How the IRS is taking digital government to a new level

Best listening experience is on Chrome, Firefox or Safari. Subscribe to Federal Drive’s daily audio interviews on Apple Podcasts or PodcastOne.

In some ways, the IRS is defined by its some forms. Many of the 800 forms represent a process or transaction. People in the agency’s Enterprise Digitalization Project have been exploring quite a few projects that go far beyond simply digitizing forms, to modernizing business processes and case management. For an update, project director Harrison Smith spoke to the Federal Drive with Tom Temin

Interview transcript: 

Tom Temin: Mr. Smith, good to have you on.

Harrison Smith: Good morning, Tom, thanks so much for having me. I appreciate it.

Tom Temin: This office has existed now for, what, a couple of years, and you’ve made some pretty good progress. If you would, just give us the quick background on the purpose of the whole project.

Harrison Smith: Appreciate it, Tom. Yeah, we joke that we’re almost a teenager because we’ve been around for I think it’s about 15 or 16 months. I mean, the general purpose of our office is to spearhead the IRS efforts to empower taxpayers, and IRS employees to rapidly resolve issues in a simplified digital environment. And that obviously crosses quite a few efforts, as well as issues across the IRS. What we really focus on is helping stitch together efforts across the service to create a more holistic and enterprise-wide perspective. And a perfect example of that is to look at not only improving, you know, technology activity, right, because a lot of folks talk about modernization and improvement in that space. And there’s certainly opportunities there, but that’s the prerogative in the authority of the Chief Information Officer Nancy Sieger. We frequently work with Nancy and her team, but we look at improving business processes and policy positions in order to optimize both the processes that exist without technology, but also to optimize our existing and new technology and investments. So we want to make sure as we are investing or continuing to work with technology, that we’ve got the business processes and the policies lined up in such a way where we’re making the best use of that effort that’s driven and governed by the chief information officer. The other piece that’s really important is, as you noted in the lead, the forms are frequently a focal point for the IRS. We have, I think the number is 1439, and that’s certainly a large number, but it really does focus on how we engage with the taxpayer and the public. And making sure that that we optimize our processes. So that is a straightforward and frankly simple solution and an experience that it can be. We want to make it easy, as easy as possible, for folks to engage with the IRS to provide the documentation that they need to provide, in order to understand and fulfill their tax obligations.

Tom Temin: And digital service, digital transformation, that’s a word bandied about a lot these days in the government, your project has an extra syllable — digitalization. “Al” is in there. And so I’ve heard you speak before, and there’s a definitive meaning to that word digitalization, as opposed to just digitization. And to tell us that distinction.

Harrison Smith: Yeah, Tom, sometimes we struggle with it ourselves. We sometimes just go with Team Digi to explain our office. But no, so I’ll start with digitization, because that’s the first piece. Digitization, in the way that we use, is simply the process of making a paper item digital. And the example, I’ll tell a quick story about my mother in a moment to highlight the differences between digitization, digitalization and digital transformation. So digitization is simply making a paper item, or perhaps recording into a digital item that can be moved digitally as opposed to physically. Digitalization is, the substantive difference between that and digitization is machine-readable data. Rght, so if you have something where you’ve got, you know, you scanned a piece of paper on a copier or multifunctional device, you have an image, but you can’t search for things, right? It’s not as though it is a native PDF. You can’t necessarily search for a number or name or things like that. So that’s the substantive difference between digitalization and digitization as we use it. Digitalization involves machine-readable data, and being able to search and utilize the information that’s contained within that image. Now, digital transformation is some slightly different, this doesn’t necessarily have to do with machine readable data, or moving images around. It has to do with optimizing technology in order to improve the taxpayer experience and our functioning. So as we talked about a little bit ago, you know, forms are obviously and cases are something that the IRS focuses on. But part of the digital transformation may simply be focused on, in an appropriate and good way, may simply be focused on improving the taxpayer experience in a way that does not generate a form, in a way that does not generate a case. And one of the things that we’ve worked on fairly recently is trying to expand the aperture and expand our understanding of what might be possible for a technology perspective in digital transformation, by using mobile devices in order to look at forms are publicly available information that the IRS publishes in a way that’s more beneficial or more easy to use for the taxpayer. So again, that digitization is just that piece of paper, which is now which is now digital and can be sent via email or something like that. Digitalization is that machine-readable data. And digital transformation is, again, using this business and technology processes to improve the taxpayer experience. And I noted that I would tell a story, a quick story, about my motheer, and she knows that I tell the story about her and we joke about it. But my mother at one point took a picture of an email with her phone and then texted it to me. And that’s a perfect example of digitization. I now have a picture. It’s not as though she printed it out and mailed it to me. And yes, she has done that before — my dad likes to send a cut-out newspaper articles and send them to me. But digitization was that was her activity of taking a picture of an email and sending it to me. And she said — it was fairly long email honestly — and she said, well do you see the part where she says, you know where your sister says this word? I was like, Mom, I can’t find that I have to look through the entire email, because it’s a picture of the email. If you’d sent me the email, that would be a digitalized item, because I could search for a word or I could search for a number. And ultimately, digital transformation is would be the equivalent of FaceTime or something like that. Because it doesn’t necessarily rely on an image or set of information. It, again, relies on an activity or a business process that helps improve how we function and helps improve the taxpayer experience.

