Executive search

Attio for executive search: an evaluation plan for a relationship-led firm

Evaluate Attio for executive search with a synthetic mandate model, separate person and assignment context, plan-specific permissions and an import test plan.

By LaisyPublished

Attio's flexible CRM model raises a useful question for a search firm: can the team represent its people, client organizations and mandates without losing the context that belongs to each assignment?

The answer needs a model and a test. A flexible list is not automatically a confidential search workspace, and a custom object is not automatically a complete mandate workflow. The firm should examine record meaning, access, review and maintenance using a small appropriate example before making a wider decision.

This is Laisy's evaluation plan, based on Attio's public documentation checked on November 3, 2026. It is not a deployment report or supplier endorsement. No live Attio workspace, paid plan, import or Laisy integration was used for this research.

TL;DR

  • Model a person separately from their context in each mandate.
  • Choose attributes, views, lists and objects according to what the information means.
  • Verify effective object and list access in the actual plan and roles; names and filters do not establish confidentiality.
  • Test matching and mapping with synthetic import material before considering any permitted live data.

Copyable Attio evaluation plan

TestQuestion to answer
Record meaningWhat is one person, company, mandate and mandate-person involvement?
Shared factsWhich values apply to the underlying person or organization?
Assignment contextWhere do stage, evidence questions and client-specific decisions belong?
StructureIs the task served by an attribute, view, list or distinct object?
Plan conditionsWhich required structures and controls are available in the proposed plan?
Effective accessWhat can each intended role actually see and change?
ImportWhich unique values identify records, and which columns map to which attributes?
ReviewCan another consultant recover sources, uncertainty and the next decision?
MaintenanceWho owns configuration, corrections and later model changes?
DecisionWhich requirements were demonstrated, unresolved or outside the test?

Begin with the meanings, not the interface

Describe the real-world entities and relationships before creating structures. A person can be involved in several mandates. A client organization can have several contacts whose authority differs by assignment. A candidate-related note can concern a particular role rather than a permanent characteristic of the person.

Use the mandate kickoff checklist to identify the assignment's business purpose and decision boundaries. Then ask which parts are durable shared facts and which are specific to this mandate.

Avoid a universal “candidate status” if the same person can have different simultaneous contexts. Likewise, a company-level contact field may not establish who can approve a particular role boundary. The model should preserve those distinctions rather than force the operating process to fit a convenient column.

Choose the appropriate Attio building block

Attio's data-model guide distinguishes attributes, views, lists and objects. Views change how records are displayed; list entries can hold process-specific values separately from the underlying object attributes.

That distinction can help a search team test where assignment-specific information belongs. A view may be sufficient to display a subset of existing data. A process needing its own values may require a list. A distinct entity with its own relationships and reporting may justify a custom object.

These are evaluation choices, not a prescribed Attio architecture for every search firm. The same documentation states that custom objects and custom relationship attributes require Pro or Enterprise, with object limits depending on plan. Verify the current conditions against the actual intended model before buying or designing around a feature.

Keep assignment evidence attached to its context

A research account should retain its source, date and the question it answers. If it belongs to one mandate, the model should make that scope visible rather than spreading the assessment across every future use of the person's record.

The candidate research-note guide supplies a practical review standard. Can a colleague distinguish a supported fact, attributed account, inference and missing information in the proposed model?

Test a changed instruction as well. If the client revises the geographic scope, can the team find which work depends on the old boundary? The test does not need an automatic answer, but it needs a clear route to the relevant evidence and responsible person.

Verify object permissions in the actual plan

Attio's object-access documentation describes plan-dependent controls and the most-permissive applicable grant across workspace, team and member settings. A narrower individual setting does not simply cancel a more permissive workspace grant.

That behavior makes an actual role test important. Define the intended readers and actions, then ask an authorized administrator to demonstrate the effective result with synthetic records in an appropriate evaluation environment. Do not infer access from a view filter or a label alone.

Read the action table too. Object “read only” does not mean that every possible related action is prohibited; the documentation lists notes, tasks, files and other behaviors separately. Verify the specific operations relevant to the firm's boundary rather than relying on the ordinary-language impression of the label.

Review list access separately

Attio's list-permission guidance distinguishes access to list data from access needed to change underlying object attributes. Its documentation also describes administrator management access and the interaction of workspace, team and member grants.

A private mandate list should therefore not be assumed to make every shared person or company attribute private. Decide which information is intended to be shared at the object level and which belongs within the assignment context. Verify the actual result with the intended roles.

