How Much Does Data Integration Cost?

data integration costs

How Much Does Data Integration Cost?

If you’ve never performed a data integration before you may be wondering, “How much is this going to cost me?”. Well, that depends. You see, integrations have a lot of moving parts, and each project has its own unique requirements. In addition, where your solution lives and which integration company you work with can also affect the costs. If your solutions are on-premise for example, you must account for the cost of hardware and data backups in addition to your other costs.

In this post, we’ll offer a realistic price range for four of the most common categories of data integration: Pre-Built, Custom Cloud, Web service, and Legacy. We’ll also offer some tips on reducing these costs.

Pre-Built Integration Costs

The beauty of pre-built integrations is in the name: they’re pre-built; they already exist. That means the standard fields for your connecting systems are already configured for you, out-of-the-box, when you purchase the integration. This allows you to perform your data integration faster, and for less money. Pre-built maps often allow the option to add in custom fields too, which is a great bonus. You get all the benefits of a basic integration with the added opportunity to incorporate your crucial customizations.

It’s important to investigate whether your integration provider offers pre-built maps for your integration. Not all providers do. If your integration requires the provider to build a map, that will increase the investment you must make.

Pre-Built Integration Cost:  $1,200 to $6,000 annually depending on the complexity and products.

Custom Integration Between Cloud Services

Right off the bat, data integrations between Cloud services is going to be cheaper than legacy integration. This is because legacy integrations often host a slew of integration stumbling blocks: Lack of access methods, data access only, security vulnerabilities, API complications, IT governances, cost of additional hardware, etc.

Custom data integration costs less in the Cloud because Cloud-based integrations have Web services and APIs. This allows most integration companies to access the system’s functions and not have to worry about any additional data administration or security. Even if the API is a little primitive, it can be wrapped in as a Web service for any missing API functions that are needed.

Custom Cloud Services Integration Cost: $4,000 - $8,000 for Web service, API to a Web service, or API integration.

Integration Between Applications Using Web Service

When an application is hosted in the Cloud and it has a Web service layer, we would define this as a Cloud service. But many applications installed on-premise have Web service layers, and they are usually treated differently than the scenario above when one of the systems is in the Cloud and the other is on-premise. In this case, either the application will need to have visibility to the internet (which is usually not recommended) or it will need to communicate through an intermediate application like an ETL tool.

Application Integration Using Web Service Cost: $6,000 - $10,000 for a system where one application is on-premise.

Legacy Integrations

Legacy solutions can run in either a private Cloud or an internal data center. Either way, our definition of “Legacy” is an application that has a limited API that allows you to retrieve data access, no API but direct data access, or a closed system with simple import / export features.

The major issues with most legacy systems, is that without a fully supported API to provide business logic when inserting data, you are forced to reconstruct the program’s business logic as part of the integration. This means that every time the application is updated, the integration will need to be re-tested and potentially rewritten.

Integration cost drivers for a legacy solution are:

  1. The type of access that is allowed
  2. The need to develop Web services to simulate a non-legacy solution
  3. Data access capabilities

Legacy Integration Costs: $5,000 to $15,000 for retrieval from a SQL database, $15,000 - $35,000 for retrieval using ODBC, $35,000 + for building a Web service.

Other Integration Considerations

Integrations aren’t static, they’re ongoing. No matter how simple your integration may be, there are additional costs you must consider for the success of the project. Let’s look at those considerations.

Data Cleaning & Preparation

Make no mistake, data cleaning and preparation is essential to a successful data integration project. Poor data quality undermines the goals of your integration. It can reduce your return on investment and any benefits that may have resulted from improved transparency and collaboration. Data cleaning can be approached in one of several ways:

  1. The provider will use a data cleaning tool built into their integration solution to prep your data for you. For example, we use InsideView within the StarfishETL solution.
  2. The provider will outsource your data cleaning to a trusted partner.
  3. The provider will ask you to find your own data cleaning service to prep your data and deliver it to the integrator.

While it may be tempting to try and clean your data yourself, be aware that is extremely time consuming without the right tools. The ideal situation is to have your integration provider clean your data for you. In most cases, data cleaning will charge pennies on the dollar for every piece of data being augmented. A quality data cleaning solution will be able to reference thousands of data sets from around the internet to verify the information’s accuracy.

