Data migration is a critical process for any company that undertakes it. There are many reasons a company may migrate data to a CRM. They might be moving from one CRM to another. They might be adding a new CRM to their tech stack. Or they might be migrating data from multiple other systems to their existing CRM.
But no matter the reason, moving data is rarely a hiccup-free process. Unexpected twists and turns can complicate it. Every platform has its own nuances, and ensuring that two systems play nicely together takes careful coordination.
We recently spoke with Remotish HubSpot Strategist Camille Balhorn about the CRM data migration process. Camille has played an integral role in many HubSpot migration projects—including some that migrated data from as many as seven different platforms.
She provided detailed insights on how those projects work, the primary considerations when undertaking a data migration project, and some of the quirks and nuances that you need to be prepared for. She also shared how she uses Insycle to aid her projects.
With a little foresight and some understanding of the intricacies of these projects, companies can put a plan together that will speed the process while minimizing bumps in the road.
Any successful data migration project begins with planning and analysis, Camille explained. You can’t just begin exporting data from one system and importing it into another and expect a successful outcome.
First, there are considerations for how companies collect and use the data in the systems that they are migrating from. The goal is always to migrate to a new system with as few disruptions to existing processes as possible. But your new CRM may not have features and interfaces that allow for a seamless transition. To limit organizational interruptions during and after the migration process, you’ll need to understand those nuances.
Camille recommended kicking off your data migration process with interviews and conversations to better understand and assess how companies collect and use the data that will be migrating down to the individual field.
“Data migrations require so much discussion upfront, then a lot of work on the back end. A large part of pre-migration planning is discussions with key team members about the objects, properties, systems, and processes,” she told us.
Camille noted that these discussions can benefit from visual aids including tech stack maps, downloads from the various systems, and any process diagrams. And the discussions themselves are critical.
“There are rich conversations to be had about properties during a migration, as this is quite literally how we transfer data between teams and systems,” she noted.
Camille suggested speaking about a few specific things in the early stages of the project:
With these kinds of questions answered you can then begin to design the specific solutions required to migrate the data to a new system.
But relying solely on explanations from users, while helpful, is often not enough. You have to roll up your sleeves and get your hands dirty, analyzing the data in the systems that you are migrating from to get a full understanding.
“You need to analyze the systems themselves,” Camille said. “What has this system been used for historically? Are there any unique qualities that will need to be kept in mind that will change how we migrate it? Additionally, which systems are being decommissioned versus integrated into the new system?”
Often this means transferring the data to spreadsheets to find areas where it may not align from one platform to the next.
“Spreadsheet work is certainly part of it as well. This would include making sure all property data is in the correct column. An easy example of this would be columns with combined first name and last name, which I know Insycle can help with,” Camille said.
But understanding all of this information and how one system will translate to another is not always so easy. To deal with this, Camille recommended using software to help you chart out how these different aspects will translate to new systems.
“This is where a tech stack map can be helpful. Use a program like Whimsical, Miro, or Lucid Chart to map out your current and future tech stack and what you'd like it to look like post-migration,” Camille said.
She recommended creating a legend for the connections between applications to show the difference between manual data entry/export, integration, and iPaaS connection, such as Zapier. Additionally, indicate a direction for the data flow. These visual clues will help you get a full understanding of which systems need to be turned off and when they need to be shut down, as well as what connections could potentially present a problem during the migration process.
“This helps us get into some process discussion as well, as this helps explain why and how certain tech stack applications are connected and which departments use which applications in which ways,” Camille said.
|
Related articles Four Phases of Customer Data Management Evolution |
Once you have a deep understanding of your current systems, the data they contain, and how that data will translate from one system to another, then you can begin the task of actually migrating the data.
Camille explained that a migration project often entails moving data from different objects—contacts, companies, deals, and custom objects—into your new CRM.
To do this, you have to download the records from these different objects and import them separately into your new CRM. You can’t just generate a CSV with all of the records that you need across different objects. And you need to ensure that each record contains the ID number from the original system to ensure that it can later be identified for association in the new system.
As you migrate data, Camille explained, you want to ensure that records are correctly associated with other records in your system:
“You need to ensure that all of the data that you want associated is in the same spreadsheet, on the same row, when you're importing it. For some objects, it's easier than others. For contacts, if you have the contact email, you can usually easily import that data and ensure that HubSpot is able to recognize whether the contact already exists using that email,” Camille said.
But other objects may not be so straightforward. For instance, deals don't have a unique identifier in HubSpot.
“That can become a real pain point because, if you don't import the deals along with the other data that you want the deals to be associated with, such as companies, it requires a lot of VLookup work to later associate those deals based on their name or other properties that you do know,” she said.
Data quality is a big challenge for most organizations. Add in the fact that during your migration, data could have wildly different formats in two systems, and it can be confusing when and how to tackle data cleansing. Overall, Camille recommends to divide your data issues into categories.
