What Is G2 Buyer Intent? A Practical Read on What It Catches and What It Can't
Nine page-view triggers, one account-level blind spot, and the free review data most teams skim past. A practical read on both halves of G2 buyer intent.
Two different things go by the name "G2 buyer intent," and the difference between them decides what you can put in the first sentence of an email.
What Is G2 Buyer Intent?
G2 Buyer Intent is a paid signal that reports which companies are researching your product on G2.com. It fires when someone at that company views pages connected to you, resolves the visit to an organization, and scores how actively that organization is researching. It measures attention on the platform, not what a buyer said.
G2's own documentation lists nine signal types that trigger it: profile, pricing, alternatives, category, compare, sponsored content, licensed content, reference page, and competitive. That last one is the interesting member of the set, because it fires on activity around your competitors rather than around you.
Each signal arrives with a company attached. You get headquarters location, company size, industry, the date of last activity, pageview counts broken out by signal type, which competitors the account compared you against, and two scores G2 calls Buying Stage and Activity Level. G2 describes the product as showing you "which companies are researching your category, your profile, and comparing you with your competitors".
The scale claim behind it is that G2 can see which of the 100M buyers on G2.com are actively researching your product, category or competitors. That description is accurate, and it's worth reading closely. Every noun in it is a company.
What Do G2 Buyer Intent Signals Actually Tell You?
Two useful things: presence and timing. Someone at a specific account is in your category right now, and you didn't have to guess.
That's real, and worth saying plainly before picking at it. A rep working a territory of four hundred accounts has no honest way to rank them without a signal like this, and an account that just spent twenty minutes on a comparison page involving your product is a better use of Tuesday than one that didn't. Independent measurement backs the effect up. Dreamdata published benchmark work on how G2 intent shows up in B2B buyer journeys, which is the kind of third-party read worth having before you buy any signal product.
Where activity data sits in the wider picture is covered in our practical guide to buyer intent data. For the stage-by-stage version, the buyer intent funnel breaks down what each phase of research implies.
So the signal is good at what it does. The question is what it was never built to do.
What Are the Limits of G2 Buyer Intent Data?
Three limits. None of them are flaws, exactly. They're consequences of how the signal gets collected.
It reports an account, not a person. G2's documentation is explicit that it doesn't collect, process, or store personally identifiable data for Stack, and that collection happens at the organization level. You get a count of unique visitors and their geolocation. You don't get a name. A rep holding an account-level ping still has to work out whose inbox to open, and that decision is where most of the week goes.
It carries no reason. A pageview records that attention happened. It doesn't record what the person was trying to solve, what broke last quarter, or which part of their current setup they've quietly stopped defending in internal meetings. Volume is not motive. Two accounts with identical Activity Level scores can be in completely different conversations, and the score can't separate them.
It fires late. This is the one that matters most, and it isn't specific to G2. Research from 6sense found that 95% of purchases come from the day-one shortlist, the set of vendors a buyer already had in mind before formal research began. TrustRadius's 2026 buying report found that 83% of buyers shortlist three or fewer vendors. By the time someone is comparing options on a category page, the list that decides the deal has usually already been drawn.
Activity data is strongest at confirming a race you're already in. It's weakest at telling you a race is about to start.
That gap is where competitive pipeline quietly goes missing, and it's the same gap we mapped in the dark funnel: the research that happens where no pixel reaches.
What Is Review-Derived Buyer Intent?
Review-derived buyer intent is the signal carried by the review text itself rather than by activity around it. A review is written, attributed to a named person at a named company, timestamped, and specific about a problem. It's the rare artifact in go-to-market that counts as evidence rather than inference.
It also sits on the same website as the activity signal, in public, free to read.
We analyzed 29,677 G2 reviews as of August 2026, and two numbers from that set reframe how most teams read a review platform.
The first: 88.6% of those G2 reviews are rated 4.0 or higher. A buyer who filters for four stars and up excludes almost nothing, and a vendor watching for rating drops is watching a metric with very little room to move. The score isn't where the discrimination lives.
The second one changes the workflow: 41.8% of five-star G2 reviews contain a pain point. Not four-star reviews. Five-star. Across the 14,067 five-star G2 reviews we hold, roughly two in five carry a specific, articulated complaint sitting inside what reads as an endorsement.
We first published that pattern when we looked at how B2B software buyers actually use reviews, and it has held as the corpus has grown. It matters here because of what it does to a monitoring habit.
