured data, that represents the various applications that we use. Maybe they are ECM systems, ERP, accounting systems, HR. We have a number of different applications that we use, anywhere from maybe a few dozen to, in some cases, several thousand applications if you're a large worldwide organization. I've seen organizations that have 2,500 or 3,000 applications in use, and so you might have all kinds of things that you're managing there, and the underlying databases that typically go with those.
Of course, websites: we've all got them. We've got information that is posted there, so that's going to be very important, and that's not necessarily an input, but it is something that we have to manage. Something that's not always thought of, but we have video and audio recordings. Those could be surveillance videos, they could be webinars like this, or they could be audio recordings such as your voicemail, it could be recordings of meetings, all kinds of things there as well. And also logs and audit trails. If you're in IT, you know that logs and audit trails are being generated by the, I don't even know what number to apply to that, but they're happening just all the time, everywhere, even just basic devices now like cameras and so forth have to have logs so you can go back and trace what's happening.
So those are all behind the scenes, we don't see them a lot, but it's just phenomenal how how much data is being generated in logs alone. Anyone who works with Microsoft Exchange, for example, at the server level, there are entire volumes of storage devoted just to logs, and there are all different categories and types, and it's a process just to manage that alone.
This is going to be another poll, because I want to get your input here regarding what types of short-term information might you be able to forget. So thinking in terms of memory, those things that just come in and we can just keep them short-term and let them go. Of the five that I've posted there, and this is just sort of a short list of what I had put on the prior slide, of those items there, which do you think you could simply forget as institutional memory that you don't really need? And of course, each of these could be shorter or longer. Some of these can be forgotten immediately, others might need to be kept for maybe a few days or maybe even a month or so, but we're talking about things that can be forgotten maybe within a year, for example.
So it doesn't mean that they go away instantly. So the top category of things we could get rid of is instant messaging. Looks like other ones were sort of close ties. Looks like the next one, the logs and audit trails. Some said that we could get rid of unstructured documents, and obviously that's going to probably depend on what's in those documents, depending on the content.
Collaboration platforms, a few of you said those as well. But I would agree, instant messaging is probably one of those things that we can most likely get rid of in a short time, and of course, there could be some exceptions to those. If you have decisions being made within that context, then obviously we need to save those off, but for the most part, that is temporary information. That's great. Appreciate your feedback on that.
So now get ready for another poll. We're going to do the same kind of thing with long-term information.
So which of these should you remember for a long time? And of course, long time is relative, but we'll find out which ones you think are most likely to be kept for long-term based on your retention schedule. [poll] All right, so the ones that we should remember the most: the most respondents picked the structured data, makes sense. Looks like the second was email, followed by the video and audio recordings, and collaboration platforms, and lastly instant messaging. Great.
Yeah, you're all on track with what I would say as well. I could have obviously thrown in the unstructured documents in there, I only had five to choose from, so I just picked a few interesting ones to see how that would come out. Great. Well, thanks so much for your participation, and that helps to see if we are in alignment as far as how that process works.
All right, so let's actually dig into some processes for how we can actually figure that out for each of those different types of systems. So as we begin to think about the stuff that we're dealing with, I'm going to give you some tools that you might be able to use to sort of mimic that brain process that the hippocampus does. We'll just pretend to be a hippocampus at this point. And I've coined this the SLIM process. It's just kind of easier for me to remember if it's an acronym. It just stands for short-term, long-term, in place, and move, meaning that whenever I look at a system and apply a process to it, I'm going to be thinking in terms of what is short-term, what's long-term, do I manage it in place, do I move it somewhere else.
So it's kind of handy to have an acronym. It kind of works for a presentation, so hopefully that will be helpful for you as well.
So let's start with a little diagram to help you see how this information flows and how you might apply it to your records. So to begin with, we're going to say, if we're looking at some system, and I call this System X, as you can see in the title slide there: is the system creating or storing information that is short-term only? Let's take that instant messaging, for example. If we're dealing with instant messaging, for the most part, we can say that it's only short-term information. So if that's a yes, then we can say, great, well, all we need to do then is forget that information after a short period of time.
