What is our primary use case?
SEEBURGER, along with four or five other big vendors, focuses in the integration space. When you talk about data integration, there are two major aspects of it. There is the transactional messaging side and the batch file-based side. My team is focused on the batch file-based side of things. We have a completely different team (with a different set of software) who does the transactional messaging aspects. We are using it for all secure file transfer use cases throughout the organization with multiple different data patterns: moving data within the company, moving data to and from outside of the company, and ad hoc file transfer. Any type of file-based secure connectivity goes through our team using this product.
Currently, it is on-prem. We have a cloud initiative, which has been rolling for a couple of years now, like most companies. It is on our radar for later this year. We are going to spin up another project to consider either moving to our own AWS or SEEBURGER's AWS and their iPaaS environment.
How has it helped my organization?
No matter how much you automate in the file transfer space, there is always more to be uncovered in a big company. Users, especially in the business area, will start doing their own thing and interacting with some external webpage to upload or download files manually every day. They kind of incorporate it into their daily tasks. When we discovered that, we were like, "Why are you wasting all that time? What if you are out?" That has been one part of starting to consume different website APIs, to push and pull files to and from various vendor websites that was historically done by users manually. So, there is an automation aspect to it. Beyond that, application-to-application connectivity, which historically went over protocols like SFTP or batch files, is conforming over to APIs now because everybody wants to be faster and use APIs. Therefore, a lot of these application-to-application data flows have changed over to APIs over the last year. For example, I used to get one API request a year in past years. This year, I get one or two new ones every week or two. APIs are just taking over.
It is good anytime that you can take a user who is doing something manually every day out of the picture and automate a process, e.g., going to a vendor's webpage to pull or push a file every day. Although there was a one-time cost to do the development work, you reap the ongoing benefits because now you don't need to have that user spending time doing it every day. You don't have to worry about if that user is out or gone for a week on vacation. Things can just happen automatically. There is definitely a benefit.
Because we are in the financial services industry, PCI is huge. You have to comply with PCI regulations. That has primarily to do with credit card numbers, but really any account number or sensitive data. What is nice about SEEBURGER BIS is it has made it easy to patch our yearly PCI audits with this thing called PCI realm. You can configure any of your data flows into that PCI realm that you need to. It automatically complies with regulations, offloading the data as soon as the data flows through the system. It doesn't store any copies. It offloads the data encrypted in an encrypted state to our PCI zone to be stored for X days during our backup period. That is all out-of-the-box functionality, so you don't have to waste your time trying to figure out how to comply with PCI compliance rules because it is already built into the software. You just have to configure it in that PCI realm.
We are getting these use cases now that APIs are coming into the picture where, historically in our company, the data integration has been broken into the two major areas. Transactional messaging is on one side and batch file-based transfers are on the other side. Now, you are starting to see those two areas kind of merged together. Because usually when you get a batch file over API, they want us to break that batch file up into individual transactions, iterate over all the records in that batch file, and post them as transactions into our messaging system.
There is now this interesting sort of convergence between the messaging space and the batch file-based space which is now sort of coming together because of APIs. So, this is another area that I am seeing a lot of requests for lately, "Hey, I want you to still get a batch file like you have always done in the past, but it is going to come over API now instead of over something like SFTP. In the API, I want you to iterate over every record in that batch file and post those transactions individually." This is another big growth area right now. Therefore, I am working on solutions to be able to support that. This would be an area of growth because we will be using these batch files to post into more internal systems and do it more flexibly as individual transactions, instead of as a big batch file. This is an area that we are looking to grow in the next year.
You could make the argument for time to market if there was a user doing something manually every day, then we came along with SEEBURGER BIS and automated that. So, instead of waiting around for the user to load the file once a day at whatever time they do it, we had the system automatically pull the file, possibly early morning, and pass it through immediately into the back-end processing system. There are some cases like that, where historically it took eight hours, because a user had to get engaged, do something manually, maybe convert something manually to transform the data themselves, and then they would have to manually load it. Here, we come along and run a workflow in two seconds that streams it right through. Therefore, you could make some arguments that we sped up time to market by automating previously manual processing. However, as far as just general B2B file transfer or application-to-application, I don't know that you could claim that the software itself has sped up time to market, other than just coding workflows a bit more efficiently to take out seconds or minutes to make them run faster.
