What needs improvement with GEP SMART?

Before introducing GEP SMART, we were using all the supplier databases in Oracle Financials. When we implemented GEP SMART, we wanted to do the supplier evaluation online on a real-time basis. We just wanted to give the link, but we faced some problems with the synchronization from GEP SMART to Oracle. Even though we have GEP SMART for the supplier registration, we use Oracle for our delegation of approval (DOA) processes. The DOA process is happening in Oracle Financials. Therefore, each time we register a supplier, it needs to have approval from finance and the business unit as well as the supply chain. Because we pushed the existing supplier database to the newly installed GEP SMART process, we have faced some problems with this function. While it captures all the fields and requirements, when any supplier updates or new supplier registration happens, then the supplier needs to update their bank details. Our IT team has made it so every half an hour the GEP SMART system will push the supplier to the Oracle Database. So, we have faced some problems, like bank details, etc. We addressed this issue with GEP. They are working on it to make bank details a mandatory field for any supplier and match our requirements. Their engineering team has taken up creating a customized process for us. We are planning to do the supplier performance evaluation this year, for which they have a module. Last week, we were having a discussion with GEP, and they had sent me a document, but I had to go through it in detail about how it would function in GEP SMART.

We are delighted with the product thus far. Based on our business, I see a continued need to leverage some of the capabilities further. Analytics has been the main driver for us. I see constant improvement. They released an update in January 2020 or maybe fall 2019. I want to see continued investment in analytics and reporting.

There is a level of complexity when connecting suppliers to the platform, which is something that could be improved. In terms of preparing the first RFP, certain documents need to be uploaded. Uploading this content should be easier and more user-friendly. The support for additional languages should be improved. In particular, for some of the additional languages, or key languages, the tool works well. However, there are others for which it could be better. The manuals, especially the ones for the suppliers, are currently available only in English. If these can also be translated into different languages, that would be excellent.

On the spend side, we had some difficulty with the usability, initially, but then they rolled out SMART and they built out a new spend cube, and that was light-years better. Part of the reason I hadn't rolled it out fully to the team was because it wasn't as user-friendly as I would have liked. But they addressed that in a newer version last year. I rolled it out to a subset of my team earlier this year. It was almost a proof of concept type, phase-one rollout, and it actually went very well. We plan on doing a full-scale rollout training in 2020 for the rest of my category teams. Everybody will be running their own spend reports and using this to manage their businesses. I'd like to see drag-and-drop reporting. They have the old model for reports where you have to click the "run" button. The thing runs and then you have to export it to PowerPoint. If you're doing a presentation, you have to export it out as Excel, and then you have to go through all this stuff. There is a concept called portlets, which are like an app or a window within a window. If they had a page with four different portlets on it and four quadrants, then each one would be independent and you could run and filter down each individual portlet in each quadrant. That would be beautiful. If you wanted a nice view that has spend data from a particular business unit or a particular region, you could do all of those individual filters on one page as opposed to having to export it to Excel and run four different reports. That's a big one for us. There is some stuff related to RFP on their roadmap, like the ability to pause an RFP. It could be that you're running an RFP but the business changes; you acquire a company, or the leadership or initiative changes. Instead of canceling and then reissuing, you may want to pause it. That's something we brought to their attention. That's something that may be on their roadmap. They have a track record of making changes and implementing those updates, so I'm sure they'll address that.