Now how you forget it will depend on the system that you have. A lot of different instant messaging apps are now providing tools to forget information after a period of time. You might have to, in some cases, have IT involvement to purge that information. But in this case, it's kind of the simplest scenario, because all you're doing is saying we need to just get rid of that information after a specified period of time, usually as short as possible, saving out, of course, what might be the exception. So that's the easy part.
So let's say that it's not just a straightforward list of short-term information. So we'll say no, it's not short-term. So the next question we have to ask is, is the platform or the system that we're talking about, does it have retention capabilities? So if it does not, or I should say, if it does, can it store that information long-term? Actually, the better question is, is it a good place to store it long-term?
So let's say it does have the ability to tag it for one, two, three years, 20 years, whatever the case might be, but you have to ask, is it a great place to store it long-term? And by that, I mean, can you do things with it, such as searching? Can you apply holds? Can you apply destructions and so forth? Is it a good place to manage that in a long-term basis?
Well, if so, great. Then you can just apply your retention schedule policies to that, store it long-term. Maybe some things get stored short-term, and you can manage it all in that system. That's great. So maybe that refers to, say, an ECM system where they have applied or provided tools for the ability to manage those retention periods, so that's great.
You can manage it all in that system. So, sort of two forks on the road here: short-term only, or not short-term but you have those retention capabilities.
Now, of course, the trickier part gets in when the questions are no to those answers. So for example, if I say that either it does not have retention capabilities, or it's not a good place to store things long-term, then I've got to start doing some more work. So that work would first involve how do we manage the short-term stuff. We would hopefully want to manage that in place. In other words, there's no sense moving it out of the system if we can just manage it in place, and that will depend in part on what tools we have available in the system.
So for example, in this first question, does it have retention capabilities? If we say no, then we might have to manage it short-term by some API, an application programming interface, with a system that does have those capabilities of doing retention. If it's the case where it does have retention capabilities but it's not good to manage it long-term, then we're simply using the built-in capabilities to do that short-term management. So the next part of that, of course, is what about the rest of it. So the long-term information: the best practice is going to move that into an inactive repository where we can actually do the long-term management.
Now that moving process could be manual, it could be something like a click and drag, it could be an export by generating reports, there's a number of different ways that might happen, or it could be automated. So you might have automatic processes you can apply either through external systems or there might be automated tools within the system to automatically export or work into the systems. Either way, you hopefully will find automatic ways to do that. But the goal is to move it into a place where you can search and retrieve, apply holds, go through disposition processes, have audit trails and so forth. Generally, a good repository for long-term storage, where you can apply all those tools.
In fact, this whole bottom section here relies on the underlying premise of having good records and information management tools that you can apply and you can use to apply retention, maybe applying retention in other repositories for that in-place management that we talked about a moment ago here. But also notice that the other part would actually deal with this prior piece we mentioned here about applying retention policies in place. Hopefully you can have a feed from a records management system to actually apply those, so you can manage them in one system and apply those to another repository. So that's the diagram as it is, that you can hopefully apply to any system. It's pretty simple when you think about it.
We're basically just trying to make those decisions about: is it short-term, is it long-term, do we manage in place, do we move it elsewhere?
But the goal is to actually apply this to multiple systems. So let's actually try that. Instead of just being theoretical, let's actually take a look at how we might work that through in some actual systems that we've talked about already. For example, if we take email as the first one, since everybody has a struggle with that, let's ask that first question. Is it only for short-term information?
So if you think it is, then go ahead and raise your hand and we'll see what the response is there. And of course, this may be sort of relative. So if you say generally it's only short-term information, that's fine as well. So several of you are saying that that is just short-term information. That means there's a bunch of you that say it's not short-term information as well.
And in my thoughts, it was kind of more on the no side. It's sort of a relative question. As far as it being only short-term information, no, you're going to have probably a mixture. So you've got some that are not important, but you've got several things in emails that become important, whether it's because they have attachments that are important or if there are decisions being made in there. So a lot of people need to manage that long-term.