What is most valuable?
The most valuable features are the solution's stability as well as its ease of use and flexibility to configure a workflow with as many (or as few) steps as you need. It offers us more value to our internal customers because we can do so much more than if we had a software that was super rigid. If you are looking for a software that emulates writing a piece of code where you can have as many steps as you want in whatever order you want to, this is the one. SEEBURGER BIS is so much better than other solutions because of this.
Their entire software suite is 100 percent homegrown. Every component that they have built was built to be integrated with another component. It is all one product. Due to acquisition and the integration space being a big thing, other vendors tend to go, "Buy this, buy that, and then buy that." They try to bolt them all together and make three different vendors' products work in concert as one. My experience has been that it usually leads to confusion as well as bugginess and problems. SEEBURGER BIS is all one product, all homegrown, and everything is fully integrated.
API adoption is on the rise everywhere. Even in the file-based space, APIs are being adopted a lot, especially at our company. One thing I like about SEEBURGER and their transformation engine is it is completely integrated with JSON file formats, which are typically used in API calls. I just did an API over the last couple of weeks with a complex data structure and SEEBURGER BIS Mapper handled that with no problems. As we look to the future, APIs are really taking hold. It is nice that SEEBURGER BIS can totally handle whatever comes our way that is API related.
What needs improvement?
We occasionally get ZIP files. Sometimes the ZIP file has one file inside of it, and sometimes the ZIP file might have 30 files inside of it. We have been working with SEEBURGER to enhance their PKUNZIP process to be able to unzip multiple files in a single workflow instead of just one file. This is still something that is in process.
For how long have I used the solution?
I have been at my current organization for five years and using SEEBURGER BIS during that time. Our company's relationship started with SEEBURGER in 2013.
What do I think about the stability of the solution?
In five years, we have never had an unplanned outage that was caused by the software. Every company has outages, but in our case, they have always been caused by our own infrastructure, e.g., a router went down, a cable went bad, or a switch had a problem. It was not caused by the software layer; it was always caused by the infrastructure and things that are under our control, not the vendor's. There is a peace of mind as a developer with not having to worry about having to be up in the middle of the night with software not working like it is supposed to. The stability of this software is by far so much better than other software solutions that I have used in the past.
I have come from a place where other software was up and down all the time. I would have to be up at night, burning sleep, and trying to support outages. To be in a place where we have no unplanned outages is great, you get a lot more sleep, which is good.
What do I think about the scalability of the solution?
It is mostly our team who has access to the front-ends to develop configure processings. We don't make it too widely available across the company. We have a few pockets of people who can log into the portal to view their data on their own kind of self-service. Other than that, we mostly do behind the scenes data integration, moving data between applications and external partners. That is all done by our team. It is not done widely throughout the company, it is just one small data integration team. We serve every application in the company and are connected with a couple of 1,000 external partners as well. The touch points are many, but the people who have access to the GUI to actually do the work are few.
The scalability is pretty good. We haven't had to do a ton of scaling exercises over the years. Our volumes have stayed fairly static and grown at a certain rate every year. We just reassess them with our professional services person once a year and make sure that we are watching our metrics, memory, and storage really closely. We have that in our daily monitoring. As we see it going up, we just go, "Okay, we need to add some more RAM memory or more Java heap space."
I would say the process of scaling is pretty easy as long as you stay on top of it and monitor your throughput really closely to know the numbers, knowing when something is growing. If you put in a new integration that brings in a whole lot more traffic, obviously you have to reassess and make sure that you are scaled to handle it. I have got a project like that coming up soon which will take us from thousands of files a day up to millions of transactions a day. This is something where we are looking closely at scalability and figuring out what is needed to be able to support that new volume, and not have an impact on the existing.
