Data Management: Data Deputies

Posted on December 1, 2004 by Chief Marketer Staff

If you’re reading this magazine, chances are good that your business depends on high-quality, reliable data. But access to that information is being questioned on a number of legislative fronts as consumers become more and more concerned about privacy in the electronic age.

For those in the data gathering business, these are challenging times. To discuss the issues, Direct brought together representatives of some of the industry’s largest information compilers for our first database marketing roundtable.

Among the hottest topics were the difficulty of tracking down new data as public sources become threatened; incentives to get clean self-reported data; convincing state and federal legislators to listen to the direct marketing industry’s economic case; and training.

“Obviously, stepping up the education level makes us all safer legislatively, especially with the constant changes in the business,” said Stefanie Pont, who moderated the discussion.

PONT: With the ongoing access challenges to public data sources, what steps are you taking to maintain quality and quantity in your databases?

COLCLASURE: That’s a very important objective for our company, to replace sources as they dry up. A great example is when the Shelby Amendment took much of the [Department of Motor Vehicles] data out of circulation for marketing purposes. The element we lost that was most important to us and to our clients was age data. We had to move pretty aggressively to try and recover. It was a hard blow, so we began looking for alternative sources, survey data, product card information and other sources of public-record data, like non-restricted state voter files.

MALLIN: We’ve looked at ways to leverage the Internet and online surveys and create opportunities to gather self-reported information. We’ve utilized that resource fairly heavily to at least start to fill the hole. We’re also looking at some out-of-the-box alternatives, besides some of the traditional ways like warranty cards, [such as] working with various data owners and talking to them about ways we can incorporate their data into arrangements we’ve worked out. There’s more onus on tactical strategy in terms of how we can find information to add to our respective databases and meet all our requirements, both from privacy issues as well as the ability to populate our database.

CONNOLLY: We’ve struggled with the same issues. We have a self-reported piece of our business that we’ve had for some time and we’ve taken advantage of that. We also have increased our online presence with the acquisition of MetaRewards. The thing we struggle with now is there’s a different profile of individuals that we’ve been getting from newspapers, magazines, whatever — and what you’re going to get online. So there’s still a gap you have to look at, and we’re searching for alternative sources to fill in the older gap. The Internet [responder] is typically younger. Online we’re getting 10% of those 60 years old and up, and the population is 23%, 24%, 25% and growing in that area. We’re really missing that gap. You can’t just go to online, you have to use a mixture. The data sources have age, and you’re really looking for birth information, not age. It’s been a challenge but it’s there.

RATLIFF: Not only do you lose key pieces of information like age, but access to the base names and addresses has been compromised a bit. And while some of the sources like warranty cards, online registrations and so forth are geared to supply that kind of data, we’ve found we have to scrutinize those sources more than we did in the public-record realm. In some of those situations the consumer just [doesn’t have] the incentive to be as forthright. So our testing and data polling processes have become more important with some of the sources going away.

PONT: With a mix of sources, how are you balancing online vs. offline information in the database?

CONNOLLY: You really have to review every source and look at what incentive consumers had when they filled out the survey. If you’re going to give them a new Porsche, and you’re going to deliver it to their house, they know you’re going to verify everything about them. So they’re going to be pretty truthful. Look at the example of insurance to get a different [perspective]. If you’re applying for life insurance, you might say you’re 55, because over 55 the rates go up. If you’re a young person applying for car insurance, and you’re 24, you might say you’re 25 or 26 because the rates go down. You really have to look at what’s happening with the data. We test every source online and offline to make sure all that data is accurate, that the name and address is accurate. Maybe if you’re after age, and the age is good, but you’re not sure about the names and addresses, you might just keep the age. You have to look at what’s good and what’s bad from every source and take what you want.

MALLIN: Obviously, we’ve had less challenges on the business-to-business side than on the consumer side. Our major opportunity there is that we’re still able to do our standard compilation and then do heavy telephone verification of the database. I think the public data sources for B-to-B information from that perspective have allowed us to stay fairly true to the original core of how we’ve been compiling. If you look at online vs. offline on the B-to-B side, the real resource of information continues to be printed material we can compile from, or direct contact with businesses. The pressure for information is much more on the consumer side. It really is much more than acquiring data. It’s matching it, analyzing it, cleaning it, verifying it and looking to see if you get the same data from three sources or multiples so you can have the confidence to potentially integrate that into your consumer files.

