Personalizing Government for Every Resident

An effective smart city uses technology to make it easier for households and businesses to interact with their government. To meet this standard, a key variable is whether or not people use the technology that local government makes available. A panel at Smart Cities Connect in Denver, CO, in April 2019 brought together experts in government and the private sector to discuss how we can make an everyday interaction a positive, personalized experience, and use those interactions to promote deeper engagement between a city and its residents. Below is a transcript edited for length and clarity.

Panelists

• • Micheal Carroll, CIO, Central Ohio Transit Authority (COTA)
  • Mike Duffy, Chief Executive Officer & Founder, CityBase
  • Seyi Fabode, Founder / Product & Go-To-Market Strategy, Varuna
  • Michael Pegues, Chief Information Officer, City of Aurora, IL
  • Karen Pellegrin, UI Designer/ Web Content Developer, City & County of Denver
  • Liz Fischer, Chief Marketing Officer, CityBase (Moderator)

Moderator Liz Fischer: What do you think is the value of personalizing public sector, either technology or services especially in the context of some of the “smart city” initiatives that we are thinking about as we’re at this conference?

Mike Duffy: Especially in government and city services, everyone uses city services somewhat the same but somewhat differently. They’re in different regions, they have different interests, sometimes they need certain services and don’t need other services. And so, creating something that’s more personalized creates more of a holistic approach for that individual as far as all of the services that we need in the city.

Micheal Carroll: Personalizing data gives a better experience for all customers. I work with Central Ohio Transit and we are working very closely on customer experience. So, how easy is it to do business with us? If you don’t know how to get on a vehicle that we’re running on the streets, you’re going to be embarrassed, especially in front of 50 to 60 people sitting on a bus. We need to make it easy for you to ride the bus. We’re also looking at different modes of transportation, working with the City of Columbus on a multi-modal trip planner and a common payment system, so that is something that no one has done yet, and hence the fact I’m going grey. We’re working really hard to get that implemented in 2020.

Seyi Fabode: As a public entity or as a utility, you are now competing for the minds of your residents with companies that can give them a price that is unique to them on whatever product they are trying to buy. I probably see a different price for the same product on Amazon as Mike would see. And that level of personalization makes me feel known, and at some point, we need to experience that. Residents need to experience that from the services and products that the city provides them as well.

Moderator: What have the barriers been in the past to providing a personal level of service, when it comes to government and utilities?

Karen Pellegrin: Active Directory integration. I know that’s really specific. We were trying to put in more personalization for an application we have called pocketgov, which is a great application because you can create your account. But one of the challenges that we had in getting those accounts created properly and making sure that all of the information was speaking across the system was that Active Directory integration and single sign-on—all of that stuff to make it work properly and to make it’s secure because our security is pretty darn robust. And so breaking through and making that work with security was really tough.

Mike Pegues: Two examples pop up for the City of Aurora, once we talk about the personalization of data or actually the usage. One was our FOIA data. Up until March of last year, that whole process was manual. So people had to come into the 311 helpdesk, the customer service helpdesk, fill out a form. Or they had to go to the Clerk’s office or the Legal office, fill out a form and submit it. So we designed a process that allowed people to access a web-hosted FOIA system where the actual citizens could go in and submit their FOIA request, automating the process. I know we talked about data but the service that the data provides, that customer experience will actually grow on with regards to that.

Being able to enhance that customer experience for our citizens was great. To accomplish that, we also had to keep in mind security — being able to pull the data based on retention periods. There are so many different operational efficiencies that added to that overall customer experience.

MD: To add, the challenges we see today in personalizing government is that there’s a legacy of 30 years of IT procurement at the agency or departmental level, so it’s not uncommon to go into the city or county and see three to four dozen source systems. And those source systems are organized around unique values that are not household or business-oriented necessarily. And so correlating any two records across those datasets is difficult.

And then we add in the fact that the datasets use government-speak, which is not always in line with how constituents speak. As my grandmother used to say, “No one on the planet knows their parcel ID.”

Moderator: Karen has a focus on user experience and design. Would you talk about some of the issues that you’re working on for Denver?

KP: We’ve had a few initiatives to streamline the user experience, and a lot of it’s involving branding and making sure that everything across our digital experiences is consistent enough so that the user feels like this is an official Denver.gov City and County of Denver product. There are a lot of strange third-party products that cities purchase, and they’re from vendors, they’re branded unusually, their structure is unusual. So we went through a very concerted effort to give a consistent look and feel across the site and then also created standards for things like calls to action, to really pull things out that are service-oriented.

