Automation makes a good process faster. It can also make a bad process fail at scale.
Table of Contents
Every sales organisation is being asked what it plans to do with AI. The promise is appealing: more personalised outreach, faster research, better call summaries and less administration.
Those can be useful gains. But technology doesn't remove the need for commercial judgement.
If your targeting is wrong, AI may help you contact the wrong people more efficiently. If qualification is weak, an automatically populated CRM can give poor opportunities a more polished appearance. If nobody coaches the team, call summaries simply create another archive that nobody learns from.
Start with the constraint, not the tool
Before investing in any sales technology, define the problem in commercial terms.
Are salespeople wasting time on account research? Is CRM administration consuming valuable selling hours? Are managers struggling to identify coaching moments? Does the team lack timely insight into deal risk?
These are different issues, with different potential solutions and risks.
“We need AI” is not a business case. “We need to reduce research time while maintaining account relevance and improving meeting quality” is a much more useful starting point.
Four sensible places to apply AI
Account preparation
AI can consolidate public account information and help sellers prepare hypotheses and questions. The salesperson still needs to verify facts and decide what matters.
Meeting notes and follow-up
Accurate summaries can reduce administration and clarify next steps. Review them before they enter the CRM or go to customers, especially where commitments, figures and sensitive information are involved.
Coaching support
Conversation analysis may highlight recurring patterns in discovery, objection handling and talk-to-listen balance. Those signals should inform human coaching, not become an unchallengeable performance score.
Deal-risk identification
AI can help surface stalled opportunities, missing stakeholders or repeated close-date changes. It is most useful when the business has consistent stage definitions and reliable source data.
The questions leaders must ask
How will we measure commercial impact rather than tool usage? Who owns data quality? What customer information is processed and under what permissions? Can the team validate AI-generated claims? How do we prevent more automated activity from damaging buyer trust?
Pilot one clearly defined use case. Establish a baseline, track adoption and quality, then compare the outcomes. Retain human review where mistakes could affect customers or commercial commitments.
Fix the system AI reveals
Sometimes technology exposes just how inconsistent the sales operation has become. That is useful information.
If different people describe the same sales stage in different ways, fix the definitions. If every meeting produces a long summary but no clear next action, improve the meeting discipline. If outreach volume increases while replies deteriorate, revisit relevance and targeting.
AI should support a stronger commercial operating model, not disguise a weak one.
The competitive advantage won't come simply from having more AI tools. It will come from knowing which problems matter, designing better processes and using technology to reinforce them.