COLCLASURE: Our clients need business intelligence and they need more of it today than ever before. All the raw materials we want increasingly are regulated at the state level, at the federal level. You have a layer of state regs, a layer of federal regs, you have industry code, you have your own privacy policies, and it’s a real ‘gotcha.’ Our clients’ needs break down into two categories. They either need intelligence to do things like identify verification or fraud detection, or they need it to help them grow their relationships, either to acquire new customers, [maintain] their [existing] customers or upsell or cross-sell them. We reverse-engineer and go through all the regulatory metrics back to the raw materials and acquire those bits of data. And it’s complicated today. You’ve got to understand the data element, what regulations bear on that particular element, when you multisource and compile. Then you’ve got to look at the contract’s prohibitions and permissions and bring forward all the regulatory components. Sometimes a particular data element may be used for customer relationship management by Acxiom clients and sometimes it may only be used for risk-management purposes. You have to manage that.

PONT: What do you see as the next big threat to data access?

COLCLASURE: Public-record data. This country was founded on open public records. It’s unique to the United States economy like no other country in the world. We have open public records where you can go and verify information. Take property data. When you’re buying a house you can go and verify whether the person you’re buying a house from really owns the house. That data is under threat. We all rely on that data as one way to multisource and compile our information products.

PONT: State or federal?

COLCLASURE: Most public records are managed at the state level, and there’s a lot going on. Most of it is regulated under the Freedom of Information Act. And you have court records, property records like county assessor records, you have voter file records. You have supplemental regulations — often mitigating regs that govern the voter files — and we see those voter files which historically have been relatively open being closed and regulated in a way that says [those] files may only be used for political purposes and not commercially. That’s a real challenge for all of us, because that’s one of our important sources. Real property data — used in the real estate industry, for mortgage lending, for title search — we see those under assault right now. The county clerks who are stewards, some of them are zealous. And they themselves, contrary to their states’ laws, say ‘You know what, I’m just going to stop any commercial entity from acquiring the data.’ And then we have to go in and challenge and say, ‘Well, your law says we may have this because we build these important tools for the mortgage lending industry, for the real estate industry. So it’s sort of segmented at the state level. At the federal level what we see is federal politicians trying to reach into the states like Shelby.

MALLIN: One of the challenges is the difference between the state and local laws being proposed. It becomes almost unimaginable to consider that we would have to do data strategy and compilation of potentially 50 different regs that contradict each other. I think the reality is that as public-sourced data becomes more restricted, it’s going to put more and more pressure on us to come up with strategies for self-reported information. Some people are willing to give you information about themselves and their preferences because they feel they’ll receive some added-value information. But that’s easier said than done, considering that we’ve all been very much focused on saturation marketing and volumes of information. It’s sort of a balance. How do we maintain our desire to be appropriate and privacy conscious vs. the [need to deliver] marketing information?

RATLIFF: I think you hit on a key point. You’ve got to make sure the data has value in it for the consumer, and that there’s a value in it for us. To get back to public records, there’s still some pieces of information I even hate to bring up because you worry that it might bring on further legislation. But there’s still things like occupation data that you can get at the state level. There’s hunting and fishing licenses and so forth. We rely on the tax assessor and the fee transactions to compile data and those are all great sources of names and addresses. And some of those bring along key pieces of information. The fact that someone is a homeowner, has a swimming pool or whatever is information I think is valuable and we worry that there’s going to be more erosion of that data.

PONT: Is a pre-emptive strike to start compiling self-reported data wise? Should the industry start gathering so we’re not all caught short?

CONNOLLY: We’re already doing that. You have to be involved, you have to have teams out there checking how things at a state level will affect you. You can put up a survey for self-reported data, but what do you ask? [Take] ailment data. In the old paper environment, you’d ask the respondent [in a survey], ‘Do you have asthma?’ A secondary question would be if anyone else in the household suffers from asthma. Valid question, [since] we’re looking for a household that has an ailment. There’s legislation out there now that will take that second question away in California. The person responding can only [answer] about themselves, they cannot respond about other members of the household. So you have to do your acquisition differently. You can ask different kinds of questions. You have to tailor your self-reported data. I think if you’re strong in the self-reported market, you’re going to be OK. Public data is under scrutiny and it’s a good target for politicians. They can hang their hat on it. ‘Do you know that this realty file that’s coming out of Saline County, Nebraska has Social Security numbers? It’s an identity theft opportunity.’ We all know that’s not true. You have to look at all of it, even at the self-reported level.

PONT: Public focus on direct marketers, especially compilers, has become fierce over the last five years. What steps are you taking to make that public persona a little more palatable to protect your businesses, to move along and work well with Congress?

COLCLASURE: Everyone around this table is my colleague when we formulate a strategy to tell the story to legislators at the state and federal level. The unfortunate thing is that many politicians view information regulation as a cost-free political issue. It is not. Information is fuel, it fuels our businesses, it fuels the economy and it delivers real tangible value to the consumer. That is a point many times lost on the politicians. It’s a very sexy political issue to run on. ‘I regulated information, therefore I stopped identity theft. I saved your privacy.’ The truth is that the appropriate transfer of information in commercial entities is not how identity theft happens. What we’ve got to do — and I’m not sure we’ve done it yet — is to figure out a compelling way to tell the economic value story, because that’s the truth.

