Intent data providers compared: where each one's signals come from

This page is for sales and marketing leaders deciding whether to buy intent data and from whom. It assumes you know what intent data is; if not, start with the intent data definition, which covers the types, how a surge score works and the privacy rules. Here the focus is the providers: what each one watches, how it sells, and how to test whether its signals make your team more productive.

How we compared

Blackpearl Group wrote this page, and one of the options is its own (marked "we make it"). Every fact about a provider comes from its own website, pricing page or documentation, read on 26 October 2026 and listed in the sources. Where a provider publishes no price, we say "quote only". Signal volumes, topic counts and coverage figures are each vendor's own claims. We haven't measured any provider's accuracy, so there are no scores; the test below is how to measure it on your accounts. We left out providers whose product pages we couldn't read on the day.

Four kinds of intent data provider

Most teams get more from combining two sources (one that sees research about the problem, and one that sees interest in you) than from buying the biggest single feed.

Intent data providers at a glance

Checked on each provider's own pages, 26 October 2026.

How a third-party surge is measured

Bombora publishes its method, so it's a useful reference for what "surging" means. It says it monitors company research against 25,300+ topics and evaluates each account-topic pair by comparing "the most recent three weeks of activity against a 12-week historical baseline". A surge is a change against that company's own normal, not a count of page views. Two things follow:

• Big companies don't surge just for being big. A company that always reads a lot about freight software won't flag unless its research rises above its own baseline.
• Topic choice decides what you see. Bombora suggests choosing topics in groups: your brand and products, and the pain points of your buyers. Too few topics and you miss buyers; too many broad ones and every large company surges on something.

When you evaluate any provider, ask the same question: surge against what baseline, over what window, refreshed how often? The intent data definition works through a surge score example.

Worked example: does intent pay for a six-rep team?

Illustrative example. The company, account counts, meeting rates and deal values are invented to show the arithmetic. They are not benchmarks.

The team: Routekeel (routekeel.example) sells a transportation management system to mid-sized shippers. Six reps each work 100 new accounts a month, 600 in total, from a list of 3,000 companies that fit its ideal customer profile. Today reps pick accounts by size and region.

The intent setup: Routekeel chooses 10 topics (its product category, two competitors, and pains such as freight cost and carrier management). In a typical month the provider flags 300 of the 3,000 accounts as surging on at least one topic.

Without intent: 600 accounts worked, an illustrative 2% book a meeting: 600 × 0.02 = 12 meetings.

With intent: reps work all 300 flagged accounts plus 300 others. If flagged accounts book at 4% (an illustrative 2x lift) and the rest at 2%: (300 × 0.04) + (300 × 0.02) = 12 + 6 = 18 meetings. That's 6 more a month.

What 6 meetings are worth: if 20% of meetings become opportunities, 25% of those close, and a deal is worth $30,000 a year, each meeting is worth 0.20 × 0.25 × $30,000 = $1,500 in expected first-year revenue. Six extra meetings = $9,000 a month, or $108,000 a year.

The catch: all of that rests on the 2x lift. At 1.25x (flagged accounts book at 2.5%), the gain is 300 × 0.005 = 1.5 meetings, worth $2,250 a month or $27,000 a year. At no lift, it's zero. The price of intent data that's worth buying depends entirely on a number only your own test can produce:

What Routekeel should do: start with intent it can buy without a new contract (Bombora topics in Apollo, or its own website visitors), measure the lift for one quarter, and only then ask enterprise providers for quotes, with its own break-even number in hand.

Which provider fits

Run a two-week signal check, then measure lift

Two weeks is enough to check whether a provider's signals are sensible. Meeting lift takes longer: plan on 60 to 90 days.

1. Pick 10 topics in two groups, your category and your buyers' pains, and write them down so they don't drift during the test.
2. Take 50 flagged accounts and 50 unflagged ones from your ICP list, chosen at random.
3. Days 1 to 5: check fit. Mark each account +1 if it fits your ICP and you can find a visible reason it might be buying now (a job post, news, a technology change), 0 if it fits with no visible reason, and -1 if it doesn't fit or no longer exists. GTM-Bench, Blackpearl Group's June 2026 benchmark of AI prospecting systems, scores weak records -1 the same way so volume can't hide noise.
4. Days 6 to 10: start outreach to both groups with the same sequence, the same reps and the same week.
5. Days 11 to 14: compare early signs: replies, positive replies and meetings booked so far.
6. Days 15 to 90: keep counting. Compare meeting rates between the groups. Lift = flagged meeting rate ÷ unflagged meeting rate.
7. Decide with the table above. Put your own lift, deal size and conversion rates into the worked example and compare the expected value with the quote.

Provider:                     Topics (list all 10):
Dates:                        ICP list size: ___

SIGNAL CHECK (days 1 to 5)       Flagged (50)   Unflagged (50)
Fit ICP with a visible reason (+1)   ___             ___
Fit, no visible reason (0)           ___             ___
Doesn't fit or gone (-1)             ___             ___

OUTREACH (same sequence, same reps)
Contacted:                        ___             ___
Replies / positive replies:       ___ / ___       ___ / ___
Meetings at day 14:               ___             ___
Meetings at day 90:               ___             ___

Lift = flagged rate ÷ unflagged rate = ___
Expected value a year (from the worked example): $___
Annual cost of the provider: $___
Decision and why:

What intent data can't tell you

• Who to call. Most third-party intent is at the company level. You still need the right person, and the reason they'd care.
• Why they're researching. A surge on "transportation management" could be a buyer, a student project or a competitor's analyst team. Check fit before acting.
• Anything about companies outside the source. A co-op only sees member sites; a review site only sees software buyers; your website only sees people who already found you.
• How accurate it is. Topic counts and coverage figures are the vendors' own. Your lift test is the figure that applies to your market.

Next step: before you talk to any provider, read the vendor questions in the intent data definition, then see how intent fits with the rest of your stack in lead generation software compared.