Is integration with existing systems a major obstacle for cloud-based ERP?

Being the President & CEO of an ERP solution company, I feel that if the integrations are done correctly, they are not a problem for on-premise or in-the-cloud. The key is to have the integrations be seamless and automatic.

One of my fundamental beliefs is that for an integration to be truly successful, the user of the integration must be able to have access to all the critical information they need to do their job in the application where they normally perform their function. For example, a salesperson should not have to log into the ERP to gain access to information they need that resides in the ERP. The ERP should automatically feed the CRM key account information including historical customer information, prices and products for quoting, and serial numbers for warranty tracking.

This approach has many benefits including:
1) eliminating the need for buying additional ERP user licenses
2) the sales reps do not have to learn the ERP application
3) information is timely and accurate

With regards to certain integrations, the integrations need to be bidirectional. Using the CRM as an example once again, if a detailed quote is provided in the CRM, there should be an option to send the quote to the ERP, so that it can be turned into a Sales Order after approval by the appropriate function(s).

If the above approach is used, then you will have a valuable extension of your ERP as opposed to a "bolt-on".

Bob,

How would you describe a bolt on?

Thanks for question Steve and allowing me to add some clarification.

From my perspecitve, the top four things that distinguish a bolt-on from seamless integration are:

1) the integration requires a fair amount of manual intervention to keep the information timely and accurate

2) the integration does not have safeguards to prevent undesirable events from happening.

Sticking with the CRM scenario, if a salesperson can add a new "Customer" and have it automatically be added to the ERP, this would probable cause the Customer Master in the ERP application to be quickly polluted with prospects as opposed to true customers. The integration should have a process step that puts the "Customer" on-hold until someone in Finance can verify that this truly is a customer and add all the important information regarding the account such as credit limits.

3) Vailidation; the ERP should validate all the imports from any source before being accepted. In the CRM situation, a new quote being passed to the ERP should be rejected if the quote has an invalid part number.

4) The user has to perform significant additional steps, outside their normal workflow, to make the integration work

I agree with all of the previous comments. There is not much of a difference between integration using a Cloud versus an on-premise system.

The only major difference is that customization and integration code will travel forward automatically during upgrades with Cloud applications like Salesforce.com and NetSuite while one must re-implement integrations and customizations when you upgrade on-premise systems.

However, to Bob's point, there are substantial ROI and productivity benefits related to using one complete system versus integrating multiple applications. Here is a link to a study done by analyst firm Nucleus Research on the ROI of using a seamless system: http://www.netsuite.com/portal/pdf/roi-evaluation-report-netsuite.pdf

Rob

Hi Rob,

I do have some disagreements with your comments:

1) While customizations in the cloud will "travel forward automatically", the degree of customization that one can do in the cloud is typically less than an on-premise deployment. Said another way, while you are able to avoid the pain of migrating your customizations with the cloud, the amount and the complexity of the customizations that you can deploy with an on-premise solution are typically greater than with a cloud solution.

2) With regards to the the ROI comments, I have to assume (but, I may be wrong) that since it is analyzing NetSuite vs a generic study, the paper will be biased towards that conclusion.

More importantly:
1) The analysis does not take into account the quality, depth, scale, etc of an "all in one system" vs integrations. For instance, if the "all-in-one" system has certain Quality modules to handle customer complaints, manufacturing defects, etc, how strong is that module as compared to a seamless integration with a 3rd party best-in-breed Quality Management System.

2) The report was written in 2008 and much has changed in the world of Integration thanks to companies like Pervasive.

3) The long term cost of SaaS vs on-premise is not a clear cut answer. In many cases, the cost of SaaS is higher than on-premise as the payments do not end. Being an ERP vendor that offer both solutions and being a former CFO, I am constantly watching this metric. My analysis conclusion, which I have bounced off an industry expert, is that typically the cost "cross-over" point is approx 5 years; some instances happen sooner and some a bit later. Again, being an ERP vendor that offer both solutions, I am completely indifferent to which deployment method the customer chooses, as I know that with a SaaS deployment for an ERP solution (which tends to be rather "sticky" and subject to minimal customer churn), I will increase my monthly revenue annuity stream.