The fact that the solution is available in the cloud, on-premises, and as a hybrid deployment makes it flexible and scalable for us. Every company has cloud initiatives going on right now. To know that there are options out there gives our company more things to think through and price out. We have done a lot of pricing exercises around those three different options, so we have a pretty good sense now of the cost differential between on-premise versus cloud versus vendor iPaaS cloud. We know where things are going to fall cost-wise, so now it is just a matter of, where do you want to put your money? For example, do you want to have more staff to maintain the system yourself with all your infrastructure people, or do you want to outsource that, pay that money and some more to the vendor to maintain it for you? So, it kind of just depends on where your priorities are as an organization. I think it is a good thing that there are multiple options. It has given us a chance to slice and dice the numbers.
How are customer service and technical support?
Since the very beginning of the relationship between my organization and SEEBURGER, they have been solid for everything from their support team to account managers and sales to professional services. We have always had a contract with professional services. We have used their developer personnel throughout the year on various efforts, and it has always been a solid relationship. It is one of the better ones that I have worked with. They are just an email away when we need to reach out to them.
In the case of opening support cases, it is pretty simple. You usually hear back within a day. Even though they are based in Germany, they turn around your request pretty quickly.
You occasionally get a use case that comes along, and it is something new that hasn't been done before. SEEBURGER BIS out-of-the-box might do part of it, but not all of it. In those cases, we submit an enhancement request to SEEBURGER who gets it to their engineering team. Usually, those are turned around fairly readily.
The enhancement onboarding process at SEEBURGER is pretty good compared to other vendors that I have worked with. We have a monthly meeting with our SEEBURGER contact in professional services. We keep a list of things, and say, "Here are the things that we need help with and our enhancement requests." Then, that person reaches out to engineering and gets an ETA, so we have a date of when that thing will be delivered. It is not like you give it over to SEEBURGER and forget about it, then it never happens. There is ongoing communication to the right people who know the software, what is needed, and when it will be delivered. Even when it comes to things that we don't have but need, like the multi-file PKUNZIP, usually it is a fairly good experience.
Which solution did I use previously and why did I switch?
Before we started with SEEBURGER BIS, we had as many as 13 different integration software spread out across our company. Over the early years that I was here at my organization, we were able to consolidate that down to just SEEBURGER BIS. We reduced a lot of extra costs from using other software products and having a lot of extra things to support. Our support costs went down for infrastructure, etc. Thus, it is nice to have everything fully integrated into one product that can do everything.
The number one reason why I would not want to go back and use another software after having experienced SEEBURGER BIS is its flexibility when it comes to file transfer workflow. You can configure your file transfer workflow completely customized. You can put the steps in any order that you want. Your file transfer flow might have three or 20 steps. You simply bring in those steps as activities in a workflow in any order that you want. For example:
- If you want to receive a file and immediately transform it
- Call a database, get some data, and bring it into your workflow.
- Do a transformation or adjust the line feed.
- Write it out to multiple destination systems.
You can put all those workflow steps in any order that you want. That makes it just like a piece of code which has been abstracted into a front end webpage. If you think about how your code would flow, that is exactly how you can make the software flow.
Other vendors that I have encountered in other jobs have been a lot more rigid than that. Other vendor software tends to have one canned way that you can run data through the system. You receive a file, maybe call a map and transform the file, and then you write the file out. At the very end of the process, you might be able to call a post-process, where you want to run a shell script. However, usually other software is pretty rigid in nature, such that you have to conform your data flow to how that vendor software works, because it only works one way. Because of that rigidity of other vendors' software, in order to accomplish a full end-to-end workflow, sometimes you have to spin up three different workflows and tie them all together to get all the different steps done that you want. That is not at all the case with SEEBURGER BIS.
With SEEBURGER BIS, you can string as many different activities together in your workflows as you want. You can put them in any order, like a piece of code. One leads into the next, which leads into the next. It is just very flexible from that vantage point. This makes it so easy to use and reduces the number of moving parts that you need to have. It is just a lot less frustrating not having to conform to how some other vendor software works.
