Staying in touch with the customer
Podcast with TOLERANT Software Managing Director Stefan Sedlacek
In a podcast interview, Stefan Sedlacek gave an insight into the problem of customer data and how this can be solved with the help of the relocation data tool TL Move. The podcast interview was conducted by Ashley Steele. You can read an extract from the interview here.
You can listen to the podcast in German here:
Ashley Steele: Hello Stefan. Today’s topic is the quality of customer data. TOLERANT Software has a solution for this, TOLERANT Move. But before we get into the solution, can you please explain to us what the problem of customer data quality is all about? What is it actually about?
Stefan Sedlacek: Hello Ashley. Nice to see you. TOLERANT Move is about the fact that up to 9 million people in Germany move house every year. This means that the data in customer databases becomes outdated relatively quickly. Addresses become invalid, which means you can no longer deliver letters or similar items as you would actually like. That’s the background to the problem.
Ashley Steele: I have a question about that. Nowadays, many invoices are sent digitally as PDF files. Are postal addresses still a problem these days?
Stefan Sedlacek: Yes, they are indeed still a problem at the moment. It is true that many letters are sent digitally these days. But just as many invoices are still sent by post. Or even bank statements. In most cases, account statements are still sent by post these days. If they then no longer reach the right recipient because they have moved, this becomes quite a problem. What’s more, many companies have not yet digitised their invoices to the extent that they could also send them digitally by email. When an insurance company sends out its annual invoices once a year, for example, there is always a relatively high rate of up to ten per cent of recipients who do not receive the invoice at all. The insurance company then has ten per cent returns that have to be reworked.
Ashley Steele: So invoices are sent digitally, but not all of them. There are legal regulations that force banks and insurance companies, for example, to send invoices in paper form. The problem of the correct postal address also affects consignments of goods, i.e. when something is sent physically and the customer has not corrected their address in advance, for example. Despite digitalisation, this is still a very big issue in Germany. How many addresses does a typical customer have? It is probably difficult to define what a typical customer is. If we’re talking about an insurance company or a bank, for example, how many addresses does a company actually have in its database that need to be kept up to date?
Stefan Sedlacek: Our smaller customers “only” have 100,000 addresses. The typical bank or insurance company, however, works with several million customers. A well-known insurance company here in Stuttgart typically has 6 million addresses. In terms of volume alone, the amount of mail sent out here is already a very large amount. Processing the returns here is really no fun at all. So of course we are happy to have any kind of electronic help to ensure that addresses are correct and letters are not returned.
Ashley Steele: You’ve now made it clear where the problem lies: Someone moves house and their address is not kept up to date. Now comes the interesting question about TL Move. Where do you at TOLERANT Software get the information that someone has moved?
Stefan Sedlacek: We get the data from Deutsche Post Adress. Deutsche Post Adress collects the relocations via relocation notifications that people make. Of course, in most cases, someone who moves would like their mail to be forwarded. If they then authorise the data to be processed electronically, we also receive the data via Deutsche Post Adress.
Ashley Steele: What percentage of people who move fill out a forwarding order like this?
Stefan Sedlacek: Up to 80 per cent.
Ashley Steele: That’s a relatively high figure. So you have to have the customer’s consent to process the data. Before we get to the solution: Relocations are one issue. What happens when someone passes away? Does TOLERANT Software also receive information that someone has died?
Stefan Sedlacek: Yes, we also receive this information. This is a separate data module in the software. However, I have to say that information about deceased persons is not used that often in practice. We would receive the data, but this option is not used that often by customers.
Ashley Steele: You said that consent is required for forwarding orders. So the people who use TL Move are allowed to process this data. Are there any legal requirements that you have to comply with? Is your software tested or certified in any way?
Stefan Sedlacek: Yes, we are of course subject to the guidelines that Deutsche Post Adress also stipulates. The software is certified in accordance with these guidelines. The data itself is secured and encrypted in such a way that it cannot be passed on. Only the software itself can use the data and accordingly it only releases the data “in the case of legal interest”, i.e. if the user actually has a customer with the same name. It is not possible to pass on the entire database and then sell it.