Tom Temin: Well, if your mom ever gets audited, she’ll probably end up sending the auditors away screaming by making things so complicated. They’ll never be able to figure it out. We’re speaking with Harrison Smith, project director for enterprise digitalization management at the IRS. Tell us about some of the projects you’ve done so far in the 16 months. Did you start with those that are really hard to digitalize? Or did you start with one that many, many, many people do with the IRS to try to get a big win quickly? And you mentioned looking at a forum with your camera, and maybe that’s a good place to start?

Harrison Smith: Yeah, and I think that’s an example of a transformative project, which is something that is subsequently different than how we function. And what we do, to answer your question directly Tom, is we look at those transformative projects, we look at what we call adjacent projects, which are kind of like midterm efforts, and we look at incremental projects. So we do have, again, the project around utilizing augmented reality, and similar types of solutions with your phone to help gather and provide publicly available information to a taxpayer when they point the phone at an image or a document or a form. That’s certainly a transformative activity. It may frankly be time that that’s not something that works out or something that we’re ready to to invest in right now. But again it helps us understand what might actually be possible. Adjacent activities, which is, again, sort of the midterm area that we work on, are items where we use or expand on existing processes, or change existing processes so that they’re more effective, or more helpful to other business lines within the office. An example of that would be how we utilize a fairly mature technology, such as eFax, for forms that have previously been submitted only manually. By doing that, by again using that adjacent technology in that adjacent perspective, we were able to accelerate throughput, for more than 30,000 forms that the IRS receives on an annual basis, we’re able to accelerate throughput by 81%. And that was, again something that was an existing technology and existing business process, but we expanded it and tweaked it in a way where it could be used for other items. And then there’s the incremental piece. And the incremental is really the near-term items that we’re able to look at in hopefully certainly less than than nine months, hopefully less than six months, and maybe even closer to three months. And those could be things like policy decisions that are fairly straightforward that the IRS has a lot of latitude on, or it could be something a little bit more robust, like what we tackled for retention periods for one of our forms. We recently reduced the retention period for a form from 75 years to 40 years. That seems like a fairly mundane activity, but it will enable us to save somewhere in the neighborhood of $19 million over the next 30 or so years. And that’s just just one of the forms that we retain. And again, as I noted before, we do have 1439, and so making sure that we look at forms, making sure that we look at business processes, making sure that we look at things like policies around signatures and wet signatures, and also there’s more transformative pieces. We look at all three of those areas — that incremental, that adjacent, that transformative — across our portfolio of projects.

Tom Temin: And the IRS, of course, has been struggling largely with modernization for probably 25 or 30 years now, different projects, different offices and so forth. But one of the enduring issues for government is they often chase projects long after it’s clear maybe they’re not worth chasing. And so good money goes after bad. You’ve developed a much quicker way of evaluating whether to continue investing in a project. And I think that’s useful for everybody in government to hear.

Harrison Smith: I think the piece here that we sometimes as individuals — or whether or not we’re in the government, or another type of bureaucracy, or just in our home life — that we struggle with is a sunk cost bias, right? We say hey, we’ve worked really hard on this, we’ve got a lot of time built into this, what’s another six months, what’s another $6, what’s another $60,000, whatever the metric is. And what I get especially grateful for honestly, especially during during the holiday season, is the support of senior leadership, not only from the commissioner and the deputy commissioner standpoint, but across the leaders within the IRS. It’s a support of, really from our perspective, this may not be the right time to do this. And that’s not a bad thing. And so we’ve looked at projects, which we’ve set aside for six months to say, hey, this is not the optimal time to do this. That doesn’t mean it’s a bad idea. But we need to set it aside. Or, hey, this is this is a fantastic idea, there’s a great return on investment in this space, but it’s not something that we’re going to realize for the next 10 years. So we’re not going to work on that right now, because we have other opportunities that are more near term and more immediate that we can help address. And it’s that slightly different mindset. And it doesn’t work everywhere, Tom, like it just doesn’t. But in areas where that ability to change things and really look at an opportunity, from a perspective of what is our return on investment, is it something that we can realize, and reinvest and learn from? That’s the way we look at things. And if we have to cancel or shut down a project early, and we’ve learned something that’s in no way a failure, I think that’s where we get into struggle, and where we get into the largest problems, are the projects that we have to wait three or four or five or six years to find out whether or not they work, and we’ve invested a lot of time and money and effort at that point. It is harder and more difficult to look at a place where you shut something down. We have near-term projects that we can say, this just isn’t working out as we expected. Or we didn’t realize that there was a requirement for us to do things this way. We have to continue to do it in the same exact fashion that has been done it. And that’s okay. And if we can make those decisions in, you know, six months or six weeks or six days, that’s a win. That’s not a failure; that’s actually a win. And so that’s the cadence and the perspective that we bring.