How was the initial setup?
We had a variety of software products, so it wasn't a straightforward effort to consolidate all those different use cases and patterns down to their software, because all the existing ones were hard to discover. Sometimes, you experienced that the people who originally set them up were gone. There was a lot of work trying to understand the existing use cases in order to migrate them into SEEBURGER BIS.
With any large first time installation of a completely different vendor's product, I think there is going to be some pain. That has just been my experience. You are trying to understand how to fit their product into your custom network. It isn't a one size fits all, so you need to tweak and tune it to get it to fit right in your network. The types of machines and network that you use are usually custom by company. I wasn't around for the original installation, but from what I heard, there were some of those sort of pain points during the initial install. From people that I talked to who were here at the time, it sounded like a lot of it just had to do with getting the platforms that were in use (at that time) to be configured and work with SEEBURGER BIS properly. So, I don't know if it was necessarily that the install of their software was bad or hard to work with. I think it had as much to do with the specific systems that were being used here at that time, so tuning was needed to get it to work right for memory, storage, etc.
While there was some pain, I think it was equal parts their software compared to our systems and infrastructure and trying to pair the two together. At the same time, we were consolidating 12 or 13 different vendor products down into one. A lot of time went into understanding all those different use cases and how to properly configure them the first time and SEEBURGER BIS. So, there was just a lot of learning and discovery that went along with the initial install.
What about the implementation team?
The relationship started in late 2012 to early 2013. It didn't actually get put into production for usage until 2014, so there was a year to a year and a half of planning that went on between the staff at SEEBURGER and the staff at my organization, from the infrastructure and the integration teams, to try to lay out how it needed to be installed as well as what kind of machines were needed and how much memory. That was a long, drawn-out process. I wasn't here at the time, but I imagine it wasn't the highest priority here at the time. So, there was a good period of time when they were just in that discovery mode trying to understand what they wanted to buy, for example:
- What would we need to install it?
- What team was going to support it?
- Were we going to outsource it to a managed services provider?
- Were we going to hire a staff to do it internally?
All that type of stuff had to be worked out in advance, so there was a lot of planning that went into it.
They went with an outsource managed services provider at the start because they needed staff quickly to be able to start on the migration work to get all the integrations understood as well as out of the old software and into the new one. There is always this learning when you go full-on managed services, as you don't get high levels of expertise. You mostly get people who can turn stuff around quickly without thinking about it too much. Then, towards the end is when they hired a couple others and me to try to get some more senior staff in to steer the ship and standardize everything that had been configured so far as well as come up with design patterns. So, towards the tail end is when they hired some senior developers to try to get things under control and standardize use cases.
During the migration, because it was a managed services provider, there were a whole lot of resources involved during the big part of the migration, like maybe 20 people just cranking out and moving stuff from the old systems to the new system. As things got closer to being migrated, we got down to a team of four or five people. There was a bit of managed services assistance offshore to kind of help out off-hours, and that has not really changed a whole lot. There are four or five individuals who handle the bulk of operational aspects and support as well as the engineering aspects, so it is a pretty small team.
What was our ROI?
The two areas of ROI that stand out:
- User manual task automation because you're reducing manual time.
- Two or three pockets of different data integration patterns that were outsourced to vendors over the years for various reasons, so there has been a return on investment there to be able to in-house those integrations from a vendor and save on that vendor cost because you have to pay a vendor to do that integration work. Why not use your own software and do it in-house if you can?
What's my experience with pricing, setup cost, and licensing?
I have had exposure to other big vendors over the years and would have to say the pricing is pretty typical. They all fall into a common pricing range, at least the bigger vendors: Axway, IBM Sterling, Globalscape, and SEEBURGER. They all fall into that mid-tier pricing. So, SEEBURGER is commensurate with other large integration vendors operating in this space. Maybe it is lower than some of the really high-end ones. You can get some of these high-end transactional messaging integration systems, like TIBCO, that tend to be kind of on a higher echelon of pricing. I would say SEEBURGER is more mid-level.