PONT: When we talk about public persona, we have to consider it from the perspective of the government and the consumer. In some cases, your names are well known to consumers, while others are not. Have you looked at your impact on consumers to help you survive?

MALLIN: Those are two really weighty issues. I think in the past few years our organization has attempted to be fairly active and talk to Washington. The biggest challenge we’ve had is making the economic issue to Congress. When you start to look at how many industries we really affect, in terms of the whole marketing segment, it’s enormous. We support every aspect of sales and marketing in this country, going all the way from a small lettershop to the largest ad agencies. They all feed off what we do in many ways. We haven’t been able to make a galvanized statement to [outline] that impact vs. what groups [in other industries] have done to state their case and protect their core industry. It’s incumbent on everybody here to stay active in that role. And that’s a tough call. How much attention do you want to draw to your company? The other issue is how we position our issues to the consumer. I think we have to make the case that we’re above board, we’re high profile and we want to do best practices. That’s tough when you have political expedience pushing against it. We’ve got to be in public forums to say we care about privacy — and we also care about delivering good marketing information.

CONNOLLY: We have a pretty active government affairs group. We also have a consumer advisory council that met recently in Washington. It’s a collection of about 20 marketers, including privacy people. What we’re trying to do is give them information and get feedback from them on what they see happening on the consumer side.

PONT: Is it time to take the conversation to the consumer, or are you still focused on keeping it at a legislative level and working backward? Or is it a twin-pronged thing?

CONNOLLY: I think you have to do both. You definitely have to work on the legislative side, because many of the legislators don’t understand what one word — [like] ‘opt-in’ vs. ‘opt-out’ — will do in a document. It’s a huge difference.

RATLIFF: At a more fundamental level, we have to make sure that the customers we’re dealing with are out there doing the right thing in the industry and they’re not out there giving us compilers — the people who bring all the data together — a bad name. We’ve got to make sure that the offers going out are appropriate, we’ve got to make sure that how the data is being used is always appropriate, and again, we’ve got to make sure that our customers are providing a value to consumers. I think that even if you don’t have some proactive message out there to consumers, if they’re getting things that are perfect to them, if they’re getting contacted in ways they want to be contacted, in the end the consumers will be happy with the situation.

PONT: That brings up an excellent point. With the increased focus on data use, are you directing your salespeople to look at mail pieces? Are you checking companies? Has the focus really sharpened on who you’re selling your data to and who you’re allowing to use your data?

CONNOLLY: We review mail pieces and telephone scripts for all of our sensitive data. When we get a new client, we look at how they use the data. Salespeople don’t like to hear [it, but] the first question [new clients are] asked is, ‘What are you going to do with the data?’ You can tell pretty quick by looking at that [element] and [what the] company is whether you want to do business with them. You really have to control what you’re doing. We’re eventually going to have to review everything.

PONT: It’s interesting, because for a while with mailing pieces to response lists, standard operating procedures may have gotten a little lax, unless it was COPPA (the Children’s Online Privacy Protection Act of 1998) data. I’m sensing from what you’ve said that you’re shifting ahead of the regulations.

CONNOLLY: That’s what you have to do. You have to be on the bleeding edge. Is it tough sometimes, because you’re refusing a mail piece and that business goes somewhere else? Sometimes, but in the end you’re going to be better off. We have a lot of brokers and resellers that we use and their contract calls for them to review the mail pieces. We’re probably going to be looking at how you ensure all those mail pieces — ‘Do you have samples? Have you looked at it?’ — because it’s a risk.

MALLIN: In earlier days there was that distinction between response and compiled and I think that line has become very fuzzy. Will everyone look at every single order, thousands upon thousands? I doubt anyone here could stand up and say that’s going on. But I do believe what we’re going to see [is companies] starting to find ways [to conduct] certain audits if we see things we don’t like.

COLCLASURE: [Determining an] appropriate use by the client goes beyond screening the mail piece and seeding the list so we can see what the consumer is receiving. It’s beyond that. I’ll give an example: exact date of birth. Age data has been a high-need data element. But because of the perception — not the reality, but the perception — that exact date of birth, like a Social Security number, contributes to identity theft, we have put a team into place to manage exact date of birth when it’s delivered in our marketing products or used to develop a household segmentation score. We have taken exact date of birth out of availability in our list products, in our enhancement products. It’s not a standard offering anymore. If a client needs it, they have to present a use case and we evaluate the case. If they have appropriate security controls, appropriate access controls and a valid use case where they really need exact date of birth and if we’ll make an exception, we [offer] the data to them under a [provisional] contract. The appropriate use by the client of the data is really what it’s all about.

