How Delphinus Moved 6,000+ Records and 2,700 Historical Stage Transitions from Affinity to Attio Without Losing a Thing

TL;DR

Customer: Delphinus (Investment firm)

The Move: Affinity to Attio

The Scope: 6,000+ records across companies, people, deals, notes, and files — plus a fully recreated deal pipeline, a custom Syndication Partners object, and 2,700 historical stage transitions extracted from Affinity change logs.

The Roadblock: Not a copy job. Affinity's data model does not map one-to-one to Attio. Deal notes needed to live on company records. Vertical tags needed to propagate from deals to companies. Attio overwrites system timestamps, making historical date preservation impossible without custom attributes. A custom Syndication Partners object with two-way deal relationships had to be built from scratch.

The Outcome: A production-ready Attio workspace with a clean data model, preserved investment history, historical reporting data, and a validated delta migration for seamless go-live.

By the Numbers

• Companies: 1,778 (739 after post-migration cleanup of enrichment duplicates)
• People: 1,989
• Deals: 691 (including 14 Won-stage portfolio investments)
• Notes: 832 (846 after delta)
• Files: 715 (730 after delta)
• Historical stage transitions generated: ~2,700
• Delta migration additions: 10 people, 3 companies, 43 updated list entries, 22 notes, 5 files

The Challenge: Rebuilding an Investment Workflow, Not Just Moving Records

Delphinus needed to leave Affinity for Attio, but the migration was not a straight record copy. Their investment workflow depended on specific data relationships and views that Affinity structured one way and Attio structures differently.

The requirements went beyond record counts:

• Data model transformation. Affinity objects and fields needed to land in Attio's structure the way Delphinus actually wanted to work — not as a carbon copy of what they had before. This meant migrating into Attio's proposed data model, not replicating Affinity's.
• Historical date preservation. Attio controls its system-created timestamps, meaning original Affinity creation dates would be lost on import. For an investment firm tracking when relationships started and when deals entered the pipeline, losing those dates was unacceptable.
• Pipeline reconstruction. The full deal stage pipeline needed to be recreated with historical transitions intact, so the team could report on deal velocity and stage duration from day one.
• Custom object creation. A Syndication Partners workspace with two-way deal relationships needed to exist in Attio before the team could start working.
• Relationship integrity. Every company-to-person-to-deal-to-partner link had to survive the migration. Broken relationships would make the CRM useless for a team that operates on relationship context.

The ClonePartner Solution: Model-First Migration with Custom Object Construction

Rather than extracting Affinity data and dropping it into Attio field-by-field, ClonePartner built the migration around the target data model. The destination workspace was designed first, and then the transformation logic was written to land records in the right shape.

Data model transformations:

• Deal notes propagated to companies. In Affinity, Delphinus accessed deal-related context from company views. To preserve that workflow in Attio, deal notes were copied to the related company records during migration.
• Vertical information propagated from deals to companies. Deal-level Vertical data was transformed into a multiselect attribute on company records, giving Delphinus the filtering capabilities they relied on.
• Original dates preserved via custom attributes. Since Attio overwrites system timestamps, ClonePartner created "Date added (from Affinity)" attributes on companies and deals, storing the original creation dates where the team could reference them.
• Historic stage transitions extracted and transformed. Affinity's change logs were processed to generate approximately 2,700 historical stage transition records, each containing the company/deal, stage name, entered date, exit date, and days spent in stage. This gave Delphinus reporting data they had never been able to surface in Affinity itself.

Syndication Partners object:

ClonePartner created a custom object in Attio with 12 attributes — Firm Relationship, Firm Type, Investor Type, Sector Focus, Preferred Deal Stage, Status, Tier, Ticket Size, Average Ticket Size, Investor Network, Owners, and Amount. Each partner record was linked to the relevant company and person records. A two-way relationship was established between Syndication Partners and Deals, so the team could navigate from either direction.

Delta migration:

A final delta migration before go-live captured changes made in Affinity after the initial migration: 10 new people, 3 new companies, 43 updated list entries, 22 new notes, and 5 new files. The delta brought final counts to 739 companies, 1,989 people, 691 deals, 846 notes, and 730 files.

Data Cleanup and Validation

After the initial migration, ClonePartner ran a validation and cleanup pass:

• Enrichment duplicate resolution. Attio's automatic enrichment created duplicate company records. These were identified and merged.
• Auto-logged company removal. Irrelevant companies that Attio auto-created from email interactions were removed.
• Name alignment. Company names were aligned with their original Affinity values where enrichment had overwritten them.
• Location backfill. Missing location data was backfilled from Affinity records.
• Deal-level location attributes. Location data was created on deals from the related company information, giving Delphinus geographic filtering on their pipeline.

The Results: A Workspace Built for How Delphinus Actually Works

The migration delivered a production-ready Attio instance with:

• A fully recreated deal pipeline with 14 Won-stage portfolio investments.
• 2,700 historical stage transitions available for reporting — data that was locked inside Affinity's change logs before the migration.
• A Syndication Partners workspace with two-way deal relationships.
• Preserved original dates on every company and deal.
• Company records enriched with deal-level context (notes and vertical tags) matching Delphinus's actual workflow.
• A clean delta migration for seamless go-live.

The team started working in Attio with a workspace that reflected their investment process, not a dump of their old CRM's structure.

What Made This Migration Different

For investment teams evaluating CRM migrations, three details from this project are worth noting:

• Migrate to the target model, not from the source model. Copying Affinity's structure into Attio would have created a workspace nobody wanted to use. Building the migration around how the team wanted to work in Attio meant the workspace was production-ready on day one.

• Historical reporting data can be created during migration. Affinity's stage transition history was buried in change logs. By extracting and transforming it during migration, Delphinus gained reporting capability they never had before — deal velocity, stage duration, pipeline progression over time.

• Custom objects with relationships need to be built, not just imported. The Syndication Partners workspace was not a simple record import. It required object creation, attribute definition, and bidirectional relationship setup. Treating it as a construction project rather than a data transfer ensured it worked correctly from the first day.