Every vendor has professional services to offer. That is where they make a lot of their money, in PSO time. Different companies feel different ways about using professional services hours. Luckily, for us, our company has always been pretty open to it. We use that professional services time sparingly throughout the year: for critical key projects, things that we've never done before, or if we're doing a major system upgrade or a version upgrade. Those things have to be done right. Although you could probably figure it out on your own with enough time, you can usually do it faster with a professional services person in the mix. That would be the only other cost: If you choose to use some of their professional services labor as a bucket of time throughout the year, like we do.
As you spin up new components or use cases, occasionally more licensing is needed to turn on more features of the software suite, but that is common across all the vendors.
Which other solutions did I evaluate?
I have been in this field for 25 years. I have worked with a lot of the larger integration vendors over the years. Integration vendor software tends to be fairly similar in functionality. You can pretty readily move from one product to another product and not lose too many steps, as far as understanding and utilization, i.e., user experience. However, there are some things that set SEEBURGER BIS apart from the others. I don't want to go to another vendor software after using SEEBURGER BIS, because it has these "things" that make it that much better and easier to work with. It just makes my life as a developer so much easier.
A lot of the integration vendors have a software development kit that usually looks like an Eclipse plugin for Eclipse IDE. This allows you to code extensions to the base functionality of the software suite. If the software does X, Y, and Z, but you want to add A, B and C to the end of it, then you can build your own extension to the base code and plug it in. Most of the vendors have that these days. Comparing SEEBURGER to some of the other providers, SEEBURGER BIS requires the least amount of professional services know-how. For example, that PSO level of knowledge which is sometimes needed, where you get into your own development effort to write an extension, and halfway and you are like, "What the heck?" Then, you need to have the expense to get professional services involved to help you out. However, with SEEBURGER BIS, I have been able to code many of my own extensions using their software development kit on my own with very little insight from their professional services. It is very usable and user-friendly compared to some of the other solutions.
On some vendor products in order to find logging, especially if they have bolted together multiple vendor's products, you have to go explore here, there, and everywhere. With SEEBURGER BIS, it is all together. It is not hard to find the logging. It is all pretty readable too. You don't have to go through a lot of jargon to find what you are looking for. Probably the best thing about the logging is that you can't necessarily get logging down to the packet capture level in other vendors' products. So, if you're doing an SFTP or an HTTP connection and you want to capture the packets, I have needed to install things like Wireshark or Fiddler with other vendors' software to catch the packets going across the wire to see what is going on. In SEEBURGER BIS, you can turn on the packet capturing ability and just go look at it within the product itself. You don't have to waste time installing Wireshark and getting it all connected to your network because you can get that level of logging right out of the application.
I have been through migrations a lot over the years. This one was interesting to read about because they were crazy about it. The last vendor had some problems, so they went and actually started with a list of 50 integration vendors. You go, "Holy cow, are there really that many integration vendors even out there?" There are, but they range in size. So they started with this gigantic list that they probably pulled from some vendor, like Gartner. Then, they boil down 50 to 30 then to 15. Ultimately, after looking at all the use cases and what they wanted to get out of them, they boiled 15 down to three, then they had the last three come in-house and give their dog and pony shows along with their overviews of their software. Just based on the use cases and flexibility, SEEBURGER was chosen over all those different vendors. There is some pretty good amount of documentation that was written up on that process. It was pretty thorough.
Among the bigger vendors out there in the marketplace, IBM Sterling, Globalscape, Axway, and Cleo were some of the big vendors in the mix.
Deciding to go with SEEBURGER BIS was a mixture of the GUI and simplicity for the team to understand. A lot of the team here, other than just a couple of us, are operational-level folks. They don't necessarily have broad computer science foundational skills. Therefore, the GUI interface had to be easy enough to use but these types of folks could work in it without too much trouble. The big things were:
- Ease of use in understanding the screens and how to tie together a workflow without too much developer-level knowledge.