First, which actions would a spreadsheet platform excel at performing? For example, deleting property columns that will not be imported to the new system, renaming columns for clarity, or moving data with a series of VLOOKUPs. Start by identifying these opportunities and taking care of them–but beware of falling into data-cleaning rabbit holes in spreadsheets.
“Cleaning up issues in your data is enticing when you're in an Excel spreadsheet. Like, wow, I found this issue in the data that I am planning on migrating, and then it snowballs until you've spent hours fixing your data quality manually, which is such an inefficient use of time,” she said
Instead, continually ask yourself if the data action you’re performing could be done more efficiently using smart data software. For example: formatting phone numbers, tackling capitalization, and standardizing state and country names. In Excel, these actions would require a series of search and replace or complex formulas. This is where programs like Insycle come in handy, which can be used as you’re importing and post-migration. Address these aspects of data cleansing in a couple of minutes instead of hours.
Finally, consider which actions will be most effective with all of your data present in the final system. Perhaps by the time you’ve migrated, new conclusions have been made regarding properties or processes that you can now adjust. Additionally, tasks such as merging duplicate records could be redundant if done before and after migration.
“I prefer to deduplicate data after it gets into its final system. Because you always have to deduplicate after importing a bunch of new records, deduplicating before migrating the data often creates a lot of double work. If you're trying to deduplicate in Excel, sometimes you’ll miss duplicate data using Excel functions. You're never sure if you're actually catching all of the duplicates. So, instead, bring all of the data into the new system, then deduplicate there,” she said.
And don’t forget ongoing data quality management once you’ve settled into the final system. Use a single data software solution that connects with your CRM to set up continuous data automations as new contacts come into your database.
Even with thorough preparation, there will always be challenges when migrating data from one system to another. The puzzle pieces will never fit together perfectly, and you are sure to encounter unexpected hang-ups throughout the process.
We asked Camille about some of the biggest challenges that she has faced in her CRM data migration projects. She highlighted:
By keeping these challenges in mind, companies can run a smoother and more effective data migration process.
Once the data has been migrated and cleaned, the project is not complete. Now, you have to help your company create processes around its new systems and understand the data therein.
“These migrations are often occurring because the company is moving from one CRM to another,” Camille said. “That often comes with new processes. Remotish promotes and focuses heavily on documentation for migration projects.”
The post-migration documentation often includes:
“For example, a new CRM system may require a new sales process. In that case, we may outline what those pipeline stages are and what properties we brought in that will now be required for reps. Anything we need to create to help their teams learn that new system,” she said.
However, completing a migration project using standard CRM features and interfaces can be an uphill battle. Often, it is best to look for third-party tools to help facilitate the process.
In this interview, Camille has shared many useful tips for running a successful data migration project.
Camille also shared a walkthrough video on the Remotish blog, covering how she uses Insycle’s Magical Import module in the data importing and migration process.
Click the video button above to check out all of Camille’s tips for using Insycle for data migration projects.
Here are some of the most impactful points to remember from both our interview and Camille's video:
Insycle is an invaluable tool for data migration projects. Using Insycle, you can analyze your existing data and prepare for the process of migrating data between systems, import CSV data in flexible ways, and help with the process of documenting changes to aid in the post-migration process of adopting a new CRM.
Insycle delivers numerous features that are purpose-built for CRM data migration projects. Let's examine.
Insycle helps you to analyze what you have before and after migrating data.
Using Insycle, you can easily get a bird’s-eye view of all the existing fields in your database along with important information such as the field type, the number of unique values contained within each field, and the number of records with empty values. Additionally, you can download your field-level statistics as a CSV for analysis in Excel.
Doing this can help you to spot inconsistencies between platforms during a migration. For example, your "US States" field may be a text field in one system and a picklist in the other. In that situation, if you were to import the data without standardizing, you would create a mess of duplicate, unstandardized picklist values for each state, such as "NY," "N.Y.," "New York," and "New York State," which could impact segmentation, reporting, and automation that run based on that field.
By choosing a specific field in Insycle, you can get a complete picture of all of the different unique values contained within the field. For example, you might review Job Titles for contacts.
Then, by selecting a specific job title, you can analyze all of the records that have that specific job title.
This gives you a simple way to evaluate specific fields and the data that they contain, which is useful before, during, and after migrating the data. This helps you understand what is in your database, spot potential standardization issues, and create reports to help your company better comprehend the data that they have on hand as part of the project wrap up process.
During migration projects, there will come times when you want to compare your CSV data to what you have in the platform that you are migrating to.
Maybe you can't remember if you imported a specific CSV into the new system. Or, you want to see how records from one system compare to data that you have already migrated.
For example, if you have migrated data from Salesforce to HubSpot, and are now migrating data from another sales tool to HubSpot, there is a good chance that some contact and company records may already exist in HubSpot from the previously moved data.