Data Cleaning & Preparation: Expect to pay about $1 per record.

Support & Training

As with any software project, blips are possible. That’s why a go-to support resource is a must. Integration providers offer support packages at different tiers to accommodate varying budgets and needs.

If you have an in-house developer performing your integration, they’ll need training from the integration solution company to learn how to navigate the system and map the necessary fields. Training sessions can be anywhere from one hour to 10 hours long depending on the complexity of what the developer needs to learn.

Support costs: 15% to 25% of the price of the software

Training costs: $1,500 - $10,000 per developer

Custom Fields

It’s not uncommon for a business to utilize custom fields in their CRM, ERP, or other software. Custom fields help teams maximize their use of the system by adapting it to the way they work. Although these fields are fantastic resources within the system, they can sometimes become a menace when trying to integrate solutions together. As you may have guessed, custom fields require building custom mapping so the integration provider can plan where that data will transfer to the new system, and how. This takes additional work, which means additional costs.

Custom fields cost: $25 per custom field

Ongoing Maintenance

CRM’s are upgraded about 4 times a year. The new upgrade can affect the integration by modifying the Web service or changing the data structure. Although this happens about once every 5 years for a Cloud application, it can be more frequent for a legacy application.

Ongoing maintenance cost: 10% of the integration cost for a cloud solution and 25% for a legacy solution per year

How to Reduce Data Integration Costs

With so many costs to balance in an integration, the prospect of the investment can sometimes feel overwhelming. However, if you go into your integration with a well thought out plan, a vision for what you’re trying to accomplish, and a realistic timeline, your data integration will be a smooth success.

There are ways to reduce your data integration costs if you know where to look. Here are a few tips:

  1. Use clean data! – As mentioned earlier in this blog, clean data is imperative to integration success. But, beyond that, it’s also imperative for keeping costs low. Without a company-wide focus on data quality, the project will fail, and in doing so, waste the investment. It’s a good idea to establish master data management to support your long-term data quality. These rules define how your users can create, edit, alter, store, use, and dispose of information in your systems. By setting these guidelines, you safeguard the accuracy of your data and guide your teams to better data practices.
  2. Be mindful of your customizations. – Which custom fields or functions need to be carried over to the integration? Have a discussion with your stakeholders to identify the custom fields that support user-specific processes, and the ones that don’t. Too many customizations can quickly skyrocket the cost of your integration and complicate the development efforts. Sync up your customization strategy to match your budget and goals without going overboard.
  3. Future-proof it. Think building your own one-off integration will save you money? In the short term, maybe. In the long term? Definitely not. Rigid, unscalable, and incompatible integrations will end up costing you more in the long run than your initial investment would have been with a trusted partner. What happens when the system is upgraded? What if you want to extend your integration to other systems? One-off integrations make upgrades and maintenance less flexible and more expensive. Future-proof your integration with a scalable and adaptable solution.
  4. Prioritize primary systems. – The prospect of integration can be exciting. So many new relationships to analyze and reports to generate! It’s easy to fantasize about all the additional integrations you could create with other systems, but that doesn’t always mean the results will be as beneficial. Take it slow when it comes to integrating your systems. Start by prioritizing your primary systems, aka the systems your teams use to execute your business processes; the ones that are most crucial to their work. Consult with your integration provider on the best approach to additional integrations. What else is necessary to integrate? What should be left alone? Tangling up all your systems in integration can get messy, so focus on your main goals to start and build thoughtfully from there.
  5. Consider a move to the Cloud. – As we’ve explained in this post, legacy data integration can be time consuming, tricky, and expensive. Migrating your legacy solutions to the Cloud enhances the long-term benefits of your integration. Here are a few reasons why:
    1. Updates and upgrades are easier
    2. The system can be scaled faster for changing business needs
    3. It eliminates the cost of hardware and software, which are bundled into the Cloud
    4. You can take advantage of Platform as a Service (PaaS) functionality to reduce overall costs
    5. You can use low code development tools to speed up development