The following interview on the subject of how to build a business case for ERP in today's economy took place between Ray Wang, Partner for Enterprise Strategy at Altimeter Group, and Brenda Cole, assistant site editor of SearchManufacturingERP.com.
Ray Wang is the author of the enterprise software blog "A Software Insider's Point of View". The blog focuses on enterprise apps strategy, vendor selection, software contract negotiations, and emerging business and technology trends. As an enterprise strategist, Ray focuses on bridging the gap between today's enterprise landscape with an emerging class of enterprise business solutions adopting the spirit of social technologies and Enterprise 2.0 concepts. Research topics often include ERP, CRM, project-based solutions, order management, master data management (MDM) and Software as a Service (SaaS).
SearchManufacturingERP.com: With the economy the way it is, the last thing most companies want to do is put money into unnecessary new software. Before any decisions are made, what questions should be asked to determine whether ERP is really the right move for your company?
Wang: Everything begins with the right business drivers. If you think about when you're in tight economic times, the focus should be on building business cases, allowing operational efficiency or meeting regulatory compliance. Now that sounds simple in concept, so let's put something into use. For example, if you can identify a commoditized process from an inundative process, you're actually in a good position. Now what's a commoditized process? That's something that can be repeated, automated, something that isn't adding much differentiation or value in your business.
Once you do that, you can do things like apply technology strategies such as ERP, put that into a shared service, tie that back in with an upgrade strategy, consolidate instances if you've got different versions and think about helping consolidation and even environmental sustainability. After that, you've got a business case in place where you can not only build upfront savings but ongoing savings, and then you can think about which is the right vendor to work with to achieve those savings.
SearchManufacturingERP.com: What about the Software as a Service option? How do you decide between implementing an in-house ERP system and using SaaS?
Wang: SaaS plays a significant role. For companies with existing ERP implementations, we expect hybrid deployment options like SaaS to provide purpose-built solutions or innovations where many ERP vendors cannot or have not made adequate investments. You're not going to rip out your financials systems or some of your other core manufacturing systems for a SaaS option, but you want innovation. How are you going to bring that into your staid ERP environment? Well, some examples include functionality such as projects or project-based solutions, incentive compensation, talent management, e-recruitment and donor management.
In many of our studies, we show that over a 10-year period SaaS vs. on-premise makes sense for smaller enterprises -- those with 50 users, 100 users, even up to 250 users. Once you get past 500 users, it's more of a lifestyle decision for larger enterprises; you do break even, but you have to trade off some things. For example, SaaS makes sense because of the pricing model and rapid implementation, but for clients who require heavy implementation, support for high-volume transaction models or data security location issues -- the data can't leave the state or country or needs to be in a certain location -- this may not be the right option. You've got to balance the business requirements with what you want to achieve with SaaS.
SearchManufacturingERP.com: Open source software is growing in popularity. When would using an open source ERP package make sense?
Wang: There have been a lot of improvements in open source. There are some vendors, such as Compiere and Openbravo, that have done a good job in building ecosystems. Primarily, the source code is free; that's what makes open source attractive. The value of these vendors comes from their ability to provide support, maintenance and extensions. If you've got the technical know-how to self-deploy and you find the right support groups or capabilities to be adequate, then this is a realistic option.
We're seeing transaction across a variety of industries, from large public sector environments and academia to manufacturing and service-based industries. The trick is to think about your ability to self-support and the appetite you have for a scenario where you have no one to blame for the code other than what's in open source. Open source isn't that scary, quite frankly. I'm using a blog that's open source delivered and many companies are using open source products such as Linux. I think open source ERP is an evolving area and one of great interest, especially given the current economic environment.
SearchManufacturingERP.com: What about companies that installed ERP years ago, but are considering an upgrade? Would it be more cost-effective to install a whole new system, or should businesses consider refurbishing an old ERP system?
Wang: We get this question quite often. Once you factor cost, benefits, flexibility and risk -- basically what we're calling the total economic impact of an upgrade -- if you take customers with heavy customization or integration to legacy systems, it's typically going to be cheaper to redeploy than to upgrade. The key is to evaluate the business processes that need to be supported and tie those back to the user roles. Are you able to achieve your business objectives with ease? If not, then you should start from scratch. If you can and you can keep what you have, then definitely consider the upgrade. Think about what areas you need to upgrade. There are four types of upgrades you may need to consider. Some are technical upgrades, while others are something that's much simpler, involving patches or areas that you want to move in modules. Some are complete system upgrades and others are just complete replacements.
SearchManufacturingERP.com: Let's talk about contract negotiations. What should be taken into consideration while negotiating a contract with a vendor? How long do software contracts generally last?