Tom Temin: So it’s almost an iterative, agile approach to project management and even contracting that goes along with the iterative, agile approach to development itself.

Harrison Smith: 1,000%. And I think you can look at the work that we’ve done with our partners, Shanna Weber and Guy Torres at the Chief Procurement Office. We set up contracts where we went out with a very, very broad perspective. We want to be able to solve this particular problem. There are some guardrails involved, right? This is how we would like to do it, this is the approach, these are the timeframes — but we said upfront, hey, we may award three or four or five contracts. They’ll be around for anywhere between 30 days and six months. But we don’t plan on awarding options or providing future funding to everyone. We anticipate this being an iterative process, where we learn and grow and really focus our investments in particular areas. And I think it’s also really important to note that the projects that we tackle are ones that we are able to look at something very specific in the near term, like a very specific use case, or a very specific problem, or sometimes a very specific form. And if that project is successful, then can we expand it and scale it to other places? And those are the ones that are particularly interesting, and shows such a great opportunity for enhancement and scaling, are the ones where we can test something in the near term in a very specific way, with constraints and tactical goals. But then going forward, we can utilize other forms or other processes or other offices. Those are the ones that are particularly interesting to us and really show a huge benefit and opportunity for growth.

Tom Temin: And finally, you’re expecting some output from contracts on the streets now, very shortly, and then there’ll be kind of a winnowing down of contractors to proceed with some of those projects.

Harrison Smith: Yes, those are those are established as part of the option period. So the contractors and industry partners, as we like to call them, do know when they are coming. But we do, we are looking forward to identifying specific outcomes and specific possibilities. And again, if they don’t, if it’s not something that’s going to work right now, that’s okay. We may not award any additional funding. What we’re looking at, in particular for optical character recognition pilot, is are we able to extract machine readable data out of low resolution images? And we define low resolution at 120 dots per inch. And if we’re able to do that with a publicly available form that we’ve provided the industry partners, then excellent. What’s the throughput, what’s the return on investment and then can we start looking at other forms where there are other opportunities to get access to more machine-readable data for more information? And having that capacity and that understanding of what actually has worked with a specific use case, and then applying it to other use cases, that’s the area that we’re really excited for. In some cases, we’ve been able to identify a 25x increase over our current throughput, or our ability to to analyze information. And that’s something that we’re really excited about testing and making sure actually works in our environment before we expand into other projects and other forms.

Tom Temin: Harrison Smith is project director for enterprise digitalization management at the IRS. Thanks so much for joining me.

Harrison Smith: I appreciate the time, Tom. Have a great day.

 

Related Stories

    (AP Photo/Susan Walsh)FILE - This March 22, 2013 file photo, shows the exterior of the Internal Revenue Service building in Washington. The Internal Revenue Service is recalling about 46,000 of its employees furloughed by the government shutdown, nearly 60 percent of its workforce, to handle tax returns and pay out refunds. The employees won't be paid. (AP Photo/Susan Walsh, File)

    IRS digitalization office taking ‘bite-sized’ steps to consolidating legacy systems

    Read more
    (AP Photo/J. Scott Applewhite)FILE - In this May 7, 2020, file photo, the entrance to the Labor Department is seen near the Capitol in Washington. Unemployment fraud in the U.S. has reached dramatic levels during the pandemic: more than $63 billion has been paid out improperly through fraud or errors since March 2020. Criminals are seizing on the opportunity created by the pandemic and are making false claims using stolen information. (AP Photo/J. Scott Applewhite, File)

    Agencies embracing cloud for records digitization, virtual reality during and after pandemic

    Read more

    Through new Office of Digital Transformation, FDA putting business ahead of IT

    Read more

Comments