There’s some variation of this exact conversation happening in every org right now.
Someone on the leadership team pulls up a rogue stat about AI adoption, somebody else shares a tool they’ve been testing or tinkering with, and the room begins to split between the people who are AI champions, those who think the whole thing is overhyped, and the people terrified they’re being replaced by it.
The truth?
The first two are both right, and unless the third is a receptionist at a remote-first org or a data-entry clerk griping about their stapler being stolen, they likely just need to learn more deeply about AI.
It is an absolute fact that AI is overhyped in some of the ways it’s being shown in the demo vs. how it actually looks, acts, and reasons when deployed. Conversely, the AI champions clearly see the fact that the window to leverage AI before your competitors figure it out is already closing.
But first, with so many definitions floating around, let’s get aligned on what leveraging AI actually means in a sales context.
Today, a rep can use AI to qualify prospects, draft “personalized” outreach, send the email, and set up the follow-up sequence based on a single, lackluster prompt.
It’s automated, scalable, and efficient.
…or is it?
Reality is the buyer gets inundated, and the rep torches the credibility of whatever email domain they’re blasting droves of messages from, inviting firewalls to flag more and more of their future outreach. Then, the prospects either buy or they don’t – it’s binary. The AI collects the data, changes sequence timing or messaging, and repeats on a new batch of unsuspecting souls.
That’s using AI in sales. Not leveraging it.
The distinction matters, because the behavior above – volume-first, automation-heavy, relationship-non-existent – is exactly what’s burning sales down as we know it.
Leveraging AI means identifying the tasks that consume your reps’ time without moving revenue, and systematically removing them from the equation, so the people you hired to build relationships, navigate complexities in educating and penetrating the market, and close business can, well… do that. It is the inverse of what’s described above. Instead of automating the part of sales that should be human – the outreach, the relationship building, the embedded trust and brand development – you automate the part that shouldn’t require a human at all.
Things like the research from a hundred sources for a thousand contacts, data entry and logic logging, buy-signal monitoring like watching the headlines for M&A activity or hiring surges, and follow-up logistics.
In B2B tech and SaaS, the average cold email response rate sits between 3-5%. That’s it.
Then because it’s so easy to do, this space – and everyone’s inbox – is flooded.
Selling is simple, not easy. As a result of leveraging AI, reps show up sharper, better enabled, with more time to do what they were hired to do – grow the business.
True Enablement
The question isn’t whether your team should use AI. That ship has sailed.
According to Salesforce’s State of Sales 2026, 87% of sales organizations already use AI in some form. Gartner puts the figure at 89% of revenue organizations, up from just 34% in 2023.
The question leaders need to answer today is whether you’re deploying AI where it actually moves the needle, or automating the wrong things quickly?
Put more broadly, are you valuing what you can already measure, or measuring what you can value?
All the major publications and studies say it over and over again.
The modern seller spends roughly one-third of their day, their week, their year on administrative tasks.
Not selling. Not customer interactions. Not engagement. Not collaboration. Not community involvement.
Administrative, data-entry drone work.
Salesforce’s research puts it at 28–30%. Forrester’s Activity Study, which tracked 3,031 sales reps across industries, found the average rep burns nearly two full days per week on administrative work alone.
The remaining 70% of the week disappears into internal meetings, prospect research, email management, and follow-up coordination; almost none of which directly drives net-new revenue.
True enablement means using AI to reclaim that time and redirect it toward a very high-value activity – selling.
I say those who think the whole thing is overhyped are right because AI doesn’t make a mediocre rep into a great one. I don’t care how good the demo was. I don’t care if it gives real-time recommendations and call scoring. It does not make a mediocre rep into a great one from a skill perspective.
However, it can give your whole team more time to sell by doing some of the other, admin-heavy tasks they currently carry the burden of, and free your managers (you know.. the ones you hired to grow your mediocre reps into great ones?) up to spend more time coaching and enabling their reps with forward-looking insights instead of lagging indicators.
Where That Plays Out Practically:
- Prospecting Research: No matter the territory or how it’s broken out, a rep spending 5-15 minutes per contact googling a company for news, finding the right contacts, reviewing LinkedIn, reviewing the 10k report, checking job postings, or any interviews with leadership on a random podcast, and piecing together context is a rep spending hours a week on research alone. AI agents do this in seconds and surface a consolidated brief, or compile results onto a dynamic dashboard that bolsters productivity while logging all actions into the CRM on behalf of the seller. This means reps can spend more time focusing on quality in their pipeline.
- Lead Scoring & Prioritization: This is one of my personal favorites. Having an agent apply pre-defined ICP and buying-signal criteria consistently across volumes of leads, surfacing hot leads out of a haystack daily.
- CRM hygiene: Manual data entry is the most universally despised task in sales. It’s also one of the most consequential when it doesn’t happen, or when it happens ad hoc. Bad data spills over into bad forecasting, bad territory planning, and useless pipeline reviews. AI that auto-logs call activity and dispositions, updates deal stages, and captures next steps removes the task almost entirely rather than just making it slightly faster. This transforms the CRM from a chore into a revenue-enabling lynchpin.
This is what true enablement looks like: not automating outreach at the cost of your domain reputation, but reclaiming the non-selling hours that represent the majority of every rep’s week and arming them with hyper-personalized insights so they can outperform their former self.
Your Buyer Has Already Moved
If the internal efficiency case isn’t enough, consider what’s happening on the other side of the table. Your buyers are using AI far more aggressively than most sales leaders realize, and it’s reshaping how deals get started before your team is to that point in their cold-list.
Forrester’s 2026 Buyers’ Journey Survey, which collected responses from nearly 18,000 global business buyers, found that twice as many buyers named generative AI or conversational search as their most meaningful research source compared to any other source in the study; outranking vendor websites, product experts, and sales representatives themselves.