Wang: First of all, the license in most cases is perpetual. The contracts typical cover support and maintenance and last for three to four years, depending on the vendor. Key considerations are business drivers, technology adoption strategy or how quickly you're going to adopt the product and budget. The first step is to make sure you don't over-license for software you don't need. This concept called shelfware -- software bought but not deployed -- is really expensive, because you pay maintenance regardless of implementation. If you get a great deal on 1,000 or 10,000 licenses of software and you only deploy a third of that license, what you end of doing is paying for maintenance and, at 20-25% of the net license cost, it's the equivalent of buying new software every four or five years. It's very important to make sure you sign the licenses correctly upfront. The others areas where you should focus follow what we call the software ownership lifecycle. This includes selection, implementation, utilization, maintenance and retirement of software.
SearchManufacturingERP.com: Usability is a hot topic these days. How have ERP vendors improved the user experience of their products in recent years? Do you think usability plays a part in an ERP system's ROI?
Wang: Most definitely. This is an area that's been quite exciting in the last two to three years. We've seen some massive improvements in user experience and usability, and what this does is lower training costs, improves productivity and provides business flexibility. We are seeing a significant movement to improve usability to meet enterprise principles such as a richer user experience -- role-based scenarios across barriers, usability paradigms -- so whether it's desktop, mobility, kiosk or device, we're seeing those experiences carry over. We're also seeing business process orientation. This is support for end-to-end business processes across these usability paradigms, and in some cases they're tying training to some of these business processes so you can do real-time training while people learn on the job.
Third is configurable change: designing with flexible models and rules instead of customizations. This means the business person can make a change; you don't need a developer to come in and configure something that can be done at a process or meta level. Fourth is actionable insight. While upgrading, we're seeing a lot of incorporation of key information to roles so people can make decisions at the lowest level and in the context of business processes and the user's role. Fifth of all is collaboration. We're looking at how people provide secure, private interactions and open and innovative connections with stakeholders.
Then there's intelligence response. In terms of usability, it's responding to contextual models and business events and the ability to pool information in areas that make sense and set thresholds. The last piece that's important is hybrid deployment, which means being able to support different models of delivering that software, whether it's on-premise, hosted, instance virtualization, multi-tenant SaaS or even cloud-based business process outsources. If you check out my blog at http://blog.softwareinsider.org, you can see highlights from Epicor, IFS, Lawson and Microsoft Dynamics. We've captured some of the elements of this new user interface (UI), which is really powerful with many great paradigms and definitely an improvement in terms of what we've seen in terms of boring, staid ERP UI and user experiences in the past.
SearchManufacturingERP.com: We hear stories about ERP implementations running horribly over schedule and over budget. Is there a way to guarantee that won't happen to your deployment?
Wang: There's a three-legged stool required for success. First, clients need to be realistic about their business requirements and they've got to keep within scope -- they can't keep changing the requirements. Secondly, the vendor needs to honest about what is supported out of the box. You've got to have a good discussion and think about the demo scenarios that are "vanilla." And third, the systems integrators need to guide the clients to make the most cost-effective decisions and keep them in scope. All three are to be commended for success and blamed for failure.
So when you see these large ERP projects -- "$50 million wasted," or "50% of all ERP projects fail" -- a lot of it comes back to making sure you have good design and you understand the objectives. What I say is, begin with the end in mind. What do people want? Think about the reports that are important and tie them back to those business processes. Make sure that they fall back to the right users, so that the data is captured correctly the first time and you design the product to be used correctly. This is really about access to information and being able to make informed decisions or automated transactions, which means reducing the steps in a business process.
SearchManufacturingERP.com: Can you offer any tips for saving money during the ERP selection process?
Wang: This is an area we've spent a lot of time with; during my time at Forrester, I must have done a thousand of these kinds of spender selection and contract negotiations. They key thing is to engage with more than two vendors and focus the selection on business processes and use case demos. Unless you're a public sector entity, throw out the request for proposal (RFP); it's actually not very useful. We don't know many vendors who come back and say "Oh yeah, we don't do that." The fact is everybody claims to do something. Not that they're lying, but it's either in future plans or future products.
The best thing to do is to set realistic expectations with the vendors and the internal selection team. Get these demos going and see what's possible with the software. Work with a third-party advisor who's got the main expertise and can guide you on both technology adoption and contract strategy. That way, you can reduce the time it takes to do these vendor selections. We've seen things done as quickly as a month, where people build use case scenarios, they bring the team in for demos, they run through the rigors of the demo and they really go out and build something. This is all part of building a business case.