5 steps to prepare for a migration from Esri to CARTO

If you have started looking into what it takes to migrate from Esri to CARTO, you are already doing the most useful thing you can: preparing. Most of what makes a migration smooth happens before any data moves.

This guide is for GIS analysts and specialists who want a clear, practical starting point. You do not need budget sign-off or a project plan yet. You just need to know what to gather and who to involve, so that when the time comes, the move is straightforward.

It helps to know why migrations stall, because it is rarely the technology. They stall when the right people were not involved, the data turned out to live somewhere awkward, or no one had agreed on which use case had to work first. Every one of these is something you can sort out early.

Here are five steps to get ready.

💡 Already done your homework? The Esri migration readiness assessment turns these five steps into a personalized plan in about two minutes.

Start by writing down what you actually have in Esri: your data layers, web maps, dashboards, apps, and the ArcPy or ModelBuilder workflows your team relies on. Note who uses each one and how often.

This does two things. It shows you the real scope of the move, and it almost always surfaces licenses and content that no longer get used. Knowing what is active and what is shelfware lets you prioritize what to migrate first, what to retire, and what to keep in Esri. Esri remains the best home for some work, including desktop data editing, field data collection, and building information modeling (BIM), so the goal is the right split rather than moving everything. Our guide to choosing the best spatial analytics platform breaks down where each tool fits. When you do begin, the discover phase of the migration agent produces a full inventory for you, so think of this step as a head start that helps you plan.

Where your data sits today is the single biggest factor in how easy the migration will be. If your data already lives in a cloud data warehouse such as BigQuery, Snowflake, Databricks, Redshift, or Oracle, you are in a strong position: CARTO runs spatial analysis directly on that data, so there is nothing to copy or move.

List your warehouses and which spatial data already lives there, then note what is still locked inside Esri formats. That gap is the work ahead of you, and it tells you where to focus first.

It is also worth looking beyond GIS. Spatial data has traditionally been siloed from the rest of the business, kept in its own systems and formats. As you map where your data lives, find out where other teams keep theirs too, including sales, real estate, planning, and logistics. Bringing spatial and business data together in one place is where much of the payoff of the move comes from, because it lets you analyze location alongside everything else the organization already tracks.

Per-seat license fees are the number everyone quotes, but they are rarely the largest one. Add the infrastructure behind an ArcGIS Enterprise deployment, the ETL pipelines that keep data copies in sync, the maintenance time, and the third-party data contracts. Harder to put a number on, but just as real, is the opportunity cost of slower tools and slower decisions: maps that are slow to render, analysis that drags on large datasets, and no modern connection to the AI tools that could speed up and scale your geospatial work. You may not be able to price these precisely, but they belong in the picture.

There is also the question of how you will run geospatial work day to day, from development to production. Managing and scaling spatial analytics across those environments matters as much as the analysis itself, so it is worth asking early whether your tools are built for it. For a practical example, see this post on automating geospatial management.

Putting the full number on paper gives you a clear picture to bring to whoever owns the budget. Our breakdown of the per-seat trap walks through why consumption-based pricing fits modern teams better, and our guide to how to choose a GIS platform includes total cost of ownership as one of its eight criteria.

This is the step teams skip, and the one that stalls the most migrations. Before you start, identify the people who can actually make the move happen:

• The use case owner: the analyst or team who knows how the data needs to be used.
• The cloud or database admin: the person who can grant read and write access to the data warehouse your business already uses.
• The IT admin: whoever manages SSO and can configure access for a new platform.

Also confirm who owns data governance overall, so access policies and approvals are clear from day one. CARTO runs inside your warehouse under your existing controls, which keeps governance simple, but someone still needs the authority to say yes. Most stalled migrations were missing at least one of these contacts at the start.

Pick the one or two use cases that have to work. These are the ones where a poor result would undermine confidence in the whole move. Validate that CARTO can deliver them first, then start with a single high-impact, lower-risk project and expand from there.

Finally, set the migration in the context of where your organization is headed. Spatial data should feed your AI and cloud strategy, not sit in a silo beside it. As we saw at Google Cloud Next '26, enterprise analytics is moving toward agents that work on governed data in the cloud, and spatial belongs in that picture. Choosing where your first use case fits that roadmap keeps the migration aligned with the bigger goal. If you are shaping that direction, our guide to building a geospatial AI roadmap that actually works is a useful next read.

Work through these five steps and you will know your scope, your data, your true costs, your people, and your priorities. That is most of the preparation a migration needs.

When you are ready, the Esri migration readiness assessment takes about two minutes and turns your answers into a personalized plan with the likely effort, a realistic timeline, and the right CARTO entry point.