In that case, you'll want to compare the data from your sales tool to the existing data in HubSpot. One system may have more data collected, or higher quality data that you want to ensure that you keep.
With Insycle, you can compare data contained in a CSV to existing data in your CRM. This gives you a picture of the records you already have in your database and what information they contain, compared to the data you plan to migrate in your CSV.
This will generate a new CSV that directly compares the data contained in the CSV that you uploaded to the records in your CRM, allowing for easy side-by-side comparison. With this, you can determine how to handle your data imports on field-by-field basis.
But Insycle helps with much more than analysis. The Magical Import module also helps with the process of importing data, giving you flexible ways to move data between systems and update it in the process.
When you select a CSV to import with Insycle, fields are automatically identified and mapped for you. Then, you can save your template to ensure that all future imports use the same mapping settings as well, saving you time on all future CSV imports.
And migrating data from one platform to another may require hundreds of CSV imports over the course of the project. Not only does saving your field mappings as a template save you a lot of time, but it allows you easily hand off import tasks to others without having to worry about mistakes.
You have complete control over how the imported data is used on a field-by-field basis with the Import Mode setting. You can choose to:
This is ideal for data migrations because each field may need to be treated differently depending on the underlying data, and these import mode options give you complete flexibility for each field.
For example, a "notes" field may require that you append migrated data to existing data in your CRM, so that your sales team has full context next time that they communicate with an account.
In her walkthrough video, Camille recommended creating an 'Original Application' field in your CRM, and including that field in your CSV imports so that you can keep track of where the data is coming from, for auditing and troubleshooting purposes. The append option is useful for this because an individual record may contain data that has been migrated from several platforms.
Or, if for example, you have already migrated a contact into your CRM from a different system, you can choose the "Fill" option to ensure that you are only importing data for fields that currently have no data in them.
Then, Insycle gives you numerous tools for cleaning and associating data while importing.
After loading your CSV into Insycle, you have numerous ways to prepare, clean, and standardize the data before importing it into your CRM. In the example below, we are formatting phone numbers and ensuring that contact and company names have proper capitalization.
Automatically formatting these fields before importing reduces the cleanup tasks required after the data has been migrated.
Then, you can also associate records to other records in your CRM while importing. For example, you could associate contacts to companies, using any field in your database as a matching field.
This is critical for migrations because importing the records without those associations in place leaves you with a huge task once the data has been fully migrated. And for ABM teams, it's really not possible to use your new CRM until all of your records are properly associated so that they can have a complete view of each account and the communications and activities that have already taken place.
Once imported, you have a complete record of all changes that were made to your CRM data in Insycle’s Activity Tracker. With these templates in place, you also have defined processes for importing data in your new system.
Insycle can also assist with documenting changes and new data operations considerations after the data has been migrated.
First, Insycle's Activity Tracker is a great resource for showing the changes made to specific fields during the migration process. It delivers a fully auditable trail of every change made during the migration process.
Additionally, Insycle's Customer Data Health Assessment analyzes your database and provides a top-down view of all of the identified errors within your CRM database.
Those individual issues within these categories are detailed in Health Assessment. Clicking the "Review" button takes you to the correct Insycle template for fixing each issue. While Insycle allows you to clean up data as you import it during a migration project, there will always be some data issues that sneak through.
The Health Assessment is also an excellent tool for quantifying data quality in your new CRM both post-migration, and on an ongoing basis.
Lastly, there are Insycle Recipes. A Recipe is a collection of Insycle templates that can be bundled together and executed in a specific order. They can also be automated to run on a set schedule.
For data migration projects, Recipes can play a critical role in defining processes for data management tasks. For example, in the above screenshot, you can quickly gain an understanding of all of the different ways that contacts are being identified and merged in your CRM. This is critical for post-migration documentation and training.
When companies undertake CRM data migration projects, it’s critical that they get them right. As Camille noted, it is important to get the lay of the land prior to beginning the migration process. Companies should also document changes made along the way to ensure a smooth transition for all teams and stakeholders.
Insycle is an invaluable tool that empowers users to seamlessly manage their data in a variety of ways. When they use Insycle to analyze data and prepare for migration, companies can make informed decisions about the data transfer process. The Magical Import module offers flexibility in migration, catering to the unique needs of different businesses. Moreover, Insycle's commitment to improving data quality on an ongoing basis ensures that your data remains accurate and valuable over time.
But Insycle is not just a data migration tool; it's a comprehensive CRM data management solution. Insycle helps organizations keep their customer data well-organized, clean, and easily accessible, ultimately leading to better business decisions and enhanced customer relationships. By providing an all-in-one solution, Insycle consolidates myriad data management tasks into a singular, user-friendly experience.
Don't miss the opportunity to harness the power of Insycle for your business. We invite you to learn more about how Insycle can revolutionize your CRM data management process and help your organization thrive in an increasingly data-driven world.