How to choose a geospatial consultancy in Colombia: 10 key criteria

Choosing a geospatial consultancy is a decision that affects an organization's operations for years. A poor provider creates technological dependency, inconsistent data, and unfinished projects. A good provider becomes a strategic ally.

This guide presents 10 concrete criteria for evaluating geospatial consultancies in Colombia. These are not theoretical — they are the same criteria that the most rigorous organizations apply in their selection processes.

1. Years of experience and completed projects

Three years is not the same as thirty. A consultancy with decades of track record has survived regulatory changes, technological transitions, and market crises. That means operational resilience and adaptability.

More important than years is the number of successfully completed projects. Ask for specific numbers: How many projects have they executed? How many municipalities have they operated in? How many parcels have they processed?

A firm with over 150 completed projects has a level of operational maturity that cannot be improvised. Each project generates learnings that translate into more efficient processes and fewer errors.

2. Tech stack: ArcGIS only or also QGIS/PostGIS?

Colombia's geospatial ecosystem requires versatility. The LADM-COL Assistant runs on QGIS. Many organizations have ArcGIS infrastructure. Modern cadastral databases run on PostGIS.

A consultancy that only handles one tool is a limited consultancy. Evaluate whether they master ArcGIS, QGIS, PostGIS, GeoServer, and satellite image processing tools. The ability to integrate open source with commercial technologies is a real differentiator.

Also ask about their experience with spatial databases. A consultancy that works with PostGIS and Oracle Spatial understands data management at scale. One that only handles shapefiles has a low technical ceiling.

3. Do they have proprietary software in production?

There is a fundamental difference between a consultancy that sells hours of work and one that has built proprietary technology products. Software in production demonstrates engineering capability, R&D investment, and long-term vision.

Ask whether their platform is in real use — not in a demo, not in a pilot, but in production with real data and active users. How many organizations is it deployed in? How many concurrent users does it support?

A proprietary cadastral platform operating in multiple municipalities simultaneously is evidence of technological maturity that cannot be simulated.

4. LADM-COL compliance and XTF/INTERLIS experience

LADM-COL is Colombia's national standard for cadastral data. Any consultancy working in multipurpose cadastre must demonstrate full command of the model, including the generation and validation of XTF files conforming to the INTERLIS schema.

Ask how many XTF deliveries have passed IGAC validation without rejections. Ask whether their software generates XTF natively or relies on manual conversion processes. The difference between automatic export and manual conversion is the difference between a scalable process and a fragile one.

Consultancies that participated in the early definition or implementation of the standard have a natural advantage in interpretation and application.

5. Legacy system migration capability

Many organizations have data in legacy formats: ArcInfo coverages, personal geodatabases, Access tables linked to scanned PDF plans. Migrating this data to modern systems is a project in itself.

Evaluate whether the consultancy has documented experience in legacy system migration. Have they migrated data from ArcGIS 10.x to PostGIS? Have they integrated historical cadastral records with the LADM-COL model? Do they have automated transformation and validation tools?

Poorly executed migration destroys value. Well-executed migration preserves decades of institutional information and makes it accessible on modern platforms.

6. Do they offer AI with demonstrable results?

Artificial intelligence applied to geospatial work is powerful — when it works. Many consultancies mention AI in their portfolio but cannot demonstrate concrete results.

Ask for metrics: What accuracy percentage do they achieve in land use classification? How much time do they save with automatic change detection? Do they have models trained with Colombian data or do they use generic models?

AI without local training data and field validation is marketing, not technology. Look for consultancies that can show AI results compared against field verification with verifiable accuracy metrics.

7. Verifiable references

Ask for specific project names, contracting entities, and results achieved. A serious consultancy has no problem sharing this information.

Go further: contact the referenced entities. Ask whether the project was delivered on time, whether quality met expectations, and whether post-delivery support was adequate. References that cannot be verified are not references.

Look for diversity in references: national, departmental, and municipal entities. Projects in cadastre, land-use planning, infrastructure, and environment. A consultancy with experience across multiple domains has a more complete perspective.

8. International vs. local-only experience

International experience adds perspective. A consultancy that has worked in other countries has faced different regulations, different standards, and challenges that enrich their technical capability.

This is not about preferring international over local — it is about valuing breadth of experience. A consultancy with projects across Latin America understands contexts similar to Colombia's but with variations that force flexible thinking.

International experience also indicates the ability to work with demanding quality standards and geographically distributed multidisciplinary teams.

9. Continuity: will they take the project through to operations?

Many consultancies deliver the product and disappear. The project is technically complete but operationally abandoned. Users do not know how to use the system, there is no support, and there are no updates.

Evaluate whether the consultancy offers post-implementation support. Do they have a continuous support model? Do they offer training? Do they update the software when regulations change?

The difference between a successful project and an abandoned one frequently lies in the six months after delivery. A consultancy committed to long-term operations is one that trusts its own product.

10. Pricing model: project vs. hourly vs. retainer

The pricing model reveals the nature of the consultancy. Per-project pricing implies commitment to the outcome. Hourly pricing can create perverse incentives to extend timelines. Monthly retainers work for long-term relationships with variable scope.

Ask how they handle scope changes. Ask whether the price includes software licensing or if that is an additional cost. Ask whether there are hidden costs per user, per parcel processed, or per XTF export.

Transparency in the pricing model is an indicator of business maturity. Consultancies that cannot clearly explain their cost structure probably do not have it clear internally either.

The decision

Choosing a geospatial consultancy is not about choosing the cheapest or the largest. It is about choosing the one that best aligns with your organization's specific needs, technological context, and long-term vision.

At GeoSAT, we meet all 10 criteria. 166 completed projects, 30 years of experience, proprietary software in production, mastery of ArcGIS and QGIS/PostGIS, AI with verifiable results, and international experience. Contact us to discuss your next project.