HP World '99 and ERP World '99 Conference Proceedings

ERP Vendor Care and Feeding

Mark Ripma

Menlo Park Research, Inc.

101 Yale Road, Suite 400

Menlo Park, CA 94025

Tel: 650-888-2330

Fax: 650-327-3432

mripma@mpresearch.com

www.mpresearch.com

Introduction

This presentation describes ways to plan and execute a software implementation project to ensure that all parties have a chance to perform at their best. It is based on my many years of experience as both a vendor and a user of enterprise-wide software such as ERP, Sales Force Automation, Customer Interaction Systems and Internet Services. In addition to my own experience I drew upon my interviews with service and marketing executives from ERP vendors.

The steps to a successful partnership, as describe here, apply equally well to the purchase and implementation of any enterprise-wide software system, not just to ERP.

Every vendor of every enterprise application has some customers who thrive on their relationship. These customers are groundbreakers, constantly pushing the system's functionality and the vendor's expertise. They are the first to encounter problems, but also the first to realize the benefits of the system. Working together, they and their vendors achieve mutual success.

Likewise, there are those who use inevitable shortcomings in their systems to delay the implementation. While demanding perfection is a legitimate position, waiting for perfection is forever. These companies do not achieve the benefits from the time and money they've spent.

If you want to be a groundbreaker, or at least get your money's worth, I have 9 suggestions.

1. Decide to Look for a System

If you're not sure if your current manufacturing system will do the job for you in the future, then it probably will.

It's like buying a car: if you're planning a long trip over treacherous terrain, you might want to trade the Geo in for a Jeep. If you're not planning on going anywhere special, just the same commute, keep the old car and save your money.

The point is, don't spend your or your vendors' time looking for a new system unless you're planning to do something that you can't do with your old one. If you're merely curious, go to shows, to presentations like this and talk to colleagues to keep yourself informed.

When you have decided that something different is required, put together your preliminary objectives and some budget numbers before calling a vendor meeting.

2. Define the Purchase Process

I'm a firm believer that anything worth doing in business is worth doing quickly. This means aggressive time-tables, delegation of authority and a minimum of bureaucracy. If your company has the time for endless meetings on any topic, even one as important as a new enterprise-wide application, you've got too many people or not enough business to keep them busy. These problems reflect a more serious management need than anything caused by an outdated or incomplete manufacturing system.

Buying an enterprise system is not like buying a wine, where the tasting process is supposed to be enjoyable. If someone is enjoying the search for a new system, they may have their priorities wrong and the consequence will be wasted time and energy for all concerned.

Fine tune your objectives before meeting with the vendors. Write down the specific benefits of the new system and your timing and the criteria for selection. These should be objective criteria and measurable benefits.

The criteria will be used to prune the vendor list quickly (cost, platform, products, features) so that you can review just a small number of viable alternatives

The benefits will be repeated endlessly, in every meeting; have them become your mantra. If you're not specific about the benefits, then people will have to rely on memory of why the system is going to help them. Memory, particularly in the face of trouble (and there will be trouble), is notoriously fickle.

If you're even considering "rolling your own" product, then you should. You and your new system will deserve each other (i.e. let the professionals do it).

3. The Project Team

Before contacting a vendor, put your project leader and project team in place. I strongly suggest that the project leader and the project team for the system selection be exactly the same as for the actual implementation. Ideally these people will be involved in using the product after the implementation as well.

If the project leader is not recognized as one of the best people in your company, get someone else. If you're not prepared to give the project leader authority at a level on par with the senior executives in your company, get someone else. The project leader should be full time unless it is a very small operation. It can come in handy later on if the project leader has a boss. For this reason I don't recommend the company President as the project leader.

In most system purchases there's a business buyer, an executive who needs the system to meet his or her business goals. Depending on the company it may be the President or a Vice President or a manager who's making things happen. This person is key to selling the idea across the company.

Occasionally there's someone I refer to as the "executive meddler." It's been said that the higher you go in a company, the thinner the air. The resulting light-headedness can lead to heavy- handed tactics to derail a cross-functional system. This is because it may change the established authority at your company. The executive meddler uses his or her exalted position to ask questions that can't be answered (such as "can your guarantee....:) and then uses any resulting silence to justify the status quo. Surprisingly often, and this is only a slight exaggeration, they have a brother-in-law who has a great system and they can get a great deal on it. Respond to all their impossible questions, keep their brother-in-law out of it, and keep repeating the benefits to everyone else.

4. Tell the Vendor (Almost) Everything

A successful relationship is honest. You can demand the facts from your vendors about their products, their history and their plans. In return you need to share your goals, decision process and people with your vendor.

Identifying the process is simply a matter of laying out the plan for purchasing the software and identifying the steps where a decision must be made and who's going to make it. Keep it simple. Use project planning software if you wish (this makes it look official for all of the participants). Share the purchase process with all of your potential vendors. Showing them that you are organized makes them eager to work with you. I don't know if it will save you money, but it sure will save everybody's time.

