AI has taken over go-to-market, and that’s not hyperbole.
In Salesforce’s 2026 State of Sales report, 87% of sales organizations said they now use some form of AI for prospecting, forecasting, lead scoring, or drafting emails. More than half use it specifically to prospect. Marketing isn’t far behind. Open your inbox, and you can feel it: The volume, polish, and personalization are up.
I’m not here to tell you AI is bad. I use it every day. It’s the most important thing to happen to our profession in a decade. But we’re being lazy with it.
We’ve pointed all this incredible technology at one goal: Do more, faster. Send more emails. Build more lists. Spin up more variations of the same nurture. We use AI to scale our go-to-market volume and, in the process, quietly automate the humanity right out of it.
The person who pays for that is the buyer. They’re sitting under a growing mountain of automated, personalized-sounding yet instantly forgettable slop. They learned to ignore all of it.
The pendulum is about to swing hard the other way. Back to relationships. Back to actually knowing the people we sell to. The teams that win won’t automate every part of the GTM process. They’ll use AI to scale the right parts of the process so humans can do the parts that actually matter. Here’s why 1-1 ABM is the model for getting there.
10X your SEO with Semrush for Enterprise.
The world’s most powerful SEO platform, purpose-built for Enterprise.
Request demo
We automated the volume and forgot the point
The channels we’re flooding with AI are the ones buyers have already tuned out. 6sense’s 2025 Buyer Experience Report, a study of nearly 4,000 B2B buyers, found that unsolicited SDR and BDR outreach plays a minimal role when buyers first engage.
Buyers initiate roughly 80% of first conversations themselves, and they do it on their own timeline, after they’ve already done their homework.
Now layer AI on top. Salesforce frames the real problem as a capacity gap: The average seller now spends only about 40% of the week actually selling, with the rest lost to admin and manual research.
What’s the industry’s answer? Hand the channel to AI and run the volume up. Fifty-five percent of sales pros already use AI to prospect, and another 38% plan to. We’re pumping more output into the exact place buyers already stopped looking.
The personalization doesn’t save us because, at scale, it stops being personalization. It becomes a pattern.
Every buyer alive can now spot the shape of an AI-written opener in half a second. It’s the “I saw the great work you’re doing at [Company],” the manufactured compliment, the pivot to a calendar link. The more we generate, the faster people learn to delete on sight.
That’s the trap. We measured success by how much we could produce. The buyer measures it by whether any of it was worth their attention. Those two scoreboards have never been further apart.
Relationships still decide the deal, and the data is brutal
If volume doesn’t move buyers, what does? The same 6sense research is the clearest answer I’ve seen, and it should be tattooed on every GTM leader’s wall:
- 95% of the time, the winning vendor was on the buyer’s Day One shortlist, chosen before any active “campaign” reached them.
- In 85% of successful purchases, the buyer had prior direct experience with the vendor they chose.
- 75% of buyers say they personally know sellers at the vendors they evaluate.
Read those again. The deal is decided before the first conversation. It’s decided by familiarity, reputation, and prior experience. By relationships. Not by who sent the most emails last quarter.
This is what the “automate everything” crowd keeps missing: AI is accelerating the swing back to humans, not away from them.
In a Gartner survey presented in May, 69% of B2B buyers said they prefer to validate AI-generated insights with a human sales rep at the moments that matter. Buyers will happily let a machine help them research. They still want a person they trust to tell them it’s true.
You can’t automate your way onto that Day One shortlist with volume. You earn your way onto it slowly, with trust built over time, through genuinely useful content, real conversations, and being a name a buyer already respects when the need finally surfaces.
This relationship building is what AI slop actively works against. Every forgettable, automated touch doesn’t just fail to build the relationship. It chips away at one. It teaches the buyer that you’re noise, and noise never makes the shortlist.
What 1-1 ABM actually is (and what it isn’t)
In account-based work, you generally operate across three tiers:
- 1:Many: Broad reach across a wide account set.
- 1:Few: Clustered by segment or use case.
- 1:1: Bespoke, highest-touch, aimed at a small number of your most important accounts.
1-1 ABM is the top of that pyramid. It’s the fewest accounts and the deepest level of care. This tier is all about account plans, real research, custom messaging, and human-to-human engagement with the actual buying group.
By design, 1:1 ABM is the opposite of scale. It costs the most per account and returns the most when you do it right because it’s the only motion that consistently builds the kind of relationship the 6sense data says decides deals.
People hear “AI plus ABM,” and their first instinct is to make 1-1 cheaper and automate it until it becomes 1-many wearing a 1-1 mask. They generate a thousand “personalized” emails and call it account-based.
That’s not 1-1 ABM. That’s slop with better merge fields. It kills the one thing that made the motion worth running. The opportunity with AI isn’t to make 1-1 cheaper. It’s to make 1-1 possible at more than a handful of accounts without gutting what makes it work.
Get MarTech Insights That Matter
Platform news, strategy analysis, and industry trends. Trusted by 40,000+ marketing professionals.
