Intent data is worth it for a small team only when three things are already true: you have a defined ICP and a verified list, someone owns the signals every week, and one extra deal pays for a year of the tool. If any of those is missing, start with the free tiers or skip it.
The price spread is wide. As of September 2026, Apollo includes 1 intent topic on its free plan and 6 on its $49 per seat per month plan. At the other end, Vendr's anonymized purchase data puts the median buyer at $25,000 a year for Bombora, $62,440 for 6sense and $68,360 for Demandbase.
For a founder-led team with nobody working full-time on sales operations, the cheap tier is enough to learn whether signals change anything. The enterprise platforms are built for teams with someone whose job is running them.
What intent data is, in plain words
Intent data is a record of research behavior that suggests a company is looking at a problem you solve. It comes in three kinds.
- First-party: activity on your own properties. Who visited your pricing page, who opened your emails, who replied and then went quiet. You already own it and it costs nothing.
- Second-party: another platform's first-party data, sold to you. G2 Buyer Intent is the clearest example: it flags companies viewing your G2 profile, your pricing page there, comparison and alternatives pages, and your category (G2 documentation, checked September 2026).
- Third-party: research activity collected across many sites you do not own. Bombora's Company Surge flags a company when its reading on a topic climbs above that company's own usual level, based on data from a co-op of more than 4,000 B2B publishers (Bombora, checked September 2026).
Most intent data identifies companies, not people. G2's documentation says its buying stage and activity level are calculated for each company. That matters for how you use it, which I come back to below.
Why timing is the whole pitch
The argument for intent data is that buyers decide early. The most-quoted recent numbers come from 6sense, which sells intent data, so read them as a vendor's own survey.
Its 2025 Buyer Experience Report surveyed more than 4,000 buyers across North America, EMEA and APAC. In that survey, 94% of buying groups already had a ranked shortlist before they spoke to any seller, and they bought from the vendor at the top of it 77% of the time (6sense, November 2025).
If that holds for your market, reaching a company while it is still researching is worth something. The open question for a small team is whether the signal is accurate enough, and cheap enough, to act on.
What intent data costs in 2026, tier by tier
All prices checked on 2026-09-28. Public list prices come from the vendors' own pricing pages. Enterprise figures come from Vendr, a software buying platform that publishes medians from anonymized purchases, so they reflect what buyers actually paid.
| Option | Intent signal you get | Price | Source | Small-team fit |
|---|---|---|---|---|
| Your own site, inbox and CRM | Pricing-page visits, replies, clicks, past deals | $0 | Your own data | Always start here |
| Apollo Free | 1 intent topic plus intent filters; identifies 20 website-visitor companies a month | $0 | Apollo pricing | Good for a first look |
| Apollo Basic | 6 intent topics plus intent filters; 100 visitor companies a month | $49 per seat per month, billed annually | Apollo pricing | Good for a 60-day test |
| Apollo Organization | 12 intent topics plus intent filters | $119 per seat per month, billed annually, minimum 3 seats | Apollo pricing | Only if you use the rest of the platform |
| Leadfeeder Lite | Website visitor companies, 7 days of history, a cap of 100 identified companies per month | EUR 0 | Leadfeeder pricing | Good for first-party signals |
| Leadfeeder Activate | Adds intent filtering and verified contacts | From EUR 369 a month billed annually, EUR 527 monthly | Leadfeeder pricing | Only with steady site traffic |
| G2 Buyer Intent | Companies researching your G2 profile, category and comparisons | Not public; only some G2 plans include it | G2 documentation | Only if buyers in your category research on G2 |
| Bombora | Third-party topic surges across its publisher co-op | Median $25,000 a year, from 35 purchases | Vendr, updated February 2026 | Rarely |
| 6sense | Third-party intent plus predictive account scoring | Median $62,440 a year, from 385 purchases | Vendr, updated February 2026 | Rarely |
| Demandbase | Account-based platform with intent | Median $68,360 a year, from 186 purchases; range $24,000 to $164,151 | Vendr, updated February 2026 | Rarely |
Two things stand out. The three platform medians are 21 to 58 times what two Apollo Basic seats cost in a year (the math is below). And the cheap tiers mostly give you website-visitor identification plus a small number of topics, which is a narrower thing than the co-op data the platforms sell.
What that means for a founder-led team
Run the arithmetic against your own deal size before any demo.
At the Bombora median of $25,000 a year, a company whose average deal brings $12,000 in gross profit needs about 2 extra closed deals a year from the signals just to break even ($25,000 / $12,000 = 2.1). Those deals have to be ones you would not have won anyway, and proving that takes a clean comparison.
Two Apollo Basic seats cost $49 x 2 x 12 = $1,176 a year. At that price, one extra deal pays for the tool many times over, and the test is cheap enough to run without a business case.
That is why the answer changes with team size. A team of two can afford to be wrong about a $1,176 experiment. It cannot afford to be wrong about a $62,440 contract that also needs someone to configure topics, route alerts and chase them every week.