I don't think their supplier management module is ready, from a full supply-relationship management perspective. The scope is very limited for supplier management. Also, in the contract module, the searching capability is sub-par. Most people are used to Google search and Amazon search. That's what's available on a day-to-day basis. GEP SMART's searching capability is extremely difficult to use. Their logic is different from Google and Amazon, so they return a lot of search results, which is something we're not really happy about. Another issue that I want to provide feedback on is that their icons are not as visible and as accessible as they should be. If we have someone with some form of disability, it's not easy for them to figure things out. When you hover an icon a little message pops up and tells you what it is, but the icons are so small and look very similar. It's a design element that looks great but it's a little hard to hover and makes for an accessibility issue. But overall, in general, it is intuitive to use. Also, I know that GEP, as a company, has been in business for close to 20 years, but they still operate as more of a startup and smaller company. They need to revamp their processes and put in controls in terms of quality assurance and quality control. In terms of their processes, there are no mechanisms for them to check up on how they refresh their data. So oftentimes, when we go into the tool and we don't see something, we alert them and they say, "Oh, the data hasn't been refreshed, so we'll go ahead and refresh it," and then what we're looking for pops up. We should not have to be the ones to tell them they need to refresh. There should be a mechanism in place for that. They're a technology company and their product is where we put our data. The refresh cycle should be automatic. As for quality control, I don't know if they do enough testing. When they release new enhancements, they do testing and it passes and that's why they push it to production. But it seems they only test that particular functionality and that they don't test how that functionality interacts with and impacts other functionalities. So that particular functionality itself is working, but oftentimes functionalities are interdependent and when we try to click on certain things which should behave in certain ways or bring us certain results, they don't work. We have to tell them, "Hey, it's not working. You need to fix this." Quite frankly, we're tired of informing GEP. They should be scrubbing from the left and right and from top to bottom to make sure that anything that they're releasing is fully tested. I get it, that one or two fall through the cracks. But it happens consistently that we're concerned with the data integrity because a refresh cycle didn't happen or there's a stability issue where something is working but when coupled with other functionalities it fails and we get an error message. Those are some of the things that the GEP is lacking. If they do want to compete in this market with other, bigger players, they need to up that game. GEP, as a company, is putting so much focus on artificial intelligence and machine learning. That's great. We love it. And that's where the industry is moving. We like seeing those on their roadmap. However, foundationally, they need to fix some of their processes before they bring in the above-and-beyond bells and whistles. If their basic functionalities are not functioning the way that they're supposed to and they keep building on a broken foundation and adding more, eventually it will collapse. It will become too top-heavy. This has been a message that we've been sending to GEP over the course of the last two years. I understand that to keep up with the industry, they need to bring in some of the newer technologies, a newer perspective. That's a business decision. But they still need to go back to their core technology and really enhance that, so that when they add something it really couples with that core and enhances their technology overall. In addition, from our perspective, the way that they manage their technology is very fragmented. I mentioned that we use four different modules of GEP SMART but they were built in silos. Although they actually promote this or sell it as a fully-integrated solution, the way that we search in the sourcing module is different from the way we search in the contract module. Why are they different? It should be consistent. Another example is that the supplier ID in the supplier module is different from the supplier ID in the contract and sourcing modules, although it's the same supplier. Even though we enter and register our supplier in the supplier module, when they float through to the contract module and sourcing module, they generate a different supplier ID in the backend, so it's extremely difficult to trace back to see if it's the same supplier. If it's stemming from one module, it should be duplicated across all modules, as opposed to creating another copy of it. That confirms that they're building their technology in a very fragmented way. That needs to be addressed. We have a huge number of duplicate suppliers created. When someone facilitates a sourcing event, they enter suppliers' names differently. A classic example is IBM, which can be entered as "IBM Corporation", "IBM Corp" or just "IBM". Sometimes it's entered as "International Business Machines." GEP has been trying to work with us to leverage machine learning and AI to prevent people from creating duplicate suppliers. It hasn't been fully built-out. And there still needs to be some manual intervention. The human element needs to be there. Once it's fully built-out, it might help us. One of the things they proposed was that machine learning and AI would pop up a window that says, "Hey, we have all these similar company names." Based on what was entered — not just the supplier name but the contact information and addresses — it would say, "This company is about 80 percent matching to what you entered, so why don't you use this one instead of creating another one." There is stuff coming, but it's still not there. We're testing it out right now. We're hopeful that it will at least provide some guidelines for our folks, and at least cause them to pause a moment before they create another supplier.

The loading of the contracts could be a little bit easier. I'd like to see a little bit more of AI brought in, to the point that it's actually reading the contract and automating some things, like expiration dates and renewal terms. That is an area for improvement. There is also some stuff coming that we've been asking for that should be in the next release, in terms of some automated reminders in the sourcing process. And we've also asked to be able to do a little bit more visualization of responses in terms of pricing and completeness for RFPs. Those things are coming some time next quarter and we're looking forward to them.

One of the areas where they could do better is by creating more standard templates when it comes to IT cost. IT cost has various components to it and if I want a vendor to be able to reply about cost related to IT, I would have to create that template. If I want them to tell me the hardware costs, the software costs, the professional services costs, the cloud or subscription costs, etc., I would have to build that template myself. Once I built it, I could save it and use it again and again. But sometimes building a template takes a lot of time. In addition, what we do now is write out a Word document with our requirements and submit those requirements by attaching that document to the information that goes out to the vendor. Similarly, we send a template out with it and say, "Please respond in this format." That gets attached as well and then the vendor downloads that information and creates a proposal offline in the template that I've sent. They then reattach that information to send it back to me. That's fine and it's effective. But I wonder if there is a different way where we could physically write in the requirements, rather than having to create a separate document and then attach that document. And similarly, have responses back without being put in a Word or an Excel document. On the other hand, as I think about that, there are advantages to having the separate document because you can download it and go off and read it. You don't have to be connected to the network to do that. Similarly, the vendors can type it up offline and then attach it. So I don't know if that's really something that they can move on.