Do not put restricted assignment detail into a shared object field and expect a hidden list to protect it. Conversely, do not assume a list-specific value changes the underlying person record everywhere. The distinction is part of the evaluation, not an implementation detail to discover after data is loaded.

Worked example: one person across two mandates

The following fixture is invented. It gives an administrator and a consultant specific questions to demonstrate; no Attio configuration or access test has been run.

Northbank is a fictional client. Alex Rowe is being researched for its finance and operations mandates. These labels are test identifiers, not actual Attio record IDs.

Test itemFilled contentRequired separation
Person P-07Alex Rowe; current role recorded as Finance Director; fixture source S-01, dated October 20Shared factual record, with the source attached
Mandate M-101Northbank CFO; client decision owner Priya; consultant owner LeeFinance mandate instructions and authority
Mandate M-102Northbank COO; client decision owner Daniel; consultant owner MorganOperations mandate instructions and authority
Involvement P-07 → M-101Research stage; open question: evidence of group reporting responsibility; Lee to clarifyQuestion and stage belong only to the CFO assignment
Involvement P-07 → M-102Paused; open question: direct responsibility for plant operations; Morgan to clarifyDifferent question, stage and owner for the same person

Run three demonstrations against a proposed model:

  1. Correct Alex's shared role from Finance Director to Group Finance Director using a replacement fixture source. Both assignment views should resolve to the corrected person fact without changing either open question or stage.
  2. Move the CFO involvement to Ready for consultant review. The COO involvement must remain Paused with its own history and owner. If the design uses one shared candidate-status field, it fails this requirement.
  3. Give an intended finance-assignment reviewer the agreed access, then check the shared person facts, CFO context and restricted COO context separately. Record the actual visible data and permitted actions. A hidden view is not evidence that the underlying access is restricted.

Ask a second consultant to identify the next question and responsible owner for each mandate without help from the model's creator. A pass requires the correct answer and a route back to its source. An ambiguous field name or undocumented naming convention is a repair item.

The decision record should have separate rows for documented capability, demonstrated in this configuration and still unresolved. Public documentation can support the first row; only the team's actual test can fill the second. Do not turn this fictional fixture into a claim that Attio has passed it.

Test import matching and mapping deliberately

Attio's CSV-import guide explains column mapping, unique attributes and review/preview steps. Matching behavior matters because an import can create or update underlying records as well as process-specific entries.

Use a small synthetic file designed to expose the intended relationships. Check identifiers, field formats and which destination each column changes. Inspect the preview rather than assuming a matching-looking name identifies the same person or organization.

Use a controlled environment for any authorized import. Retain the input file, mapping and observed result so a second person can explain which records changed.

Test continuity and operating effort

Use the handover guide to test whether another consultant can continue the work. They should find the applicable instruction, evidence, unresolved question and next owner without the model's creator explaining every field.

Record the operating work alongside the demonstration: who captures new information, corrects sources, maintains fields and permissions, and trains the next consultant. Obtain the price for the required plan and any additional services, then include setup and ongoing administration in the comparison. This article supplies no price or savings estimate.

Attio also describes AI workflows for account research, briefs and other sales tasks. Those are vendor-described workflows, not evidence that a particular executive-search task has passed this evaluation. Do not assume a flexible CRM lacks AI when comparing it with another tool.

If the firm's existing Claude workflow already produces an acceptable preparation note, use that reviewed note as a baseline. The next buying question is whether a proposed workflow improves capture, correction, handback or administration enough to justify its privacy review and cost. The pilot success criteria help test those tasks alongside existing tools. They do not establish an Attio connector or automatic writeback from Laisy.

Finish the evaluation with demonstrated requirements, open questions and the next decision. The useful result may be a small workable model, a request for further supplier evidence or a conclusion that the operating effort does not fit the firm's needs.

Frequently asked questions

Is Attio a ready-made executive-search platform?

This article does not establish that. It evaluates whether its documented flexible CRM model can support specified search-firm responsibilities. Any required workflow, integration or assurance needs its own current evidence.

Should every mandate become a custom object?

Not automatically. Start with the meaning and required relationships, reporting and access. Attio distinguishes lighter structures from custom objects, and plan conditions apply. Test the chosen model against actual tasks.

Does a private list protect all information on its people?

Do not assume that. List and object permissions govern different information and actions. Verify the effective role access and avoid storing restricted assignment context in an inappropriate shared field.

Does this plan include a Laisy integration with Attio?

No. It describes an evaluation method based on public documentation. Any alongside task, selected inputs or handback would need to match actual supported capability and be agreed and reviewed separately.