Before you skim this, one point to make:
This isn’t a “tips” article (sorry to disappoint!).
It talks to common pitfalls/misconceptions and the impacts it has on sales team and revenue outcomes. Nothing groundbreaking but it’s until it’s written or spoken out loud that it becomes the obvious.
Read it once, reflect on it, come back to it before your next AI rollout, discussion or own reflections. Think about what it would take for the below to be well executed within your world.
I hope you enjoy this edition and take something away from it.
AI increases activity but not performance in remote/hybrid sales teams
AI removes friction from sales work. This creates a surge in activity, but revenue performance is constrained by a few hard bottlenecks that AI does not fix on its own: unclear positioning, weak data, inconsistent process, and thin coaching.
In remote/hybrid teams, those bottlenecks are amplified.
McKinsey’s latest State of AI survey highlights the underlying issue: many organisations are still early in the management practices required to capture value, including tracking clear KPIs for gen AI solutions.
What ‘activity lift’ looks like in practice
AI tends to lift what is easiest to scale:
- More outreach messages drafted
- Faster account and contact research
- More meeting notes and follow-ups logged
- More proposals and decks produced
This is real throughput and sounds great, right?. But it’s not the same as:
- Better conversations
- Higher-quality opportunities
- Higher win rates
- Shorter sales cycles
- Higher average selling price
Bain makes a similar point in sales terms: applying AI to existing processes often creates only small “micro-productivity” gains, because new bottlenecks appear and inefficiencies get automated (and scaled) rather than removed.
So why doesn’t performance automatically follow…
1) AI scales the message regardless of systems quality
If your ICP, problem framing, and differentiation are fuzzy, AI will help you produce more average outreach at high volume. Buyers experience this as noise.
HBR has warned for years (well before genAI) that the best virtual sellers use technology to deepen relevance and insight, not to spray generic pitches – classic scatter-gun approach. The channel makes volume easy but relevance…that what will always win.
Effect on sales teams: fewer in-person cues and less informal context means reps lean more on scripts and templates and lose skill. AI accelerates that tendency.
2) ‘Workslop’ becomes a pipeline drag
When AI-generated content looks polished but adds little substance (wee see that a lot right here on LinkedIn!), it increases internal and external churn: more messages, more follow-ups, more meetings… with no improvement in conversion.
HBR has recently labelled this risk directly: lots of AI-shaped output can destroy productivity when it creates low-value content that others must sift through.
Remote/hybrid effect: disjointed work increases written communication. If AI inflates low-quality written output, the drag compounds.
3) Activity metrics are easier than performance metrics, so leaders over-measure the wrong thing
When AI tools roll out fast, many teams default to:
- Number of emails
- Number of calls
- Number of sequences started
- CRM hygiene completion (aka “CRM adoption”)
Having been in remote sales for 15 years, I’ve never seen activity-focused metrics truly move the needle. Doing more doesn’t always correlate with selling more (quality vs quantity debate).
McKinsey’s survey data shows a common weakness: relatively few organisations report tracking well-defined KPIs for genAI solutions. That makes it easy to celebrate adoption and output, while missing impact.
Gartner’s sales research adds a warning: by 2028, AI agents may be everywhere, yet fewer than 40% of sellers are expected to report productivity improvements because adding more tools can overwhelm sellers, and increase burnout, if not integrated into coherent workflows and measures.
Effect on sales teams: distributed teams often use dashboards to “see work.” If the dashboard rewards volume, the system will produce volume.
4) Data quality and process variance cap the upside
Sales data is typically fragmented across CRM, enablement, product usage, support, and marketing systems. Bain calls out data cleanliness and governance as a key hurdle for AI in sales, alongside the reality that sales processes vary widely by region and individual.
Effect on sales teams: more tools, more handoffs, more systems. Variance goes up unless RevOps actively standardises.
5) Coaching becomes the constraint (and AI often worsens the manager load)
AI can increase rep activity without improving judgement. That puts more pressure on managers to:
- Inspect deal quality
- Correct messaging
- Coach discovery and next steps
- Keep standards consistent
If managers are already stretched (I previously wrote about them only spending 14% of their time on coaching), higher activity can reduce coaching quality, which reduces performance.