The reframe: Scale the prep, not the pitch
Every 1-1 play has two halves. The first is the prep: research, signal-gathering, enrichment, prioritization, prediction, and the first rough drafts. It’s heavy, slow, unglamorous work. It’s also exactly the kind of work machines are built for.
The second half is the relationship: developing a point of view, telling a story, understanding what this specific person actually cares about, creating a genuine human connection, and deciding to put your name on something and hit send. This is the work machines do poorly, and where buyers notice the absence of a human touch.
Our industry’s default instinct is to automate the second half and generate the message. That’s the wrong strategy.
Automate the first half so completely that your people have the time and context to be more human in the second half. Salesforce’s own data points to this: Sellers expect AI agents to cut prospect research time by 34% and email drafting time by 36%. Take the research win every single time.
Use AI to buy back hours. Then spend those hours on the relationship.
What to hand the machine
This is where AI earns its keep in a 1-1 motion. Point it at the prep:
- Research at scale: Deep research on an account or a person that used to take an analyst half a day now takes minutes. Distill information about company strategy, leadership priorities, recent moves, and your customers (the actual human across the table).
- Signal synthesis: Pull intent, news, hiring, funding, product usage, and earnings commentary into a single, readable account brief.
- Enrichment and data hygiene: Make sure the plumbing works properly. Clean up records, complete buying groups, and check for accurate contacts.
- Prioritization: Of all your accounts, which small set genuinely deserves 1-1 attention right now? Let the model rank propensity so your scarce human time lands where it counts.
- Prediction and next-best action: What’s the most relevant move for this account at this moment? Use AI to suggest. Use a human to decide.
A ZoomInfo survey of more than 1,000 GTM professionals found AI users saving an average of 12 hours a week. That’s the prize. Not “we sent more.” It’s “we gave our best people their time back.” But the whole strategy hinges on how you use those 12 hours.
What stays human
Hand the machine the prep. Keep these for yourself:
- The idea and the point of view: AI remixes what exists. It doesn’t have a take.
- The story: The thing that makes a buyer feel understood instead of processed.
- The judgment: What does this buyer actually care about, which is different from what your template assumes they care about?
- The relationship itself: You can’t outsource trust.
- The send: A human looks at the output and decides, “This is good enough to carry my name.”
Here’s a simple gut check for any AI-assisted touch before it goes out: If you’d be embarrassed for the buyer to know a bot wrote it, don’t send it. That one test will quietly kill most of the slop your team is about to produce.
What the saved hours are actually for
You’ve handed the prep to the machine and bought back the time. Now it’s time for the part most teams never get to, because they spend the savings sending more: Use those hours to learn something real about a specific human, and then do something real about it.
This is what I’ve started calling signal-based experiences, and it’s the sharpest version of 1-1 I know. The machine does the deep research, and a human reads it for what actually matters: the off-work signal.
The buyer who coaches a youth soccer team. The one who’s quietly obsessed with single-origin coffee. The one who ran their first trail marathon last fall. Build one genuinely thoughtful, real-world moment around that signal, then earn a conversation with it.
There’s a discipline to this, and it’s where most personalization fails:
- Routine, not one-off: You’re looking for something a person actually does repeatedly, not a hobby they mentioned once.
- Two sources, not one: If you can’t verify a signal in at least two independent places, you don’t act on it. Guessing on a signal is how you end up looking weird.
- A thoughtful friend, not surveillance: The bar is simple: It should feel like a friend who knows you planned it well. The moment the buyer feels like they’re being watched, you’ve lost.
Get it right, and the buyer’s reaction isn’t to delete. It’s “Wow, how did you know that about me?”
That question is the whole game. It’s the sound of a relationship starting and the exact thing no volume play can manufacture. It’s also not coincidentally what buyers tell Gartner they want: a human who actually understood them, not a machine that processed them.
How to make 1-1 ABM work
In one recent 1-1 signal-based experience, 30 carefully chosen contacts turned into 11 booked meetings and $1.3 million in pipeline in 40 days, off a 53% open rate, a 37% lead-to-meeting rate, and a 66% meeting-to-opportunity rate. Not from sending more, but from sending low volume to the right people, with something worth their attention.
1-1 is the bet I’m making at the twelfth agency for our enterprise clients. We let AI handle the prep and the plumbing, and protect human time for the relationship and the message. We scale the boring half so you can go deep on the half that builds trust.
The solution isn’t more automation, but better-aimed automation
AI isn’t the problem. How we’ve aimed it is.
We pointed the most powerful tool our profession has ever had at “do more,” when we should’ve pointed it at “matter more.” The future of go-to-market isn’t more automation. It’s better-aimed automation, using machines to buy back time and attention, than spending every minute of it being unmistakably human.
The buyer is practically begging for it. They’ve told us, with their silence and their delete key, that volume doesn’t move them, and relationships do. The pendulum is already swinging back toward relationship marketing and relationship selling.
The only real question is whether you’re early or late.