We’re driving towards making especially Denver.gov, but also pocketgov, very service-oriented. Our goal is to bring it all together, bring pocketgov, this application that we have and Denver.gov into a unified experience and then have that single sign-on, very personalized experience for our users and give them the choice. For those who really do want to have that personalized experience, they have the option to have that user account. And it enhances their overall ease of use with all the services that we have.

Moderator: Mike Pegues, one of the things that you’ve done as CIO is consolidated the IT functions under one roof. Would you speak to that decision and how it impacts your operations?

MP: I started as CIO in the city of Aurora back in June 2017. But before I accepted the role, I requested three things from the mayor. One, after doing my due diligence I said, “I need to be able to reorganize all the IT resources in the city under the office of the CIO.” Two, “I need to consolidate the budget and the spending, so any IT investment that is being spent on hardware or software, they come through the office of the CIO.” And the third thing was to have direct reporting to the mayor in terms of decision making.

So the mayor agreed to it. We completed the reorganization by December 31st of 2017. I would still say that we’re going through some transformational changes within the organization. Those three critical success factors were absolutely required to be successful in doing anything going forward from that.

Moderator: Michael Carroll, would you tell us a little bit about this Smart Columbus initiative that you’re working on?

MC: Central Ohio is the original smart city. We were awarded a $40 million USDOT [US Department of Transportation] grant two years ago, and as the Central Ohio Transit Authority for a transportation grant, as you can imagine we are very heavily involved. We’re working very closely with our city and our smart commerce partners to gather all the data that we have in a central repository, also known as the Columbus Operating System. It’s a source of data that can be used and replicated throughout cities in the US and across the world. We try to personalize that data and understand more about our riders. We are more than a bus company; we are a mobility integrator.

Imagine a customized experience for getting you from where you live to where you need to go. Not just to work, but to school, to a health care appointment, to different appointments — for everyone in Columbus.

We’re anticipating to grow by over a million people by 2020, so we’re not sure where those people are going to go. But if we get them onto trains and personalize their experience, have them pick up an app and type in where they want to go, it’s a multi-mixed modal experience because it’s not just one type of transit.

Moderator: Seyi has an extensive background in companies that use the network at the intersection of constituents and utilities. How do you use the large data stats to inform personal experience?

SF: Yes just a little bit of background: my prior company was on the power utility side. In this image (below), we took data from a power utility in Chicago and converted into what we felt was a more user-friendly experience.

A utility customer could come to the site, put in their account number, and it would pull their historic usage, and run it against the average. Although one of the things we always said was, “there’s no average consumer,” most entities, utilities, government tends to think about the people we serve in this average approach to delivering a service. We ended up with a product that personalized the experience for customers, once they pulled up this user-friendly bill that they could understand.

As another example of how utilities can personalize the delivery of data they are collecting: the City of Austin had a boil water warning. Everyone had to boil their water, and we had to rely on the news for the state of the system, which, considering the utility has this customer confidence report, is just mind-boggling to us. We do a lot of stuff on the operational side of utilities, but this was the conversion of; the whole of Austin has a problem, how can we as residences of Austin go to the water utilities’ website and see a real-time feed of the water quality? What the problems might be, in real-time. So it takes it away from all of us having a problem, to, “How does this impact me?” Even though it was all of Austin that had to go on the boil water warning, there were only some parts of the city that really had problems. There was a lack of information into how it was impacting me on the north side of the city, or on the south side. A lot of the user experience work you talk about can be applied to the huge data sets that cities have, that utilities have, to communicate with people.

We need to start to take it away from those huge data sets to personalize experiences that convey meaningful information, not just scary stuff. That’s supposed to be a consumer confidence report on the top there [of the image]. I’ve not met anyone who gained any confidence from looking at it.

You already collect the data in cities on the state of the water system, but what we do is we put sensors in the distribution system that capture water quality data in real-time. It’s interactive in the sense that when it exceeds a certain band, the alert will show this is the problem with inorganic contaminants, for example.

Moderator: Mike Duffy, CityBase was founded as a payments company and we touched on payments for utilities, we touched on payments for transportation, I think it would be interesting to hear you discuss how payments fit into this overall topic of both personalization and just better delivery and design services.

