With many companies loosening the IT purse strings, enterprise IT projects are on the rise. According to the latest forecast from analyst firm Gartner Inc., spending on enterprise software worldwide is expected to increase by almost $20 billion this year to $255 billion. Even more pronounced is the increase in spending on IT services, expected to top out at $824 billion. Every company that sets off on a new enterprise deployment wants its project to go off without a hitch — but that doesn’t always happen. To find out what aspects of an ERP project bring the most risk for sending a project off track, we recently talked with Pat Phelan, a research vice president at Gartner Inc. with expertise in ERP implementations and project risk management. Following are the highlights of that conversation. Do you see ERP projects increasing in 2011? Yes, we’re seeing a lot of interest in ERP projects right now. The budget seems to be there, and we’re seeing some fairly substantial projects, both globally as well as in the US. I wouldn’t call it a renewed interest, because I don’t think the interest ever really waned, but we’re definitely seeing more budget being allocated to ERP projects. In which areas of ERP implementations do companies most often shortcut their due diligence? There are several areas I’d point to. First is the amount of time and budget the companies dedicate to implementation services. Companies spend a significant amount for the software itself, and then go on to spend from three to 10 times that amount on implementation services. But often they don’t spend enough time selecting the right systems integrator or don’t budget properly for the services involved. Even if you have the best software in the world, if you have the wrong team put together to implement what you bought, or underestimate how much the implementation process may cost, the project will suffer. Along with that, if you don’t have the right project manager or simply don’t devote enough time and budget to the project management aspect, problems will eventually arise. Another common reason projects go off-track is lack of sponsorship — there is no key business leader responsible for championing or sponsoring the project. That filters all the way through the organization and colors peoples’ perspectives in terms of being willing to work on or with the project. In terms of budgeting for services, how does a company know how much to budget for that services piece? There’s really no right answer to that because it depends on the application you bought and how well it fits into your requirements. But you should assume that the services will cost at least three times what you paid for the software and it may go up from there. Are companies getting better about building project teams that balance business and IT requirements? They are starting to. We actually did a survey recently on companies we thought were doing a good job in this area and we found about half of the roles on project teams at those companies were filled by business people, and the other half were IT professionals. That’s different than project teams of the past that were heavily IT-oriented or had a lot of custom development roles on them but few business people. Do companies with a formal project management office or staff have more success at implementations? I wouldn’t say having an enterprise project management office is the key driver to success. An ERP implementation is an aberration, much bigger than most projects. So the key is the management of that project and the discipline put in place for that ERP initiative specifically, regardless of where it comes from. You may not have formal project management as a competency, but for this project you’ve implemented something formal and that’s the key to implementation success. Do you see more companies using third-party independent project managers in their ERP projects? I see that mostly in public-sector implementations. That’s not to say it’s the only place to do it, but they tend to need an independent third-party more so than in the private sector. It can be very effective, but you have to have the right person in that role. It’s not an easy job, it’s a thankless job. There are some consulting firms that specialize in it, but often I see a single person in that role, an expert in project management. It doesn’t really need to be a deep-benched consultancy to staff that role. It can be an independent individual. Is there any difference in getting a sponsor for a public-sector project versus a private-sector one? The challenge with a public-sector project is you are more likely to lose your senior sponsor because of the turnover with elected officials. So you need to recruit from both the elected level and the ongoing level for project sponsorship to cover your bases. And if you can’t get that consistent sponsorship, that has to be addressed in your project constraints — what you will do if you lose that project sponsorship. When it comes to project scope and dividing the work up between the services partners and the company, are there tasks that tend to fall through the cracks? Yes, there are two areas where organizations may not get what they thought they were getting from their implementation services partner: interfaces and data conversions. Systems integrators consistently “underscope” these areas, so the company thinks the systems integrator is taking care of them, but, according to the statement of work, the systems integrator is not responsible for it. Or, in some cases, some standard estimating techniques are used to put numbers into the work plan for those two areas without accurately considering the volume of the work or complexity of the specific project. Every project is different and some have more interface and data conversion work than others. Are companies getting better at training and tracking ERP adoption and use post-implementation? Companies that are focused on getting more value out of their ERP investment are treating training as an ongoing investment instead of a one-time cost. And that has to be built into the budget somewhere. We recently interviewed 10 companies about their post-implementation approach and every single one of them wished they had done a better job of planning for training after go-live. None of them had budgeted properly or built a proper post-implementation training structure. Who should be in charge of that training post-implementation? If you have a strong training contingent that is formally structured at the enterprise level, possibly in the HR organization, the best practice is to leverage the infrastructure that’s already there and not try to recreate it within the support organization. However, if you don’t have that training infrastructure in place, we recommend making training a part of your support organization. At least create it there and in time it may grow into an enterprise-level group. If you have a problem in training, it’s going to come back to the support organization eventually, so it makes sense to seat it there. We don’t recommend distribution of training to individual groups in the end-user community in an unmanaged manner. For example, making a super user or a single person in each department responsible for training creates inconsistent training results, delivers inconsistent training messages to the end users, and, because nobody is monitoring training effectiveness, there’s no quality control or assurance that the right training is taking place at the right time. The worst approach is to throw together a “train the trainer” program after go-live and expect those individuals to be able to keep up with the training required for such a major initiative. Being a super user doesn’t necessarily make someone a good trainer. They have been hired for a functional role, not to be an IT trainer. But the reality is, many companies will not be able to build an entire training organization. So if you must go the super-user route for training, the key to success is putting somebody in charge of maintaining that network of super users to monitor that they are delivering consistent messages and training correctly. You need someone to position those super user/trainers for success. Somebody has to own that job. And as always, what gets measured gets done. We recommend that you make sure the job description for that position includes measurements for the training aspects of their job. And their annual performance review must include being measured on their effectiveness as a trainer. Do companies have such training effectiveness measurements available? Most do not. So the first step is getting those measurements in place early. Many companies really struggle in this area. Do you see more companies outsourcing training in these areas? For example, hiring expert training firms for ERP training? I’m not sure if it’s increasing, but, on average, about two-thirds of the firms we talk to do use some sort of external training assistance. There are definitely more options out there to automate and improve how training is crafted, managed, and delivered. Where are companies most likely to see the ROI from an ERP implementation? The return on an ERP implementation does not come from the same area for every company. It depends on what’s most important for the company. For one company it could be order entries, for another it could be an inventory management issue. So it’s hard to pinpoint where the return will come on their ERP investment. It depends on where the pain points and revenue streams are for a specific company.