The Signal-to-Meeting Audit: Where Intent Data Fails Before a Reply

Your signal feed looks healthy. Accounts are scoring, the list refreshes every morning, and this month shows more high-intent accounts than last month. The calendar does not agree. Reps worked the list, sent the sequences, and booked roughly what they booked before anyone bought intent data.

The usual next move is to question the data. That is almost always the wrong first move. A buying signal has to survive five separate steps before it becomes a meeting, and only the first belongs to your provider. Run the audit below before you renew, switch, or cancel.

A signal-to-meeting audit is a step-by-step check of the path between a buying signal firing and a reply landing, testing at each stage whether the signal is still specific, still current, pointed at the right person, visible in the message, and measurable in the result.

Why Intent Data Fails Downstream More Often Than at the Source

Intent data usually fails after the data arrives, not before. Forrester's Q1 2023 Global B2B Intent Data Survey found that over 85% of companies using intent data had achieved business benefits, most commonly higher outbound response rates and more successful prospecting. In the same research, more companies expected benefits from intent data than achieved them, with the widest gaps on sales use cases and revenue growth, and analyst Brett Kahnke pointed to weak reporting on intent's impact as a common cause.

Read those findings together and the conclusion is uncomfortable but useful. The raw material works for most teams. The loss happens between the feed and the first reply, in steps you own. Each check below has a test you can run on data you already have, and each one fails differently.

Check 1: Does the Signal Name an Event, or Just a Topic?

A signal is only useful if a rep can say what changed at the account and when. "Elevated research on data infrastructure" is a category, not an event. A new VP of Data, a posting for three data engineers, or a migration mentioned on an earnings call are events, and they give a rep something concrete to open with. If your feed is mostly topic scores, add public, unstructured buying signals that describe events, and sort what you already collect by what each one proves.

The test. Pick ten accounts off this week's high-intent list and ask three reps to write the one-sentence version: what happened, at which company, and when. If they cannot, the signal will not produce a specific message no matter how well it scores.

The second test is the one most teams skip. Export this month's high-intent accounts and last month's, then measure the overlap. If the same accounts appear every month and the list broadly matches your firmographic ideal customer profile, your score is ranking fit, not timing. That is a familiar failure in new clothing, and it recreates the problem with generic prospect lists at a higher price. Timing signals should churn, because business events churn.

Check 2: Is the Signal Still True When the Message Goes Out?

Measure the median age of a signal at the moment of first touch. Most teams have never calculated this number, and it is usually worse than they guess, because signals sit in a queue while reps work yesterday's list.

A funding announcement referenced four days later reads as attentive. The same announcement referenced seven weeks later reads as a mail merge, and by then the reader has heard from thirty other vendors. Signals also age at different speeds, which is the argument for knowing how long a buying signal stays useful before building a cadence around it.

The test. Log the signal date and the send date for your last 200 touches, then look at the median gap by signal type. If it is measured in weeks, the problem is routing and capacity, not data quality.

Check 3: Does the Signal Point to a Person Whose Job It Changed?

Account-level intent tells you a building is interested. It does not tell you which desk to knock on. Forrester's survey respondents named identifying specific contacts inside accounts showing intent as the top execution challenge they face, noting that few providers deliver signals from known contacts.

This is where most programs quietly break. The signal is real, it is fresh, and it goes to whichever VP the database had an email address for. That person did not experience the event, cannot act on it, and does not reply.

The test. For twenty signaled accounts, write the name of one person whose responsibilities the event actually changed, plus the reason they would care this quarter. If your team cannot do that for most of the list, you have an account signal and a contact gap. Settle which contacts a signal should pull into the conversation before adding a new data source.

Check 4: Can the Buyer Check the Claim in Your First Line?

The message has to state something the reader can verify, then connect it to an outcome they own. Buyers have less patience for the alternative than they used to. A Gartner survey of 646 B2B buyers, run from August through September 2025, found that 67% of B2B buyers say they prefer a rep-free experience, and Gartner argues that what earns a seller a place in that journey is value clarity, a clear sense of how a solution improves outcomes in the buyer's specific context. Gartner also reports that confident buyers are twice as likely to report a high-quality deal as buyers with low decision confidence.