PONT: What type of feedback are you getting from your resellers or your list brokers, in terms of how they deal with their clients on these issues?

COLCLASURE: It’s tough because this is baking in compliance costs and that inevitably will eat into their revenue portion of the business. But it’s a changing dynamic. The world of data has changed and we need to bring them along with us because we all depend on them.

RATLIFF: I think the thing we struggle with in the resell market is that they can tap into all of our companies respectively, so if you go to a broker who is doing something with the data you don’t think is appropriate, that’s a hard decision to make. Do we no longer let them have access to the data, because we know they can walk down the street to one of our competitors? It’s a hard balancing act.

CONNOLLY: You can talk to your brokers and resellers and train them. It becomes a cost issue for them. They have to save package samples if you audit them on a regular basis. A salesperson calling them can say they need to see samples for X number of jobs. If you know you’re not going to audit them, then they’ll probably save that cost. If they think there’s a chance you’re going to audit them, they’re going to be prepared.

MALLIN: And that’s a huge change in the metrics, because in years past the audit function was often around financial issues, especially if someone had a segment of the file in some sort of arrangement. So now the reality is perhaps there’s a financial audit around terms of how much data was used, but [there’s also] the client’s audit, which is totally different. And I think [in the end we may] have to charge more to cover this, which may not make our resellers [happy]. It’s a challenge we have to face and they have to be prepared to absorb that.

PONT: How will this affect installed data? If you don’t have your hands on it for long periods of time, now what?

CONNOLLY: I think you’ll struggle with the same things as the brokers who have the data. But if you have installed files I think it’s imperative that you audit them on a regular basis — actually an irregular basis — so they don’t have a lot of notice. If you don’t, you’ll potentially leave yourself open for a bad issue.

PONT: Do you think installation is going to increase or go away in the future, because it’s just too risky?

CONNOLLY: We’d like to say yes. (Laughter.) You really like to have control of your data, you’d like to not give the data to anybody, but then you can’t make any money that way.

MALLIN: I think it will continue, but it’s who you do business with. If you’re doing business with companies that passed the litmus test in terms of how they handle the data, how they do their marketing, [you’re probably] OK.

COLCLASURE: We are stewards of a great deal of data, and of course there are pressures. It goes back to what kind of shepherding you extend to that data when it leaves your shop. Do you have a high degree of confidence in who you’re site-licensing it to, do you go in and do some checks, some irregular audits, and what kind of contract provisions do you put in place and how do you hold them accountable? It’s a necessary part of business today.

CONNOLLY: Anyone who installs data — to someone who resells, say, to end users — [needs to have] some sort of training or knowledge of that reseller, maybe a certification process. You can’t just continue to put that data out there. You’re going to get burned. You’ve got to have some way to make sure those resellers are knowledgeable with what’s going on, [that] they know they’re going to be audited. If they’re not ignorant and they purposely misuse the data, you’ve got to be willing to hold the data back.

RATLIFF: We’ve been extremely selective in which resellers we’ll allow to have our data. Training has been a big issue for us, making sure that everyone who has access to our information is extremely knowledgeable about what they’re getting and how to use it. [In the past,] you might have given someone the data set and off they ran to resell it. Today you’ve got to make sure they’re informed, that they realize there’s contract stipulations around how you can use the data — and therefore they’re liable for that same type of stipulation because they have access to your information.

PONT: Obviously stepping up the education level makes us all safer legislatively, especially with the constant changes in the business.

COLCLASURE: We have some pretty extensive training that we require our sales reps to [take] and become certified. We also train every new hire. We’re pretty severe about it.

CONNOLLY: In addition to training, you have to know all the pieces when you’re solutions selling. How are you going to use [the data]? What are you trying to accomplish?

COLCLASURE: Well said. You identify the client’s need, what they’re trying to sell and then say, ‘This is the information we can bring to bear for you to solve your problem. And here are the ways this data can be used and here are the ways it cannot.’ It’s a very consultative sale.

MALLIN: We have monthly sales meetings where we bring in 150 reps from around the country. Every meeting has an hour dedicated to issues related to the data, like privacy and compilation.

RATLIFF: One of the other approaches we’ve taken is not only training sales folks but also all the people that help them fill orders. We focus on making sure they’re aware of what the policies are and how the data can be used and how it cannot be used. We sort of have internal company advocates for the privacy council that say, ‘By the way, Mr. Sales Rep, you may not want to use the data this way.’ We’ve got people throughout our company who know what should and should not be done with the data, vs. just training the sales reps and hoping they will weigh the revenue [against] the privacy concerns.