Ashley Steele: But that also means that guidelines such as the European General Data Protection Regulation are complied with.
Stefan Sedlacek: Yes, exactly, that’s the background to it. The GDPR guidelines are complied with. This starts with Deutsche Post Adress, where it is already being checked. These checks are then also passed on to us.
Ashley Steele: So, we’ve now explained the problem and also talked about how the data comes to you and about the fact that everything is legally checked. What does TL Move do then? If we now assume that you receive tens of millions of data records from Deutsche Post Adress. What does TL Move then do with these addresses?
Stefan Sedlacek: These data records are used to check the customer’s databases, i.e. we take each customer data record and check and validate it against the data from Deutsche Post Adress, i.e. against the relocation data. As soon as we establish that a move has taken place, this information is released. The customer database is then enriched with the new address information, so to speak. We carry out error-tolerant checks, i.e. the data we have from the customer and from Deutsche Post Adress do not have to be identical. Thanks to our search technology, it is still possible for the data to match, even if they are not written identically. This could be a spelling mistake in the name, a small difference in the house number or something similar. We get all of this so clearly separated that we can even tell spouses apart. We know, for example, whether only one of the two spouses has moved out or whether the couple has moved together.
Ashley Steele: So the hit behind it, if I may put it that way, is precisely this error-tolerant data synchronisation. Not everything has to match 1:1. The intelligence of your software gives a high hit rate and finds “Ashley” even if, for example, “Aschley” is entered.
Stefan Sedlacek: Yes, you explained that well. The name can of course differ to a certain extent. But of course we also take other characteristics into account, such as the addresses, the complete place of residence and whatever other information is available, and compare everything together in an error-tolerant manner to arrive at an overall result. And then the software decides that it is a match and issues a new address.
Ashley Steele: What technology is used in TL Move? What is behind it “software-wise”?
Stefan Sedlacek: We have our own technology behind it. This is the fault-tolerant search that I have already mentioned. It’s very close to an AI-based approach, which helps us to bring together data that doesn’t seem to belong together at first glance. This is our own technology that we have integrated into the software. Everything else around it is based on standard technologies for connecting to databases and processing files and the like.
Ashley Steele: So that means that someone who decides to use TL Move is taking a small risk, because the software is not only based on new technologies, but also on established ones.
Stefan Sedlacek: Yes, you could say that.
Ashley Steele: Can you also name a few customers who have already used TL Move successfully?
Stefan Sedlacek: Yes, I can of course name one major customer from the automotive sector. This is typically BMW AG, which uses it for many millions of data records. And in the banking sector, perhaps to name a competitor, Mercedes-Benz Bank, which also uses it. Other customers of ours include ING-DiBa and WGV-Versicherung in Stuttgart.
Ashley Steele: That means that there are really well-known customers who are already using the product or the software.
Stefan Sedlacek: Yes.
Ashley Steele: We have really talked about the problem and also about the use of TL Move. But the question is: Can you summarise for us at the end what the benefits are for a customer when they use TL Move? What are the unique selling points of TL Move compared to other types of address matching software, if I may call it that?
Stefan Sedlacek: The advantage is, of course, that the customer receives updated addresses. These updated addresses help them to optimise their processes and also to address the customer correctly. We support this with our error-tolerant technology, as I mentioned earlier, which in turn gives the customer the assurance that their data is correctly matched with the Deutsche Post Adress data. This error-tolerant technology helps to find the data accurately and then feed it back into the database. And it is precisely this error tolerance that is a unique selling point that we are very proud of and that we regularly develop further.
Ashley Steele: As you said earlier, the software is already being used by well-known customers. The risk for someone who wants to use the TL Move software as a new customer is therefore low.
Stefan Sedlacek: Exactly, that’s right. We actually have a lot of customers who also use other versions of our software. We are very proud of what we have achieved in recent years.
Ashley Steele: Great. Thank you very much for the interview, Stefan. You explained the problem and the solution really well. I’m looking forward to the next conversation with you. Thank you very much.
Stefan Sedlacek: Thank you very much. See you next time.
Ashley Steele: See you next time.