Technical Challenges

You anticipate running into these but can't be sure up front what most of them will be. The bad news: You will likely run into many more of these than you can anticipate. The good news: Almost all of them have good solutions...many of which require patience and expertise to find.

• ☐ Glue required. Though CMS, SMT and GMS systems usually have APIs for integrating with other systems, these are often insufficiently robust for what you need to do. Do you have resources who can provide the glue required to make it all work together?
• ☐ Purchased tools are not as good as you think. As expensive as they are, enterprise tools often don't do basic things you need. Have you thoroughly explored the capabilities of the systems you are considering?
• ☐ Both good things and bad things effect everyone in an enterprise system. Suppose someone creates a segmentation file that improves leveraging by 5%. In an enterprise system, this potentially helps everyone and not just the group that came up with the idea. On the other hand, if someone does something to cause the system to crash, everyone is affected. Do you have contingency plans for these situations?
• ☐ Machine Translation (MT) strategies. Should you integrate your MT solution with your translation system or leave it as a standalone, separately-integrated engine? There are pros and cons each way.
• ☐ More automation means more integration with complex systems. This is very different from the old, manual, easy-to-understand world. Is your organization ready to increase their technical abilities?
• ☐ Link Management. Most CMS systems can automate the diversion of some links to previously localized content...as long as those links go to content that is also contained in the same CMS. However, many companies do not migrate all their content to a new CMS at once; moreover, there are always links that go to content belonging to other companies. There are straightforward ways to automatically divert links to previously localized content but they require technical expertise to integrate with the CMS. How will you manage links?
• ☐ Throughput/response time. Often, the initial throughput or response time of these systems is unacceptable. There are many ways to improve this. Do you have people who can do this?

People Challenges

Though tool providers are good at showing you how their systems work, most are not adept at guiding your organization through the paradigm shifts required to make implementation successful.

• ☐ A lot of work up front to make things smoother later. When something is hard and takes a long time, people tend to lose interest. Do you know how to sustain interest, engagement, and support for the long haul?
• ☐ Sharing power. Any system that requires people to give up control for the greater good of the organization will meet resistance. How will you handle the power struggles in a way that preserves buy-in?
• ☐ People see what's worse more clearly. New things people have to do or can no longer do in the new system are immediately obvious. Less obvious are the advantages of the system. Are you ready to meet change resistance with concrete benefits?
• ☐ New problems. In a new system you will have new problems; problems people are not comfortable dealing with. How will you deal with this?
• ☐ Sifting through 'wants' and 'desires' for real 'needs'. Most people are attached to the way they've been doing things. They usually want the new system to look and behave like their old one. Often, no single approach works for everyone. Moreover, the way people have been doing things may not be the best or most appropriate way; just the best way they could think of at the moment. How will you manage this?
• ☐ Perception. People base their opinions partially on experience with previous systems. If you had an inefficient process before, it is easier to show new value with a new system. If you had a good system before, it will be more difficult to exceed it. How will you manage perception?
• ☐ Training. These systems are complex to build and use. How will you manage the ongoing training of resources who come and go? How will you manage training in a system of constantly evolving technologies?
• ☐ Strong internal team. This experience can make or break a team. How can you ensure it strengthens, rather than fractures, your teams.
• ☐ Setting expectation for change. Much will change when the new system is implemented. How will you successfully manage expectations in this new environment?

Business Challenges

These considerations lay the foundation for everything else. Time spent up front here makes everything else MUCH easier.

• ☐ Funding an enterprise system. Sometimes it's difficult to get all the people who previously funded their many smaller projects to join forces and fund a new, big one. How will you get everyone to support the new system?
• ☐ Showing value. It is important to keep reminding management of the value the effort will generate. How will you measure value?
• ☐ What process flexibility to enable and what to standardize or hardcode? All three methods have their place. Do you know when to do which?
• ☐ Management sponsorship. Do you have a sponsor? Can they obtain the resources needed to carry the effort?
• ☐ Support. Accomplishing the implementation is only the first step. What is your plan to maintain, support, and improve the system going forward?
• ☐ Big bang implementation vs. iterations. What is your implementation strategy? Big bang puts people through a big change only once. Also, funding may not last long enough to realize future iterations. However, iterations give you a chance to stabilize each major step before introducing a new one. You build from a solid foundation and lead people slowly to the end state. Which strategy is better for you?
• ☐ Too many vendors? If you look for the tool that provides the best quality for each piece of the system, you may end up with a puzzle where some pieces don't fit. Each vendor represents an additional communication channel to manage; another place where things can bottleneck or go wrong. Like sports teams, a system of parts that work well together often beats a system of 'star' parts that don't work well together. How will you balance 'technology' vs. 'overall system intercompatibility' and 'ease of maintenance'?
• ☐ Don't code for things to be 'Out of the box' soon. What are the tool vendors doing over the next few years? Can they be influenced to provide something so you don't have to make it yourself?
• ☐ Need a long term strategy. Your strategy guides all decisions. What is your long term strategy for the system? How will you create the strategy?
• ☐ Touchpoints. Every place where data travels from one tool to another is a place for problems to occur. Have you minimized the touchpoints?
• ☐ Coordinating with source effort. In most areas, localization follows the path taken by the source files. Do you have a localization plan for each place that provides source content?