The proportion of B2B buyers using AI in their purchase process has grown from 89% in 2025 to 94% in 2026, which means by the time you’re reading this it’s even higher than that.
G2’s March 2026 survey of 1,076 B2B software buyers found that 71% now rely on AI chatbots for software research, and 83% report feeling more confident in their final choice as a result.
Their Chief Innovation Officer put it plainly:
“The Yellow Pages compressed the market into the big book. Google compressed it into the first page of results. Now, AI chatbots are compressing it into a single answer.”
What buyers are specifically using AI for is worth understanding:
– 54% use it to research product information
– 55% use it to compare vendors against each other
– 47% use it to build internal business cases before engaging any vendor
Every single one of these is a sales function, and today, they’re happening entirely outside your sales funnel before your reps ever know the opportunity exists.
The downstream impact is that B2B buyers now complete 70–80% of their research before they even contact a vendor.
This makes inbound leads more valuable than ever before, because it means by the time someone fills out your demo request form, they’ve already formed a preference.
What it also means for us in sales is that the vendor who wasn’t visible in the AI-generated answer for their category’s buying questions wasn’t considered.
Not rejected, not beat out by a competitor – never encountered.
Now that used to be marketing’s headache to figure out, but it has direct implications for sales leaders beyond marketing now.
For just about any given deal in your forecast, your champion is building their internal business case using AI. Your economic buyer is running competitive comparisons using AI. Your multi-threaded stakeholders are arriving at conversations with pre-formed views shaped by what AI told them about your category, your competitors, and you.
The Agentic Layer
While Generative AI is what took the world by storm in ’23-’24, agentic AI is the 5th generation on the AI continuum.
Simply put, Generative AI responds to prompts. One prompt. One answer. Rinse and repeat.
Agentic AI pursues a goal, and along the way it reasons and takes initiative. It monitors signals, executes multi-step workflows, routes opportunities, and takes actions in the world outside its software platform – all with minimal or no human instruction at each step.
The agentic AI market was valued at roughly $7–9 billion in 2025–2026 and is projected to grow at a compound annual rate of 40–46% through the end of the decade. Those numbers are less important than what’s driving them: organizations are discovering that the same way automation can speed up tasks, agents can remove entire categories of manual work from the workflow entirely.
Bain’s 2025 analysis found that AI-assisted sales teams see 30% productivity gains and 68% shorter deal cycles. The same analysis estimates that AI could double the percentage of time sellers spend actually selling. This directly translates into a structural re-imagining and asset re-allocation of how a sales organization operates.
Deloitte Digital’s February 2026 study of 1,060 B2B suppliers and buyers found that digitally mature companies — using AI extensively and systematically — exceeded annual sales growth targets by 110% more than low-maturity competitors. They were also five times more likely to use agentic AI at all.
There’s a two-tier market forming. The gap is compounding quarter by quarter.
What Sales Leaders Should Be Doing Right Now
You don’t need to deploy the fastest, but you do need to consider some key elements, as this is unlike any deployment opportunity preceding it.
Here’s where to focus:
Start with the time audit, not the tool list.
- Map where your reps’ non-selling time actually goes before selecting technology.
The 70% of the week that isn’t selling isn’t uniformly distributed; it concentrates in specific activities that vary by team, role, and motion. Admin, research, internal coordination. Find the biggest drain first, then find the tool that addresses it. - Aim small, miss small.
It can be very tempting to produce a super-agent who will handle all tasks you have in mind, but there are a few problems with that approach. Economic: this setup becomes very (very) costly, especially depending on things like the chosen LLM and orchestration platforms in your tech stack. Best practice is to create the Minimum Viable Product (MVP) for the smallest use case which will have the greatest positive impact. - Arm your reps with deep insights about their prospects instead of battlecards about their competitors.
The outreach automation arms race is producing diminishing returns industrywide. Email deliverability is getting harder. Teams will excel in this environment if they simply show up to conversations with genuine context and genuine curiosity. Use AI to ensure your initial messaging is better prepared, not just to manufacture more messages at scale. This can be done by creating an interactive, dynamic dashboard that pulls together buying signals and organizes leads according to their quality and quantity of signals - Build for the buying journey, not just the sales process.
One massive advantage of agentic AI is that it enables us to the point we can re-imagine entire workflows. Look at each step in your process – not your sales cycle – your process. Do this with your buyer’s experience in mind. If you can tap a client on the shoulder to pick their brain about use cases that would be most valuable to them, that’s even better. - Ready your team, your org, and your culture.
IBM’s State of Salesforce 2025–2026 report found that only 33% of AI initiatives meet ROI expectations, with 53% citing poor data quality as the primary reason. The technology is accessible. Organizational readiness is the bottleneck. Whether bad data integrity or fear that employees are training their replacement, org readiness becomes a top-of-mind initiative for leadership everywhere. - Take action.
Planning for six months is no longer proper due diligence. It has become an irresponsible and borderline negligent way to lead when new, genuinely useful tech is launching and evolving weekly. Fact is, a highly useful agent with a sharp focus can be built in about a weekend. That means our timelines need to compress with that capability shift. Three weeks is the new deployment best practice for shipping your MVP. In comparison, that means we can launch eight MVP agents in the same time it used to take to conduct due diligence on a single vendor before we could sign off.
AI has already been shaping B2B sales for a while now, so the question becomes whether you are shaping how it happens on your team or waiting to react to what your competitors figure out first? Your buyers aren’t waiting.If you’re not sure how to audit where your rep’s time is going and what that’s actually costing you, you’re not alone.Start with the FrictionFILTR and calculate it in less than 5 minutes.