So if we're saying that it's not just short-term, we can't just stop there and say it's all managed within the email system, we have to go on to the next question. That is, does it have retention capabilities? So that may depend on your systems. If you're using Outlook or Office 365, one of those kinds of platforms, we have to think about whether that has retention capabilities, or using Google as your email platform, that has different capabilities. So you might have other systems as well.
So be interested in seeing the hands raised there. But I would say that generally speaking, most email systems today do have some capabilities in terms of retention. So we did get some hands raised, looks like about 15 or so, and obviously that depends on your platform. So it's not just short-term, but it does have some retention capabilities. So, next question: is it a good place for long-term management?
If you think so, go ahead and raise your hand there. In other words, can you use, let's say, your Outlook or Office 365 for a place to search, to apply holds, to do destructions, all those good records management capabilities? So, about 20, again, raised their hands on that. So about 20 have said that it is a good place for long-term management. So maybe you've got some good tools in place, that's great.
But I'd say based on the number of attendees we have, the majority of you are saying that it is not a good place for long-term management. And again, it depends on the system. But I would say for the most part, when I talk to my customers and prospects, they find that it's not a place I want to keep it long-term because it's hard to search. It doesn't have all of the different processes for managing holds and destructions and so forth that they'd like, or at least not to the extent that they'd like. So we've got to come up with a strategy.
So a couple of things that I found that seem to resonate well with the folks that I work with, one is that in many cases, it's best to try and forget most of the information after a short period. Now you can define that short period. I find that somewhere in the 90 to 180 days, or three to six months, is a good timeframe, meaning that most of the stuff coming into your inbox can simply be purged after that time. And then, of course, you want to save the rest of that to some other long-term repository, either automatically or manual, so that it's kept for the right period of time based on its content, based on what's actually in that email. And then you might also have an email vault to keep it for a longer term.
So some of the times I see that might be two or three years, some as long as five years, just that sort of just-in-case place to put things in case people haven't saved them properly, but not a place to keep them for 10, 20 years, something like that, and not according to a retention policy, just sort of a safety net. So that's a strategy that some companies do put in place and seems to work quite well, and it kind of matches the process that we're talking about.
So what about shared drives? If we take the same kind of strategy there, what is, or, is it only short-term information? If you think so, go ahead and raise your hand. All right, so in the response, we've got about 10 people saying that it's only short-term information, so the majority are saying, "Nope, it's not just short-term." I would agree with you on that.
Typically you want to keep most of that stuff, at least you want to figure out what's short-term and long-term, but it's not going to be only short-term. So does it have retention capabilities? We're talking typical shared drives, or maybe even cloud drives, things like that, so some sort of a repository where you've got folders to put things in. And I would say that most of you are going to say, "No, it generally does not." And again, it depends on the platform. There's different tools coming out, but for the most part we're saying that it probably doesn't have good retention capabilities, so we've got to figure out what we're going to do with it, so we have to have a strategy.
So one strategy that I see that works pretty well is to first divide the folders into active and inactive. Now you can do that on a global basis where you kind of just have one great big bucket for active and another great big bucket for inactive, and you've got substructures within that, or it might be within each department or each group that you subdivide that into active and inactive stages, but it's good to separate the two so you can easily handle those. And then the stuff that is short-term, you might forget them or purge them, maybe after a year or some time frame that's going to be short-term. And for a typical shared drive, IT can do this. They've got tools to do this through PowerShell scripts and so forth.
But you do have to be careful. Like in Windows, if you want to use something like the last access date as your tool to find out what's been accessed recently, that is something that has to be enabled. That was turned off a few versions ago in Windows, and so that's not a tool that's readily available unless you turn it back on. And then, of course, for those long-term documents, you want to save them somewhere where you do have those tools, in order to save them long-term and apply your retention policies. So, pretty simple strategy, but it can be effective if you carve it out in that way and be able to have a place to put those long-term.