The test. Take your last fifty sent messages and delete the sentence containing the signal. If the message still reads fine, the signal did no work and you are paying for decoration.

Weak, because the signal is decoration: "I saw your company is growing quickly and thought I would reach out. We help teams like yours improve pipeline."

Stronger, because the claim is checkable and tied to a priority: "I noticed you posted three lifecycle marketing and revenue operations roles this month. That usually means retention and revenue process maturity are moving up the priority list, which tends to expose gaps in how expansion accounts get flagged. Worth a short conversation?"

The second version does not claim to know the company's strategy. It states what is observable, offers a reading of it, and leaves room to be wrong. That is the standard for turning a signal into an outreach sequence rather than decorating a template.

Check 5: Can You Tell Which Signals Actually Earn Replies?

Most teams cannot answer this, which is why the argument about intent data never resolves. If reply rate is tracked by campaign rather than by signal type, every signal looks equally mediocre and nothing gets cut.

The test. Tag every touch with the signal type that triggered it, then compare reply and meeting rates across types against a matched control group of fit-equivalent accounts with no signal. A month of this usually shows that two or three signal types carry the program and several contribute nothing.

Run it on your own accounts rather than importing a benchmark from a vendor blog. Reply rates by signal type vary with industry, offer and seniority, and a number from someone else's funnel is not evidence about yours. Once you know which signals earn replies you can weight them, which is the point of a system that scores accounts on intent and fit together.

When the Signal Was Right and You Were Still Late

Sometimes the audit clears every check and the meetings still do not come. The most common explanation is arrival time. Forrester has cautioned that late-stage signals may represent prospects who have already narrowed to a shortlist, which means the hottest-looking accounts in your feed can be the ones where the decision is effectively made.

The same analysis makes a related point: teams that use intent only to spot accounts in an active buying cycle leave value on the table, because earlier signals can shape a need before a shortlist exists.

So stop treating signal strength as the only ranking input. Earlier structural signals, a new executive, a new location, a hiring pattern, often give a longer runway than a late research spike. Combining signals into a real buying window beats chasing whatever burns hottest today.

How to Run the Audit in One Afternoon

Work through it in order and stop at the first check that fails, because fixing a later step will not help while an earlier one is broken.

• Pull this month's and last month's high-intent lists and measure the overlap.
• Ask three reps to write the one-sentence event summary for ten accounts.
• Calculate the median age of a signal at first touch.
• Name the affected person and their reason to care for twenty accounts.
• Delete the signal sentence from fifty sent messages and see what survives.
• Tag a month of touches by signal type and compare against a control group.

Most teams fail at step one or step four, and both are fixable without changing providers.

Where Alsona Fits in the Audit

Alsona is built for the first, third and fourth checks. Its signal library tracks 31 signals across hiring, advertising, technology, funding, SEC filings, reviews, social activity, events, podcasts and company news, and it interprets unstructured sources such as job postings and podcast conversations against your ideal customer profile, so what reaches a rep is an event with a date rather than a topic score.

Each account receives an intent-based lead score with the reasoning attached, which makes the overlap test in check one answerable instead of a matter of faith. Scored accounts flow into context-aware LinkedIn and email outreach that references the specific signal, and replies land in a unified inbox. You can also point it at accounts you already have, by CSV or CRM sync, and see which are producing signals now.

If you are choosing a source rather than fixing one, compare provider types before vendors, and if the category is still fuzzy, start with what intent data actually covers.

The Takeaway

Intent data rarely fails as data. It fails as a workflow, usually where an account-level signal has to become a specific person and a checkable sentence. Audit the five steps before you blame the feed, because the two most common failure points cost nothing to fix.

Run a podcast or webinar through the free podcast signal extractor to see what an event-level signal looks like next to a topic score. Build outbound around real buying intent, not guesswork.