The difficulty we had when we first started was that we had to migrate all of our records from our previous tool which is a program called Ntracts, a contract tool. We were only using it as a repository. They have upgraded it where it's supposed to be more of a lifecycle tool, like SMART is, but I've not heard good things about it. GEP had never migrated anything from Ntracts before, and the information within Ntracts was not compatible with the information that was needed for SMART. Then, they put somebody new on the migration but it was a huge process. We had 50,000-plus contracts in Ntracts that had to be migrated to SMART. It was a big ask and it went horribly wrong, and we ended up having to do a second migration. We spent a year trying to migrate, and then we had to just tank it and start over. Needless to say, it was not well-received. It put a bad taste in peoples' mouths at first but we're past that. We got the migration done and we have happy people. Some of the rough spots in it have more to do with the things that we want and require within the tool, rather than the tool itself. Some of the forms or questions that we've asked them to incorporate and customize for us are the things that our users sometimes struggle with. It's more what we're asking them to provide, versus how to use the tool. We definitely have some asks for enhancements. One of the big ones we'd like to see is what we call "drag-and-drop." If I have an email that I want to maintain in the SMART record for whatever reason — for example, it has supporting information in it — I would love to be able to just drop that into the notes and attachments section without having to save it as a PDF and then upload it. And that would be helpful if I have a pre-signed document rather than having it e-signed in SMART. There are times when we have to have it signed outside of SMART. I would like to be able to just drag-and-drop that from my computer, rather than having to upload it. It would save three steps. We have a lot of asks that we've found over the last two years of using this. There are things that we have felt, "Oh my God, this would make so much more sense. Why isn't it like this?" I've got a pet peeve. They have different terms that they use throughout the systems. For example, on the cover sheets that we use, under Basic Terms, they might refer to the author as the "contract administrator." But if I go into the system to run a report, and I'm the contract administrator who is the author, if I want to run a report about every record that I've worked on, it's called "author" in that section. It's really confusing, although not so much for me, because I'm in it every single day. But for other people who don't, it's confusing. "Contract type" means, in some areas, the coordinating area that's working on it. But in other areas it really means, what is this contract. But sometimes that's referred to as "document type." It's confusing and inconsistent. We've talked to them and they're aware of it. To me, that's low-hanging fruit. Fix it. It's just being sloppy. It was a problem in 1.0. When they went to 2.0, it became worse because they changed terms on the cover sheet. So now, not only are they still different, they're new. So everything we got used to in 1.0 — even though it was not consistent throughout, we were starting to get used to it — was changed in 2.0. And it's still not consistent.

A lot of the things in the system are not client-facing. So we weren't able to edit certain types of master data, and we're relying on them to edit it for us. I assume it's partly how it's designed and that it's also a safety net that we're not able to essentially ruin our installation. So it's understandable. But there are some components that we would like to have a little more control over. And there are things regarding how the process for procure-to-pay works that differ slightly from how we do things, but that's expected with an out-of-box-solution. As an organization, GEP is very technically capable. Very recently, they implemented a customer success team to manage our expectations and communicate them to their technical team. That function is relatively new and some work needs to be done to build that connection so that it's a little more seamless. They need to be managing our requests for enhancements and our requests for fixes with their engineering team and getting anything that needs to be fixed, fixed.

Their contracts module is kind of clunky and It took a while for them to correct some of the basic functionality, some of the "Contract Management 101" functions, but it seems to be coming around. It wasn't working the way we'd expected. In terms of additional functionality, most of what we'd like are on the roadmap, like bid optimization functionality. Also, some of the modules don't have the same user interface as the others. We'd like to see them all made uniform.

We didn't like their dashboard initially, but they responded to that very well. They've given us some customizable dashboards and have also made it so that the dashboards can be exported into PDF and other formats, so that we can share them with the rest of the company, people who are not necessarily users on GEP. That was a weakness at the beginning, but one that they have responded to adequately and we're really pleased with the result.

I do not know how they could get better in spend. That's a pretty great module. We could probably take advantage of some of their available wizards and develop some intake efficiencies for the process so that it's not quite as much just data entry and creating certain profiles. We have not gone into the 2.0 version. The whole dynamic and interface between the modules may be improved or it may be the same thing that just looks a little bit different. There are certain things within the contracts module — how to upload the line items, for example — that could be done to make that utility a little more user-friendly, a little more like the sourcing module. I'm interested to see how that might work.

On the contract side, we have definitely come across a lot of pain points since I've been here. There were some issues with our initial implementation. It wasn't done correctly and it's been a process over a few years to recover from that. There were a lot of lessons learned on their side and our side, and there are still things that we're trying to work through that, maybe, weren't understood properly in the beginning. We're still continuing to try to build it for what we use it for, which is different than what some of their bigger clients may use it for. We've had to do a lot of cleanup and make a lot of changes. We had a lot of challenges and disagreements with SMART. It's been a long road, for sure, on the contract side. There is a little bit of pushback on their part when we need stuff done. Things aren't done very efficiently. I'm still waiting on some changes that were requested well over a year-and-a-half ago. These certain items have been bumped up to the president of the company. In addition, there is a lot of information that we have to put in that is not useful for us, but we have no control over that because it's hard-coded into the program. There's a lot of stuff there that we just don't need or use. It would be better if we were able to turn off all the things that we don't need. The way it is right now makes things seem unclean and not tidy because there's all this information we have to put in that we don't even use. Being able to turn off tabs and fields that other SMART clients use but we don't would be nice, just to simplify it and not have to see them or fill them in.