A quick way to read your own case:
| Your situation | What to do |
|---|---|
| No defined ICP or verified list yet | Skip intent data and build the list first |
| Total market of a few hundred accounts | Contact all of them; use free signals only to set the order |
| Nobody can give the signals a fixed hour every week | Stay on free tiers, or skip |
| One extra deal would not cover a platform's yearly price | Entry tier only |
| Large market, someone owns the signals weekly, and the 60-day entry test showed a clear gain | Only then get quotes for a platform |
A note from how I run outbound
Smirnov Consulting Group is a Prague-based B2B outbound lead generation agency that runs cold email and LinkedIn campaigns for founder-led B2B companies and books qualified sales calls.
Running those campaigns, I treat intent data as a timing tool, and timing comes after the foundations. The offer has to be clear, the sender's profile has to survive a prospect checking it, and the list has to be verified and on-ICP.
Once those are in place, volume does most of the work, because in any given quarter most prospects are not shopping, and the only way to find the few who are is to reach enough of them. Signals can tell you who to contact first. They cannot replace the reach.
The numbers in this market you should doubt
Intent data marketing leans on a few statistics that get copied from article to article. Three I would not repeat as fact:
- "97% of companies report positive ROI from intent data." It is usually credited to a Demand Gen Report study, but the report page it is attributed to no longer exists (checked September 2026). The study may well have existed, but nobody can check it at the source today.
- Gartner's buying-cycle percentages (how little of the cycle buyers spend with suppliers). The page they are cited from sits behind a lead-capture form, so the exact figures cannot be confirmed without registering.
- Conversion and deal-size lifts from vendors' own pages, such as one intent vendor's claim that signal-qualified leads convert 47% better and close 43% larger deals, with no method or sample given.
A useful rule for any vendor pitch: ask for the sample size, the period, and the comparison group. A lift figure without a control group only tells you that the vendor's best customers did well.
How to test intent data before you pay for a platform
A test tells you more than any demo. Here is a version a founder plus one or two helpers can run in about 60 days with free or entry-tier tools.
- Write down your ICP and build the verified list first. Intent data should rank accounts inside your list, never replace it. Start with the checks that take a lead list apart: cleaned titles, a set headcount range, verified addresses.
- Pick 3 to 6 narrow topics. Choose topics that describe your buyer's problem, not your category in general. Broad topics flag a large share of any list and tell you little.
- Split the list in two. Accounts that show a signal in the test window go in one group. A matched set without a signal, same titles and company sizes, goes in the other.
- Send both groups the same sequence from the same sender, in the same weeks. Change nothing else.
- Compare replies, positive replies and booked calls, not opens. With small lists, expect noise: the guide on split-testing cold email at low volume shows how many sends a real difference needs.
- Decide on a rule you set in advance. For example: if flagged accounts do not book clearly more calls per 100 contacted than the control, you stay on the free tier or drop it.
Give the test at least one full cycle of your sequence, including every follow-up. My team usually sees the reply arrive on the third follow-up, so a test that stops after the first email judges the wrong thing.
Mistakes small teams make with intent data
Buying the platform before the process. A dashboard of surging accounts is useless if nobody owns it. If you cannot name the person and the weekly hour it will take, wait.
Treating a company signal as a person's interest. Most intent data is company-level. Someone at the company read about your topic, which does not mean the VP you are emailing did.
Mentioning the signal in the email. "I noticed your team has been researching CRM migration" reads as surveillance, and the reader may not even be the one who researched it. Use the signal to pick who gets contacted first, then write about their problem in plain terms.
Choosing topics too wide. A topic as wide as "sales" or "marketing software" will flag a big part of any list. Narrow topics give fewer alerts, and the alerts mean more.
Using intent instead of a list. When your total market is a few hundred companies, you can reach all of them anyway, and timing adds less than you think. The choice between working a fixed set of named accounts or a wide list matters more.
Judging it after two weeks. One sequence cycle plus follow-ups is the minimum. Anything shorter measures chance.
If you sell into the EU, ask any visitor-identification vendor how it identifies companies and what data it stores, and check the answer with whoever handles GDPR for you. This is not legal advice.
Questions about intent data for small teams
How much does intent data cost for a small team?
From nothing to a few hundred euros a month for entry tools. Apollo's free plan includes 1 intent topic and its $49 seat includes 6. Leadfeeder's free tier identifies up to 100 visiting companies a month. Enterprise platforms sit at medians of $25,000 to $68,360 a year in Vendr's data.
What is the difference between website visitor tracking and intent data?
Visitor tracking is first-party: it names companies that came to your own site. Third-party intent data watches research on sites you do not own, across a publisher network. Visitor data is cheaper and shows interest in you specifically, but only covers companies that already found you.
Can one person run intent data part-time?
On an entry tier, yes, if it is a fixed weekly slot: review the flagged accounts, move the relevant ones up the sending order, ignore the rest. An enterprise platform with dozens of topics and routing rules is a bigger job than a part-time slot allows.
Should I use intent data or just send to my whole list?
If your list is small enough to contact every account within a quarter, send to all of it and use signals only to set the order. Intent data earns its cost when the market is too large to reach fully and you need a reason to pick who goes first.