Clari argues that poor targeting and unreliable account data create wasted sales effort before outreach begins, making ICP precision, qualification rates, and account prioritization more important than adding tactics or volume.
Revomnis take
Revomnis take Clari is right that outbound performance is constrained first by ICP precision, account data, qualification, and prioritization. Better sequences cannot correct a market that was chosen too broadly or data that misstates who the buyer is. Scale only amplifies the quality of the decisions made before the first send. Failure mode Teams often treat data quality as a cleansing project and ICP as a slide, then proceed with volume while both remain unresolved. They enrich thousands of records, accept uncertain job roles, and use weak fit signals because the database is available. Poor replies are then blamed on copy, deliverability, or the channel. Operating rules 1. Translate ICP language into executable account rules. Specify industries, company bands, geographies, operating characteristics, technologies, growth signals, exclusions, and disqualifiers. “Mid-market technology companies” is not enough to govern selection. 2. Separate fit data from contact data. Account fit answers whether the company should be targeted. Contact data answers who should be approached and through which channel. Both need confidence standards, but a valid email address does not make the account commercially relevant. 3. Classify before sequencing. Use account tiers and segment labels that determine research depth, buyer roles, message logic, proof, channel mix, and sending pace. Classification-driven targeting prevents one broad sequence from hiding multiple weak hypotheses. 4. Introduce volume only after a valid learning sample. Review delivery, reply categories, qualification, held meetings, and sales acceptance by segment. Expand only where the pattern is repeatable. Pause or narrow segments when the evidence is ambiguous. 5. Feed outcomes back into the data model. Record disqualification reasons, incorrect roles, stale signals, objections, referrals, meeting dispositions, and downstream sales feedback. The account model should become more accurate after each campaign, not remain a static list. How Revomnis applies it Revomnis treats fit and data as part of managed outbound, not as inputs handed off by the client and forgotten. We define the audience, build exclusion logic, validate records, classify accounts, and connect campaign outcomes back to the selection model. Managed infrastructure and execution begin only after the audience can be defended. Decision test / what good looks like A strong system can state what percentage of researched accounts met the agreed activation standard and why the rest were excluded. It can show performance by account class, not just across the whole campaign. If increasing volume is the main response to weak qualified conversations, fit and data are not yet under control.