I'm going to run through the next of these a little bit faster. I won't ask for input on all of them. And obviously, if you have points that you want to make or questions to raise on those, feel free, and we can tackle those at the end. So let's talk about things like SharePoint. Is it only short-term information?
Well, it's complicated. You've got all kinds of things. It depends on how you use it, how it's deployed, what rules you apply to it. In some cases, it is. In some cases, it's not.
Does it have retention capabilities? Well, yes, sort of. In fact, recent versions have even enhanced some of those capabilities, so it might be a good place, depending on what strategies you deploy, what tools you have enabled, so it may or may not. Is it a good place for long-term management? Well, again, that kind of depends.
It depends on how you're using it, how you've structured it, what tools you're putting in place. So this one gets a little bit messier. It's not cut and dried. So I've come up with three options that I think can be workable for most people. You can choose either one of them.
The first one is essentially keeping it all in SharePoint and using the built-in information management tools that are there, using the tools and processes that they have in place, and then using the auto-update policies, like if you can apply the retention policies from a records management system, like we talked about in the diagram, in order to automatically update those policies so that you don't have to manage that manually or do it in multiple systems. That's going to help you out. And then use the built-in SharePoint tools to purge that information when necessary. Option two might just be saying, "Well, we're going to keep the information in SharePoint while it's active, just while we're collaborating and creating things. And then we're going to move it to long-term when it's inactive, and then purge the remaining short-term items once we have done so."
So it's basically using it only as the collaboration piece, only as active storage, moving it elsewhere when it's done. And then finally, option three: using a third-party tool that kind of builds on top of SharePoint. There's lots of good tools out there that you can use to manage it in place, but not using the built-in policies and tools, but using a third party to apply records or policies to that.
If we talk about Office 365, that's going to be almost exactly like SharePoint, because it's still going to be not just short-term information, it's going to have a mixture. It does have some capabilities for applying retention. We have the Office 365 labels that have retention policies that you can apply. Is it a good place for long-term management? Not the best.
It's getting better, again, with recent things that have been applied. So we can essentially take that same concept of either managing it where it sits with the built-in tools, or keeping it there only while it's active and moving it out of that repository when it's done with its active lifespan. And then option three, again, applying external tools to that, that have been built for that purpose in order to apply those policies. So, very similar to SharePoint, because they are essentially the same system, just with kind of a different focus.
So we can do the same thing with things like video recordings. Is it only short-term information? For the most part, yes, especially if you're talking about things like surveillance cameras where it's just kind of running, and you just need to have it kept for maybe 30 days or a couple of weeks, and the information can just roll off. And, of course, if you have exceptions, then you can save those out. So, in that case, it's a much simpler option.
You just set how long you need to keep it. Usually, the devices now have the ability to set how much information you're going to keep, either based on the size of the information or the days that you want to keep it, and then save off the specific incidents. For example, if you have a break-in in your building, you need to save the recording of that. Obviously, you want to pull that out and store it in some longer-term repository. Now, there are some concerns, especially as of late, with the increase in the use of body cameras, where they might have to keep those longer-term, especially when there's a case that might come up with those for litigation.
And in that context, there are issues where you might have to keep it long-term, but you can't keep the personal information, especially if you're dealing with, say, GDPR rules. You might have to actually spend quite a bit of time and effort going through and blocking out images or people's faces, things like that, in order to make sure that you're not keeping the personal information longer than you should, according to privacy rules.
And instant messages. I'm going to just kind of buzz through that one the same way. It's mostly short-term. So, again, you can forget it after a short period of time, except for maybe some exceptions, and hopefully you can have tools within the system to be able to do so.
Then the last one here, structured data. That's going to be a little bit more complicated, and actually [inaudible] did a great treatment of this a couple weeks ago. That would be good to watch on one of the MER webinars. But it's definitely not short-term information. It usually doesn't have good retention capabilities.
There are some exceptions, of course. So, there are different strategies to handle this, but most of the time we see that people will be exporting information out of them to store into the long-term repository. You can, of course, look to see if your RIM system has APIs that can reach into that information and actually manage retention in place or be able to pull out the information that is needed to be kept long-term. So, again, Greg Forrest was the one that did that presentation a short time ago here in the MER platform. So you'd certainly benefit by going back and looking at that one.
