Customer lookalike account research

This playbook defines the end-to-end process for identifying, qualifying, and staging high-probability target accounts modeled after verified successful customers. One run evaluates a discrete cohort of closed-won reference accounts, derives verified operational criteria, identifies matching candidate companies from available sales platforms, deduplicates candidates against active CRM opportunities and territory exclusions, and submits qualified candidate dossiers to the accountable sales owner for review. External CRM account creation occurs only after explicit owner sign-off. If direct CRM API connections are unconfigured, candidate data is prepared for manual import export without claiming CRM record completion. All evidence, source discrepancies, and verification uncertainties are preserved.

• Ingest the designated closed-won benchmark accounts from the primary CRM (Attio, Pipedrive, or HubSpot) or provided export. Extract verified attributes including industry sub-sector, employee count, primary operating geographies, contract scale, recorded tech stack signals, and sales cycle length. Filter out accounts with disputed implementation outcomes, atypical non-standard deal terms, or incomplete firmographic profiles, recording the specific exclusion reasons in the case audit log.
• Synthesize a structured profile matrix from the benchmark accounts. Identify the core common operational attributes, distinguishing established factual records (such as verified employee count ranges and billing countries) from subjective sales notes or unverified enrichment tags. Establish the qualification criteria, allowable variation bands, and negative disqualification indicators (such as non-commercial entities, B2C operations, or unsupported regulatory jurisdictions).
• Query configured enrichment providers and company intelligence sources (such as Apollo) or import candidate prospect records to identify potential lookalike companies meeting the profile matrix. Collect candidate attributes, including domain, verified employee count, estimated revenue, primary tech stack elements, recent hiring activity, and key executive roles. Where multiple data sources disagree on headcount or location, retain both values with source timestamps rather than averaging or discarding conflicting points.
• Information request. If the benchmark customer cohort exhibits conflicting firmographic signals or missing core fields (such as missing employee bands or ambiguous business models), request clarification from the sales operations lead or account executive before proceeding with external candidate searches. Specify the exact missing or conflicting fields and provide the observed options. If clarification is provided, update the profile matrix; if the owner declines or confirms data is unavailable, proceed using only verified attributes and flag the wider tolerance in the case audit.
• Perform deduplication and territory conflict screening against the active CRM database. Match candidate company domains and legal names against existing customer accounts, active pipeline opportunities, recently lost deals, and excluded target lists. If a candidate matches an account with an active open deal or an existing assigned rep, exclude the candidate from consideration and log the active owner and deal stage. If a candidate matches a dormant or churned account, retain the candidate but flag the prior engagement history in the dossier.
• Check corporate parentage, subsidiary affiliations, and competitor status for each deduplicated candidate. If a candidate company is a subsidiary of an existing enterprise client or active prospect, record the hierarchy link and determine whether the subsidiary operates autonomously under territory rules or requires consolidation into the parent account team. Disqualify any candidate operating as a direct vendor competitor or partner entity.
• Assemble the structured candidate evidence dossier for qualified accounts. For each candidate company, summarize the matching benchmark attributes, verified company signals, observed tech stack overlap, recommended priority tier (Tier 1 high-fit, Tier 2 moderate-fit), deduplication status, and recommended sales rep assignment based on defined territory rules. Highlight any unverified data points, recent executive turnover, or corporate relationship notes.
• Approval. Present the prepared lookalike candidate dossier to the designated sales owner or sales operations manager for formal qualification and account assignment. Present each candidate company alongside its benchmark match rationale, firmographic signals, potential pipeline overlaps, and proposed account owner. The sales owner may approve individual accounts, request adjustments to priority tiers or rep assignments, or decline candidates based on subjective territory knowledge. The owner must confirm the approved candidate list, and silence remains waiting without implied consent.
• Process the owner's review decisions. For approved candidates, prepare the standard CRM account records, populating verified company name, domain, employee band, industry classification, lookalike benchmark reference tag, and assigned sales owner. For declined candidates, record the explicit rejection reason (such as out-of-market, recent bad-fit discovery, or territory reassignment) in the audit log to refine future research runs. If the owner requests modifications to rep assignment or tier, update the candidate profiles accordingly.
• Stage and create approved candidate account records in the CRM (HubSpot, Pipedrive, or Attio). Before executing any creation call or retry, inspect existing records by exact domain and company identifier to prevent duplicate accounts resulting from transient network timeouts or partial writes. After writing each approved record, immediately read back the account entity to verify that domain, properties, benchmark tags, and owner assignments match the approved dossier. If a write fails or verification read-back cannot be obtained, do not treat the write as successful; keep the affected account in an unconfirmed status awaiting successful read-back receipt or obtain the sales owner's explicit deferral or withdrawal with a named follow-up owner. A generated CSV export is recorded as a prepared manual import file and is not completion proof of a CRM write.
• Conclude the research case by delivering a comprehensive final summary report to the sales owner and sales operations lead. The report lists all approved accounts with verified CRM record read-backs, declined candidates with preserved rejection reasons, excluded pipeline duplicates, and any unconfirmed writes with their designated follow-up owner. For an example of operational judgment during deduplication, consider a mid-market freight tech company that evaluates twenty lookalikes: one candidate, Apex Freight Logistics, appears ideal on employee count and tech stack but matches a corporate subsidiary of an enterprise account currently in active contract renewal. The agent flags this hierarchy link in the evidence dossier; the sales AE reviews the checkpoint and declines outbound research, re-routing the subsidiary record as an expansion opportunity for the enterprise team. The remaining approved candidates proceed to verified CRM creation, completing the run with zero unwanted account conflicts.