BCG’s sales-focused AI perspective is explicit: getting outcomes at scale requires an operating model and a heavy focus on people and process change, not just the tech layer.
Again, for those who follow my newsletter, you will know that leadership work starts by redesigning workflow to be AI-led, not human-led. That takes an awful lot of planning and change management.
6) Buyer behaviour is moving digital-first, which rewards precision, not volume (that’s why I believe remote sales – coupled with digital marketing – is primed to become the primary channel for B2B transactions in any given industry!)
Gartner has long projected that the majority of B2B sales interactions are shifting to digital channels. Digital channels reduce the cost of contact, but they also raise the bar on relevance because buyers can ignore you instantly.
Bottom line: digital-first selling makes ‘more activity’ less defensible as a strategy. You need higher-quality touches.
The leadership problem hiding underneath
It’s quite simple really… If you introduce AI into an unchanged sales system, you scale the system’s current behaviour.
- Strong system à AI amplifies performance
- Weak system à AI amplifies waste
BCG’s broader research on AI value supports this pattern: most companies struggle to move from pilots to tangible value, while a small group of leaders build the capabilities to scale outcomes.
This is not an AI problem; it’s a leadership system problem.
Practical leadership implications for remote/hybrid sales
1) Redefine productivity (stop rewarding motion)
Replace ‘more activity’ targets with a small set of outcome-linked standards:
- % of meetings that convert to a qualified next step
- Pipeline created per rep per week (with quality checks)
- Win rate by segment / motion
- Cycle time by stage
- Average deal slippage (time + stage regression)
Then track AI’s contribution against those measures (not tool usage alone). This aligns with Gartner’s recommendation to redefine success metrics to capture both human and AI contributions.
2) Put ‘quality gates’ into the workflow
Examples:
- Outbound copy must reference a verified trigger + business impact
- Discovery must produce a written problem statement + quantified cost
- Every opportunity must have a mutual action plan by stage
- Proposals require an executive summary tied to customer outcomes
AI helps create the structure and content and Leaders enforce the gates.
3) Standardise the sales process enough for AI to help
You don’t need rigid scripts. What you do need are:
- Consistent stage definitions
- Required exit criteria
- Clean fields that matter (ICP fit, use case, value hypothesis, next step to name a few). Shape these to whatever matters most to your business but these would be a great foundation.
Otherwise AI has no stable structure to improve.
4) Upgrade RevOps from reporting to ‘system design’
Deloitte Digital’s B2B sales research points to RevOps as a growth driver that aligns GTM functions and reinvests in capabilities and technologies that amplify seller effectiveness.
Before being in MedTech I was in SaaS and I think this is valuable consideration.
In an AI rollout, RevOps should own:
- Where AI sits in the workflow
- Which tools are approved (and which are removed)
- How data is governed
- How impact is measured
5) Protect selling time, but don’t confuse time with performance
Bain notes sellers may spend a minority of time actually selling, and AI can free up time by removing low-value admin.
Leadership job: reinvest that time into the few behaviours that drive performance in remote/hybrid.
That means:
- Better account plans
- Sharper discovery
- Multi- stakeholders management
- Tighter deal orchestration
6) Create an “AI coaching cadence”
A simple operating rhythm example for managers:
- Weekly: 3 deal reviews focused on value hypothesis + next step quality
- Weekly: 2 call/visits reviews focused on discovery depth
- Bi-weekly: outbound quality clinic (real examples, not theory)
- Monthly: metric review linking AI usage à conversion à revenue
AI can summarise and tag calls/visits. Managers must still coach judgement.
Takeaways to reflect on:
Don’t ask, “how do we use AI more?” but ask:
- What behaviours do we reward?
- Where are our quality gates?
- What do we want managers to actually coach on?
AI doesn’t replace sales leadership. It exposes it and can amplify it for the better.
Until next time.