So those are some of the strategies. Hope you find those helpful. Just a couple of other points here about how to get started. These are just some of the practical things about how you might actually apply these principles and these plans to your system. And those start with just developing a plan.
So, in terms of developing a plan for how you actually put this into place, I would say that you would want to start, first of all, with planning out an information inventory. So when we saw earlier, there was a slide about that data mapping that we had done for a customer. That was an inventory that had been put into a graphical format. Yours doesn't have to be necessarily that fancy. It can be simply a spreadsheet with a listing of all of your different systems and those inputs and repositories, but basically getting an idea of what you have, so you can start working through the rest of that that follows from it.
Then, of course, you hopefully will have a records management, or I should say, a records retention policy because that's going to directly relate to that. In fact, you would want to map that retention policy to the inventory that you take so you can then start to put your thoughts around how you keep the information in each of those repositories, and what you do with it, and how you assign priorities, privacy details, and so forth.
And then I would recommend taking that same process that we just went through, those different exercises, and applying that to each of those systems so that you can say, "Okay, is this just short-term? Is it long-term? Do we have retention capabilities?" Just do that methodically through each of those systems in order to see how you're going to process that information. And, of course, you want to determine the order of implementation, maybe a pilot system to start with because you can't do it all at once.
And I would highly recommend baking this process into your onboarding process that you might already have. So an example is that if you are purchasing an application and you have onboarding processes to ensure that you've got storage and management and ownership and governance of the application, also make sure there's at least a checkbox or something that says, "Has anyone thought about how long this information is going to be kept and if it's short-term or long-term? And basically, what are we going to do from the records and information management perspective?" And then finally, going back to that idea of bailing the boat, you need to probably spend some time doing file analysis, purging the redundant, obsolete, and trivial information. And again, there's a lot of good tools out there that can do that, basically taking a look at the piles of things you have in terms of unstructured documents, sorting through and figuring out what can be gotten rid of, what is important.
And then, now that you know sort of this mapping of your inventory and the application of your retention policy, you can actually take the results of that and categorize those into the different buckets that they should belong to. And now you'd have the retention and the policies applied to them to know what to actually do with them.
It would require some potential tools, so you might need an inventory mapping database. It can be as simple as a spreadsheet. There are actually commercial tools for doing this as well. You might need a file analysis tool that can go out and crawl your systems and sort out what you've got. Obviously, having a records retention schedule system would be ideal so that you can manage the retention schedule in an organized fashion, apply the rules, keep up to date with legal requirements.
And of course, having a repository: all this is sort of premised on the ability to put the information someplace long-term if it needs to be following that retention policy and be able to manage it for all those RIM functions.
And this is the part that nobody likes to talk about, but I said that this is a practical strategy, not necessarily an easy one, but you do have to get buy-in, so you'll have to do some cost analysis, figure out what this is going to be costing you long-term. And that would follow with getting executive sponsor. They're going to want to know how much it's going to cost. You're going to need some executive buy-in to make sure it's got backing from management. You'll need to work with IT, to collaborate with them on these systems, and in fact, IT can often supply the starting point for your information inventory because they've got a list of systems that they have to back up, so they're a good person to have in your team.
And of course, you've got to talk to legal and procurement if you're going to be getting any of those tools.
One last option here, and that is putting it into action. I got some questions on previous presentations of this topic that talked about, well, how would you actually start this if it were you? So what I would do is determine my top 10 systems, the ones that pose the most risk or the most problem, and I would interview the subject matter experts for those systems and start to find out what they keep and where and how much is there, how long they're keeping it, and so forth, and then follow that decision tree, that SLIM process, for each of the items in that spreadsheet. And from that, develop a brief policy for each of those systems. It could just be a one-page Word document for each one saying, "Okay, here's the type of information we have.
