In order to be successful, companies first need data from a wide range of systems, from CRM (Customer Relationship Management) to PRM and SRM (Partner or Supplier Relationship Management) to PLM (Product Lifecycle Management) and SCM (Supply Chain Management) in their own company or have received from the outside and then evaluate accordingly.
It depends essentially on three fields: data collection, data integration and data analysis. Data collection and data analysis and the conclusions to be drawn from this are very individual for each company. The data hub in the middle, which ensures that the various data are delivered from various sources, assembled as needed and made available to specific target systems in their structure, can already be standardized well today. For the planned IIoT strategy to work, data integration must be highly flexible, error-free and standardized. The following checklist shows what is important.
12 tips to help you with data integration:
1. Keep it simple
Simple handling such as an intuitive user interface and drag and drop, easy installation by Software as a Service, which simply runs in the cloud without its own installation and can be docked – through such amenities impresses a good software. Try it out: how complicated it really is to work with the new software for the first time and then integrate data from one system into another.
2. Transparency for the view
Two aspects make the difference here: Does the software offer you a comprehensible and clear structure? And is there an automated and problem-free documentation of all mapping processes for later analysis and for the exact follow-up?
3. Save time
Let the clock run: How much time – and how many steps – do you actually need when you create new customers or partners or start another process? Agility is one of the basic requirements of IIoT. You should avoid error-prone intermediate steps via other systems.
4. Perfect performance
Especially if you have to process high volumes of data, then: Do the stress test. How many records does your system really handle? And what utilization does that mean for your systems?
5. Efficient of one for all
Do you use several systems for your integration topics (EAI / IOT / EDI / ETL …) or can you handle all tasks with one tool? Can you also use central monitoring for all interfaces?
6. Easy to use
Are specific programming skills required to integrate your data? Or can employees of the specialist departments implement the data integration based on their process knowledge by means of configuration?
7. Always up to date
Do you work with custom-made special solutions that require specific programming skills for maintenance, changes and updates? Or can you use standard software with automatic updates?
8. With all standards
Can your preferred system – in view of your flexibility and future security – access all current industry standards when needed (which means immediate access to almost 10,000 different templates)?
9. Cost / Benefit
When comparing the price / performance ratio between different standard softwares and standalone solutions, you should keep an eye on the direct costs, but also the future costs (e.g., costs of setting up new interfaces). Do the chosen systems allow a quick start as a web application without lengthy installations on your own servers while at the same time protecting your data in your own environment? What is the cost of changing processes (programming vs. configuring)? And what costs do you have to expect for updates and upgrades? Are licensing models easy to understand and transparent?
10. The great freedom without dependencies
Are you independent of manufacturers of certain software systems? Are you trapped in a provider system after purchase? Are you dependent on service providers (cost-time factor) and operating systems? Which hardware requirements are placed on the system?
It’s simple: think of three tricky questions and test the hotline. Fast? Friendly? Competently?
Ask someone who already knows the system of your choice. Advertising brochures and statements from sales representatives are interesting. Findings from two or three users who have been using the system for some time are more interesting. Case studies on the website are a good indication.