Most teams read the one-star pile, because that's what "monitoring reviews" has always meant. The one-star pile is also the smallest, least representative part of the corpus, and it's the part your competitor's customer success team already found and already called about. The frustration nobody is acting on is filed under a recommendation.
Buyers already read this way. TrustRadius found that on a review site, review content at 18% and reviewer relatability at 17% both outrank scores and badges as decision inputs. G2's own 2024 Buyer Behavior Report put review platforms as the number one information source for 31% of B2B buyers, up from 23% the year before. Buyers weigh the text. Vendors weigh the star.
We made the category argument for this signal in the 5th type of B2B intent data, where the case was that review-derived evidence doesn't belong filed as a subtype of second-party data. This is that argument with one platform named and the numbers attached. For the wider benchmarks on how buyers behave, the state of B2B software buying collects the sourced figures in one place.
What a Review-Derived Buying Signal Looks Like in Practice
Picture a four-star review posted last Tuesday. An operations lead at a 300-person logistics company praises the reporting, calls the onboarding good, then spends two sentences on approval workflows that break every time their headcount changes. She signs it with her name and title.
Count what that one artifact carries. The person. Their role. Their company and its size. The specific thing that's broken. The week it broke in. A pageview carries none of those, and no amount of scoring turns one into the other.
Right now most teams find that review by accident, weeks later, while researching something else. The reason isn't laziness. Reading every new review across five platforms for a set of competitors is a full-time job that produces nothing most days, and nobody's going to remember to check another dashboard between 47 other tabs.
Software is better at that kind of watching than people are. Monitoring competitor products and pushing the reviews that carry a real pain point into the tools a rep already has open is what Opportunity Alerts do, with the reviewer's company and contact context attached so the research is finished before the rep starts writing. It's one way to act on this signal, not the only one. The review-signal lead generation workflow lays out the manual version if you'd rather build it yourself.
The outcome either way is the same. You get to open with the thing she actually said, which means you're not writing a cold email. You're answering one.
Using Both Signals Together
They answer different questions, so they compose rather than compete.
Activity data answers which accounts, and when. It's the right input for ranking a queue, spotting a competitor comparison while it's happening, and catching a current customer wandering through alternatives pages before the renewal call. G2 leans into that last case explicitly, positioning its signals as a way to spot churn risk as well as new pipeline.
Review text answers who, and what to say. It's the right input for the message itself, and for finding the specific human whose problem you can speak to.
The strongest pairing available on a review platform is the overlap: an account showing category activity where a named person has also written about a specific, unresolved problem. One tells you the timing is live. The other tells you what to lead with.
A routing rule you can adopt this week. Let activity data set the order of your account list. Let review text write the first two sentences of every touch. Treat any account that shows up in both within the same fortnight as your highest-priority outreach of the week. Our five plays for turning intent signals into pipeline goes deeper on sequencing, and intent data vs contact data is worth a read if your stack currently confuses the two.
Is G2 Buyer Intent Worth Paying For?
G2 doesn't publish Buyer Intent pricing, so anyone quoting you a number online is guessing. Three conditions decide whether it earns its cost, and you can answer all three yourself.
Does your category carry enough traffic on the platform? Activity signals need volume to fire. A crowded category with heavy comparison traffic produces a usable stream. A young or narrow one produces a trickle that's hard to act on and easy to over-read.
Can your team act on an account-level ping within days? The signal decays. If an alert sits in a queue for two weeks before anyone works it, you're paying for information you aren't using, and the problem is the process rather than the data.
Do you already have a person-level path into the account? An organization name is a starting point that still needs a human attached. If finding that human is your bottleneck, solve that first, because an activity signal will just keep handing you the same bottleneck faster.
Answer those honestly and the buying decision usually makes itself. The broader method for working competitor frustration as a channel, activity data or not, is in our guide to Frustration-Led Growth, and our pricing page shows what the review-signal side of this costs to run.
The Signal Everyone Already Owns
Both signals live on the same website. One is metered, account-shaped, and tells you the room is full. The other is public, person-shaped, and tells you what the people in it are annoyed about.
Teams buy the first and skim the second, which is an expensive way to learn less. The reason a buyer is looking is usually already published, in a review, under a four or five star rating, written by someone who signed their name to it.
Somebody typed it. Nobody read it.
Start with the text, use the activity data to time it, and your first sentence stops being a guess.
Start monitoring for free and see which of your competitors' customers have already written down what's going wrong. 30-day free trial, credit card required.
Methodology: review figures come from Reechee's review database as of August 2026, covering 29,677 G2 reviews. Pain points are detected and categorized per review by the same analysis pipeline that powers Opportunity Alerts.
