Our intent data says surging but nobody answers. What is wrong?
Because a surge is not an event. Most intent products infer topic interest from content consumption across a publisher network and resolve it to a company, usually by network address. What arrives on your screen is a company name and an elevated topic score. It does not say who read anything, why they read it, whether they have budget, or whether anything at that company changed. So a rep opens a sequence with no fact to reference and no deadline to point at, and the conversation has nowhere to go.
The second reason is structural, and it is the one to check first. Resolution is company-level and noisy, so a surge from a five thousand person company can be produced by people who will never buy anything, in a department you do not sell to, at an office where somebody was researching for a class. Intent tells you the building is warm. It does not tell you which room, and outbound is addressed to rooms.
What a surge actually measures
It is worth reading your own vendor's methodology page, because the mechanics explain the symptom. The usual pipeline is: publishers in a co-operative share content consumption events, those events are mapped to topics, the consuming network address is resolved to a company, and the company's topic consumption is compared to its own historical baseline. A surge is a deviation from baseline, not a volume of interest.
That has two consequences people underestimate. First, a quiet company reading a little suddenly looks like a surge, while a large company that always reads a lot rarely does. Second, remote work has made address-to-company resolution significantly harder than it was when the technique was designed. Neither point makes intent worthless. Both make it a weak first input and a reasonable second one.
Six reasons nobody answers
Work down this table in order. Each row has a test you can run in a week with data you already have, so you can find out which failure you actually have rather than replacing the vendor and hoping.
| Row | Failure | What you would see | Test to run this week | What to do about it |
|---|---|---|---|---|
| 01 | No event behind the score | Reps cannot say why they are calling, so openers are generic | Take 20 surging accounts. For how many can you find a dated public event in five minutes? | Require a dated event before an account enters a sequence. Use the surge as a tiebreak |
| 02 | Wrong room in a right building | Replies from people who are not buyers, or silence at large accounts | Split reply rate by account headcount band | Stop working surges at large accounts without a named function or a second signal |
| 03 | Topic is too broad | The same accounts surge every month on a generic topic | Count accounts that surged in three consecutive months | Narrow topics to ones only a buyer would research. Drop perpetual surgers |
| 04 | Resolution noise | Surges from companies far outside your profile | Check what share of surging accounts pass your basic fit filter | Apply fit before timing, always. Fit filters, timing sorts |
| 05 | Late to the window | Contacts reply saying they already chose a vendor | Ask in every disqualification whether an evaluation was already underway | Move earlier by watching hiring and leadership evidence, which precede research |
| 06 | Right account, wrong ask | Opens and clicks but no replies | Compare reply rate on event-referencing touches versus surge-referencing touches | Reference the event, not the interest. Nobody wants to be told what they read |
What to do with intent instead
Do not throw it out. Demote it. Intent is a good tiebreak and a poor trigger. The workflow that works is: filter on fit, sort on dated events, and use the intent score only to order accounts that are already tied on the first two. Used that way it does real work, because it occasionally tells you which of two equally qualified accounts to call first.
- Never let a surge alone start a sequence. It has no event to reference and no owner to route to.
- Pair every surge with a public event before it reaches a rep. Surge plus a dated hiring cluster in the relevant function is a genuinely strong pair.
- Never mention the surge to the prospect. Reference the public event instead, which they can verify and did not need to be told about.
- Route by the function implied by the topic, not by whoever the data vendor supplies as a contact at that company.
- Track outcomes separately for surge-sourced and event-sourced accounts, or you will never know which one is carrying the number.
The measurement that settles the argument
- Tag the source of every worked account
Surge only, event only, or both. This is a single field and it is the whole experiment.
- Hold volume constant for a month
Same reps, same message discipline, same weekly account count. You are testing the input, not the effort.
- Compare reply and meeting rates by tag
Not open rates. Replies and meetings are the only outputs that survive contact with a forecast.
- Re-decide the spend
If surge-only performs at or below your cold baseline, the intent product is a tiebreak tool and should be priced and used as one.
This is what the category calls intent data, topic-level intent, and Company Surge.
What you are buying is usually called third-party intent data or topic-level intent, and one vendor's brand name for the surge concept has become the generic term for it. Related products include first-party intent, meaning activity on your own properties, and visitor de-anonymization, meaning resolving your own web traffic to companies.
The contrast worth holding onto is with buying signals or trigger events, which are dated, attributable public events at a named company. Intent estimates interest without an event. Signals record an event without estimating interest. Teams that argue about which is better are usually two teams with different jobs: intent is a marketing prioritization input, and signals are an outbound routing input.
Questions people ask next
Should we cancel the intent subscription?
Run the tagged comparison first. If surge-sourced accounts perform at or below your cold baseline over a month at constant volume, you are paying trigger prices for a tiebreak, and you should either renegotiate or use it only as a sort key.
Can I combine intent with buying signals?
That is the best use of it. An account with a dated public event and an elevated topic score is a stronger pair than either alone, because the event gives the rep something to say and the score suggests somebody there is already looking.
Why do the same accounts surge every month?
Usually the topic is too broad, or the company is large enough that baseline noise crosses the threshold regularly. Count your repeat surgers over three months and remove the perpetual ones, since they are measuring the company's size, not its interest.
Where to check this yourself
Primary records and published research, not vendor blog posts. Every link was checked on September 1, 2026.
- Bombora data and methodologybombora.com
A major intent vendor's own description of how consumption data is collected and turned into a company-level topic score.
- G2 Buyer Intent documentationdocumentation.g2.com
Vendor documentation for a first-party marketplace intent product, useful for seeing exactly what the signal does and does not contain.
- The Bridge Group, SDR Models and Metrics researchblog.bridgegroupinc.com
Published survey research on sales development structure and metrics, a sanity check for the reply-rate comparisons above.
- The Short Life of Online Sales Leads, Harvard Business Reviewhbr.org
Research on response speed to hand-raisers, the case where speed genuinely is the variable. An anonymous surge is not that case.
Intakra works the other way around: it starts from a dated public event, keeps it only when it can name what breaks and who absorbs it, and gives you the source link so the reason survives being questioned.
Free scan, no signup, no card