MD: Liz, much like you, I’m sure the whole audience wants to hear about payment. This industry was once a soulless pursuit of transaction fees, but it’s found itself playing an interesting role in an aggregation of services around a household or business persona. We find that the informational search is the most common and that right behind that is payment, and payment is interesting in that it is a compulsory interaction that crosses the entire demographic that is by its nature intimate, and that the resident or the constituent has to identify themselves in order to complete that.

We, as an organization, always aspire to incent people through value into doing something; opting into an account is a good example of that. So when you think about how we get as many households as we can to authenticate or create an account, how about asking them to create a password every time we touch them for a financial transaction because we already have enough information, that’s the only item we would need to finish that process? Or what if we had a program or service that was specific to that household. Let’s say they’re paying their water bill, we’re familiar with their financial status or qualifications for a program like the Low Income Home Energy Assistance Program (LIHEAP). This is free money for utility customers who could not pay a bill. In the state of Indiana, only 18% of eligible households are opting into free money.

We touch five times that number of people who are eligible through enforcement actions and through payment actions in a 12-month period, yet we’re not getting them the benefit that is so easy for them to qualify for. And so payment, the soulless payment act, is a point of aggregation and compulsory interaction. When you think about what would you do if you had the attention of every household and business in the jurisdiction on a monthly basis, how would you use that interaction to make their experience better? It’s a touchpoint or a concentration point in which to advertise other opportunities.

KP: We’re doing kind of the same thing, and we’re trying to get into that same sphere. So paying a parking ticket, we all hate it, it’s very painful. We now push out street sweeping notices, but some people still have to pay parking tickets, so we try to do targeted interactions that say, “Hey, you have to pay a parking ticket? Sign up and get street sweeping reminders.” And that’s how we encourage people to get in there. “You can prevent these tickets later on down the road because we’ll set you up with an account, and then send out reminders.” It’s working beautifully, actually. It’s one of our more successful things.

MD: Yeah, absolutely. It’s worth saying too that a water department, the utility will have the most frequent interactions and the most, the broadest, across the broadest set of a demographic. So if you have a water department or internal municipal utility, that’s a great anchor point with which to start. And if not, partnering or trying to work with the service providers in an area is another strategy.

Moderator: What are some projects, some initiatives that you’ve seen work well in other cities or other companies, or where can people look to see other examples of personalization done well?

MD: The state of Michigan I admire from afar. They took their major benefit programs and aggregated the application process into one space. Because they were all income-based but had different criteria, one application — say for SNAP — would automatically qualify you for other programs.

MC: I would say the Indy.gov site we were just looking at that the other day, I mean I think that site is great the way they personalize a service and they prioritize them based on usage. I was quite impressed with that.

KP: We’re just kicking off today, in fact, a gigantic content audit, which I highly recommend everybody to do that. The content audit covers everything from brand and standardization to accessibility, and we extended accessibility beyond required accessibility. This gets into that personalization and customization. Make the language on your site the level of seventh to ninth grade reading level — which is the standard in the United States for reading level — using plain, easy-to-read language. Then, if you put it through a translation tool for your top languages, for instance running it through in Spanish, see if it makes sense when you run it back from the Spanish to the English, and continue to refine from there.

That’s actually true for creating straightforward layouts as well. If you create layouts that are predictable so that your contact information is always in the same place, for instance, you’re not making your users guess what’s going on from one page to another.

Audience Question: I’ve heard a lot of really good examples of personalization. I think the flip side is that it can easily go too far and be creepy. Are there any principles that you’re considering during your personalization to make sure that you don’t go too far?

KP: Our rule is, make people put in the minimal amount of information, but give them the fields for optional items and let them decide what they want to buy into. It’s based on service. You have to put in your address if you want street sweeping reminders. But you don’t have to put in your address if you just want to create an account. Our pocketgov tool does 311 cases, so you can put tons of information into it but you get out of it what you put into it.

So if you’re just doing a regular 311 ticket where you don’t have to put additional information, it’s really just your name and your email address. If you want street sweeping reminders, then you start adding in your physical address. If you want to know about property taxes, you can put additional information, so you can pay your property taxes. So, it’s all based on the service that they’re signing up for and the things that you’re providing them. So I think that’s actually a good rule in general: let them decide. Make them put in a very minimal amount. And it speaks to what you were talking about with the password. You know, give us a password. You don’t have to put anything else in. We already have all that information in this case, give us the minimal amount and then we’ll get there when we get there. When you’re interested.