- The capability to handle use cases. All these different 12 or 13 vendors were doing different data flow patterns, so SEEBURGER BIS had to be able to support all of them.
- Stability, because the previous vendor had problems with it. That had been a pain point which they were trying to solve at the time.
So, it was really a mixture of those three things.
What other advice do I have?
Going into the next calendar year, we are going to migrate to some formulation of cloud, which is kind of the way everybody's going, and then we will also be migrating to version 6.7.
I work a lot in our integration for error/fault handling and reporting system metrics, making sure all the components are running, raising incidents automatically if there is a failure, and raising incidents if there is a problem with the software. Because I sort of operate in that monitoring space, I was hoping that monitoring would be easier and better in the new version. They have a new component that they have added, which has changed names a few times. This will allow us to do different kinds of file transfer, failure, and error types in any software. There are lots of different points of failure that you catch when you do monitoring, e.g., conductivity failure or transformation error. This tool will take the error logging for every kind of error and standardizes it into one stream of data into one software component. In the past, I had to read error log messages in a map and pull out very specific error reasons and populate them into the auto-generated incidents.
In version 6.7, all that code that I wrote to map out those error codes, which are all different for each type of error, is all standardized and common. It is already in the software components. You don't have to do extra code to parse out and find those different kinds of errors and metadata related to each type of error. This is because it has already been standardized and put into the GUI. The GUI can also integrate with ServiceNow and other ITSM systems to auto-generate incidents. Therefore, a lot of that behind the scenes monitoring code that I have written in the past can go away because now it has all been sort of consolidated down and standardized by SEEBURGER. I am really hoping to be able to get the funding to purchase this extra component going into version 6.7, so we can move away from our sort of homegrown code into out-of-the-box monitoring.
We are always looking for new use cases in the broader integration space to try to bring in and automate what our team is responsible for. There are some other parts of the company that have historically outsourced different pieces of integration. So, we are in the process of in-housing one of those now, and then there is another one coming down the road that is sort of more into the SWIFT area. Banking has this concept of SWIFT, which is an integration to the banking network. SEEBURGER BIS has some out-of-the-box functionality to offer there, which is another space where we are looking to expand services.
We use it mostly for behind the scenes data flows. Other than a few cases, we don't necessarily serve up screens with metadata about the data that is flowing through the system to facilitate use cases for data insights. This is an area where this new version 6.7 also has more functionality. It can present more data for what is flowing through the system, bringing the data and important parts of that data up into the GUI so you could reach out to a business team, and say, "Hey, I know you probably use some other product for your day-to-day operations. But, what if we could serve up this data that you care about on a screen? Would you no longer need that other software that you are currently using? Because then you can just log into SEEBURGER BIS and see the data there."
The solution’s ability to future-proof our business is positive because I have a pretty good sense for what is out there in our organization that we are currently not involved in as well as what we probably could be or should be involved in based on what our responsibilities are to the organization. As I look at other areas of things that we are currently not doing, then I look at SEEBURGER BIS, and say, "Can it do that stuff?" The answer has always been, "Yes." You might have to buy some more licensing, but every vendor software is that way. They don't work for free. So, I would say that the feature functionality is very broad, such that any other area of data integration that we have come across in the company that we are not supporting, which is being supported manually or by some other team using some other software, we then look at SEEBURGER BIS and we have always found the functionality there, even if we had to buy an additional license or something to turn on a new feature.
Biggest lesson learnt:I have learned a ton about APIs that I didn't previously know. The software helped facilitate that knowledge because the functionality was there and I had decided to figure out how to use it. However, in figuring out how to use it, I learned a lot about how APIs work. That has been probably my biggest personal area of learning in the last one to two years: Being immersed in APIs and the file transfer space, learning all the different SOAP versus REST, and calling service operations and methods. All of that, I learned by using SEEBURGER BIS to set up API integrations of various flavors.
I would rate this solution as 10 out of 10.
Which deployment model are you using for this solution?
Which version of this solution are you currently using?
6.5.2 SP 1.0.2