Seven Warning Signs Your ERP System Is Extinct
I enjoyed this white paper from PLEX Systems on Seven Warning Signs Your ERP System Is Extinct because there are so many ERP systems out there today that are literally technology dinosaurs and so many companies who are unaware, thanks in part to continuous efforts on the part of their providers (charging 22%+ maintenance) to keep them in the dark. It's nice to know that someone besides myself and Vinnie is willing to speak a little truth on the issue now and then.
The white-paper doesn't beat around the bush and gets straight to the seven warning signs it promised:
- Your ERP system can't integrate mission critical business data
"Silos" of data are not useful. You need integrated data to do meaningful analyses that can deliver meaningful results. - Changes to the system are costly and time-consuming
Are your new releases on a 24 month schedule? Do new features only appear years after you need them? Do custom changes start in the six figure range and take months to complete? In comparison, new SaaS platforms tend to do major releases on a 3-6 month schedule, develop new features as soon as they are identified as important, and use integration platforms that allow them to develop customizations in a matter of weeks that start in the five figure price range. - Your disaster recovery plan involves tapes
A technology that has all but disappeared because it is easily eaten by players, easily erased by nearby magnets, easily destroyed by a spilled drink, and easily lost. (When was the last time you saw a tape produced by the Music Industry?) A modern system that maintains an on-line remote backup can be recovered in minutes as compared to the hours, or days, it can take to reload data off of tapes -- which could be days, or weeks, out of date. - Maintenance fees are high
Maintenance fees on current technology should be low, not 22%! - Upgrades are Disruptive to the Business
Outdated ERP systems often require OS, database, and hardware updates which take time to plan and even more time to execute and take your business 'down' for a period of time to do the conversion. (In contrast, modern SaaS systems can be upgraded instantaneously.) - Trading partners can't easily interact with the system
There's no ability for partners to get data from, or submit data to, the system -- which is problematic considering how importance visibility is in today's supply chain. - Globalization is too difficult
Many legacy systems require you to run a different version of the system to support China, Russia, India, or another country. A modern system supports multiple languages and allows you to run the same system in every company you operate in, and only the language used in the UI is different.
Excuse me, but an OS or database update will take a SaaS system down for just as long as any other system. If the author is saying that another system can be brought up in parallel with the first, so what. Such a strategy can be executed with or without SaaS. Hardware is cheap, virtual hardware cheaper.
You're right, but how many companies have sufficient staff with sufficient competency and resources to pull this off?
The answer is not very many. With SaaS, the vendor takes care of maintaining a parallel instance, and handles the (near) real-time switch, updates the system, synchs the database(s), switches back, updates the parallel instances, syncs the database(s), and restarts the regular database backup and synchronization to allow for (near) real time switching or fail over when an update needs to occur again. In reality, 95%+ of customers would not be setup to do this with a traditional system.
Furthermore, what you are suggesting is a lot of work, and added up across a few hundred customers, an awful lot of resources (time, money, etc.) if you go the traditional solution route. With SaaS, it's done once, and then it's done.
In summary, you're right in theory, but in practice, it doesn't happen.
How many vendors are paying their SaaS hoster for a parallel instance? How many SaaS hosters are providing a parallel instance for free? How many vendors are planning for database and OS upgrades by creating a parallel fail-over instance for every customer?
This argument is another way of throwing the ball over the fence. "It's up to the vendor, and because it's SaaS, my vendor is obviously super intelligent and super careful, so I don't have to worry about anything, it'll just happen by magic."
No, it won't. You should worry deeply about it. In fact, you should write into your SaaS contract that your SaaS vendor MUST provide a parallel fallback instantiation of your solution WHENEVER there is a database or OS or JVM or [whatever] update.
Right again. I guess I keep assuming that end users know by now that the goal of the majority of vendors is to get as much money for as little as possible (on the software and services front) and that they will only see the numerous benefits a solution has to offer (which, for SaaS, includes the benefits listed here http://www.esourcingwiki.com/index.php/On-Demand_/_SaaS_Application_Platforms#Benefits, for example) IF they actually insure that all the benefits are clearly described and mandated by the contract.
The reality is that most of the better SaaS vendors and all of the vendors who have a major SaaS (related) certification (like SAS 70) have a parallel, fail over instance in a separate facility. However, not all have the same amount of computing power available in this fail over facility and those who distribute customers among multiple facilities might not have all customer profiles and data backed up at the primary fail over facility if it's not in the contract that the fail over facility must contain a complete data backup and enough computing power to provide the same level of service as the primary facility.
So, with SaaS, you can get all these benefits (and more), AS LONG AS you make sure they are clearly spelled out in the contract that it is the vendor's responsibility.