It doesn't have retention capabilities. We're going to move it here when it's met its active life cycle, et cetera." Just keep it brief and short and use that to sort of build your policy. And then, of course, get approval on those policies, and you can choose the easiest system to start with and dive into the implementation. Now, I did have one person say they want to start with the hardest one.
I guess it follows that old adage: if you have to swallow a frog, don't look at it too long. So if you want to go for the hardest one, great, be my guest. But I'm suggesting something that might just be as simple as putting together a spreadsheet where you list the system. You have those questions that we've got in our decision tree. You list the available tools, kind of summarize your decision, and maybe you add some other details as far as the subject matter expert, other properties that may apply regarding privacy, et cetera, anything that would be helpful to note on here, but just keeping it simple as far as the details that you have categorized for each of those systems.
So it doesn't have to be fancy, but at least it gives you a place to start, and you can actually maybe take this as a practical way to guide you through that process.
So as far as questions and answers...
Chelsie McGuinness: Looks like we are limited on time, and we'll only be able to get to one question today. Please note, if you have submitted a question and we do not get to it, we will be certain to follow up via email afterwards and get your questions answered. [audience question] Kathy asked, "What would you recommend for large Microsoft Outlook files?"
Warren Bean: Oh, yeah, the dreaded OST files. Those can be, for certain, problematic. In most cases, when I talk to customers, they are trying to get rid of the use of OST files. I know that's not always possible, and if you already have them out there, then of course you've got to do something with them. Part of the answer to that, I think, will depend on how you use them, like are you frequently having to reference them to go back and pull out emails.
But for the most part, I would say that those are going to be fairly long term. In other words, if you are saving those files now, you've probably saved them for a reason because maybe the person that has them or had them had important information, so you're most likely going to be saving them. So, you're going to have to have a repository to save them in and determine the retention period, of course, on those. I'm not sure there's a cut-and-dried answer, because it depends, again, on those questions about how you're using it and the purposes for saving them. But there are certainly repositories that can hold OST files, and so you can apply retention to them.
But I would say, for the most part, try to move away from that process, if at all possible.
Chelsie McGuinness: Great. Thanks, Warren. And with that, I want to thank you all for coming. And if you are interested in scheduling a 30-minute consultation with Warren Bean, specific to your company's industry and jurisdictions, or if you'd like to schedule a one-on-one meeting during ARMA, feel free to reach out to me at my email, chelsie.mcguinness@zasio.com. Also, if you're attending ARMA and want to see Warren present along with Soo Kang, Zasio's director of consulting, be sure to join us on Sunday, October 20th, for our Z-Talks pre-conference education sessions, which will be located at the Opryland Resort and Conference Center, where ARMA InfoCon is taking place.
I'll be sure to send out the link for you to register in a follow-up email to you as well. However, don't worry if you will not be attending ARMA. Please just let me know, and I'll be sure to send you a copy of the recorded sessions. Again, thank you all for attending the webinar today, and I hope you have a great rest of your day. Thanks, everyone.
Questions answered in this webinar
Which types of electronic information can usually be forgotten after a short time?
Instant messaging was the audience's top pick, and Warren agreed it is mostly temporary information that can be purged after a short period. The exception is when decisions are made in a conversation, which then need to be saved. Answered by Warren Bean.
What is a practical strategy for email, given most email systems aren't a good place for long-term management?
Purge most email after a short period, often 90 to 180 days, and save what matters to a long-term repository based on its content. Some organizations add an email vault for a few years as a just-in-case safety net, not as the place to apply the retention policy. Answered by Warren Bean.
If you were starting this yourself, how would you begin?
Pick the top 10 systems that pose the most risk or problems, interview their subject matter experts, and run each through the SLIM decision tree. Write a brief one-page policy per system, get approval, and start implementing with the easiest system. Answered by Warren Bean.
What would you recommend for large Microsoft Outlook files?
Try to move away from using OST files if at all possible. Existing ones were probably saved for a reason and are likely long-term, so put them in a repository that can hold them and apply a retention period based on how and why they are used. Answered by Warren Bean.