The easiest decision process is to give one person a great deal of authority - a sort of benevolent dictator. This might be the project leader. If you have the right project leader then no one should mind too much, besides they are supposed to be busy with their own jobs. If they're not busy, you've got problems bigger than a new software implementation and you may wish to skip the whole thing for now.

Vendors are always hankering to meet executives, particularly the ultimate decision maker, the one who can actually sign for the amount that the system or service is likely to cost. This might be the CFO, CEO, CIO, General Manager or someone else. Let the vendors meet this person and spend at least an hour listening to his or her thoughts on the process and important items. After this meeting everyone will have an insight to the motivations of the person who needs to sign and how he or she views the project. If you don't do this early in the process any competent sales person will keep bugging you until you do.

If you can't get an hour of this persons time then they aren't on board yet. Stop the process until they are convinced a new system is important.

You needn't tell the vendor everything, only what they need to know to present their products and services in a way to help your organization make a decision.

5. Negotiate Lock-In

Everyone involved in the purchase process knows that the purchaser is getting locked-in to a solution that is going to be very expensive to change. The vendors will play down the fact, the buyers will play up the fact. That's how it works. Because you'll be locked-in to this product and vendor, you're going to pay dearly later for maintenance, add-on products and service. You'll be stuck and have very little leverage. Now is the time to get your deals.

Consider what it is your vendor wants. If you're a big company, or influential in your industry or region, the vendor may need your name on their reference list more than revenue. If it's a small vendor, cash flow may be much more important than revenue; for 90% down on all products, training, consulting and three years of maintenance you may get a great deal. Public companies, or those about to go public; may need recognizable revenue immediately.

Whatever your deal, don't skimp on services. If you're going to need help, get it.

6. Get Smart

You won't be able to implement what you don't understand. If consulting and training services are not readily available from the vendor or qualified 3rd party - beware. People with experience implementing the product are a tremendous help. They've already been down many of the blind alleys that are lurking out there. They can steer you clear of them.

Have someone on your project team be, or become, an absolute product expert. This person will be able to talk the same language with the vendors' experts. The respect and efficient communication will make the vendor enjoy dealing with your company resulting in better service.

While you're getting trained you can also be cleaning up your data on the old system. A company moving from a legacy system may have hundreds of thousands of records that are now meaningless in customer files, vendor files, part files, order files and other places.

We all know the expression "third time's a charm." Nothing could be more true when it comes to implementing a software system. Your vendor may have implemented scores, hundreds or thousands of these systems. Do what they suggest.

7. Define the Escalation Process

Problems will arise during the implementation. Routine issues will be handled by the project leader and people working on the project team. Sometimes problems can be very sensitive particularly those that stem from personality issues or money.

I mentioned earlier that the project leader should have a boss. When problems arise the boss can be used as a person to whom to escalate issues. Assuming the key vendor contacts also have a boss then they can do the same. Not only does this get new blood into the discussion, it may allow the project team to get back to work while the bosses work out finances or politics. If your project leader doesn't have a boss (i.e. he or she runs the company) there is no safety valve when steam begins to rise. I have seen a relatively minor misunderstanding derail an entire implementation simply because there was no way to escalate the problem.

The escalation process usually consists of two elements: timing and path. For example, a project deliverable from the vendor that is late by some amount of time (for example, one day) will be automatically brought to the attention of the next level of management.

If there are perceived problems in the software then it is essential that your product expert be involved in any discussions.

If you train everybody who'll be using the system when it's first installed, then plan that a year later less than half of the users will be trained. This is because new people are hired, people leave the company or are promoted, new features are implemented and new releases come out. In my experience the biggest source of dissatisfaction with a system, after the implementation, is directly due to people who don't understand how the system is supposed to work.

A related topic is User Groups. A strong user group can make up for a host of vendor inadequacies. They provide a source of support <<<< and training that is very important. Consider the user group when you choose your vendor.

9. Embrace Your Vendors' New Technology Initiatives

I worked for a company that had a great budgeting application that ran only on the Macintosh. We made the decision to port to Windows and support both platforms. Not a single one of our users stated any interest in the Windows version. Indeed, some users were furious. Two years later, when many of their companies forced them to change to Windows, they should have been thanking us.>

About 12 years ago some of the major ERP vendors didn't go to open, relational systems and client server. How many of you are considering ASK, Cullinet, D&B Software, MSA, McCormick & Dodge, Pansophic or Xerox systems. They were the big names twelve years ago. Most of you are only looking at the vendors who kept up with the technology and the market demands of their prospects, not simply the immediate needs of their customers.

There's going to be another market fall-out due to the Internet. Examine your vendors' Internet strategy carefully.