How to measure meetings from signal-matched sends

Sends are not outcomes.

Teams install RB2B, Vector, G2 alerts, Clay waterfalls, then celebrate activity. Opens. Clicks. "Personalized" volume. The calendar stays empty and leadership asks for more tools.

Meetings per signal source is the scoreboard. Open rate without meetings is theater with better charts.

What should I measure instead of vanity?

Three numbers. Weekly.

1. Qualified sends: fit + intent + recency cleared, human-approved.
2. Meetings booked from those sends (attribution to the signal that triggered the row).
3. Meetings per 100 qualified sends by signal source.

Secondary: show rate, opportunity rate, closed-won. Primary for Signals ops is still meetings. Meetings, not theater.

Ehrenberg-Bass / LinkedIn B2B Institute: ~5% in-market. Your chart should show whether each signal finds that 5% or just annoys the 95%.

How do I attribute a meeting to a signal?

Tag the row at ingress.

Every Clay or CRM record that enters the send queue carries signal_source, signal_at, and fit_score. The meeting object inherits those fields. No tag, no credit.

If two signals fire, keep primary = the one that justified the send. Secondary can be noted. Do not double-count to make the dashboard pretty. That is how vanity metrics sneak back in.

What does a healthy week look like?

Thin volume. Clear tags. Sources ranked.

Example shape (illustrative, not a promise): de-anon pricing hits → 12 meetings / 100 sends. Competitor LinkedIn likes alone → 1 / 100. Firmographic-only control → near Gong's cold floor (~3.4% reply, rare meetings).

Kill or demote the bottom source. Double down on the top. This is Grow discipline applied to Signals spend.

Salesforce: ~27% of reps hit quota. Measuring meetings by source is how you stop blaming "the market" when the source mix is junk.

How does this change tooling spend?

Pay for sensors that produce meetings after fit filters.

If Factors.ai paths book and raw scraped lists do not, funding follows the path. If PostHog stuck states book and newsletter clicks do not, fix content incentives.

CAC gets honest when signal cost rolls into cost-per-meeting and cost-per-opportunity, not cost-per-send.

What are the failure modes?

• Measuring replies instead of meetings (polite declines inflate ego)
• Letting SDRs pick attribution after the fact
• Mixing in-market and nurture sends in one denominator
• Celebrating AI draft volume with no human kill rate

Forrester's ~84-day cycle and Gartner's 30%+ ghost rate punish fake pipeline. Bad measurement creates fake pipeline on purpose.

Would you rather 2,000 tagged sends and a ranked source chart, or 2,000 untagged sends and a story about how "outbound is hard"?

Adapt or fail. Tag the signal. Count the meetings. Cut the dead sources.

Start Signals, Convert, Grow

What about multi-touch attribution?

Still pick a primary signal for the send that booked the meeting.

You can report assist signals separately. Do not let "multi-touch" become an excuse to credit every tool in the stack for every meeting. That is how every vendor looks ROI-positive and nothing gets cut.

Grow budgets should follow primary-source meetings, not assist theater.

FAQ

What if marketing and sales disagree on attribution?

Primary = signal that authorized the send. Write it down. Debate once. Stop renegotiating every Friday.

Should I track pipeline dollars by signal too?

Yes after meetings are trustworthy. Dollars without meeting hygiene just move the vanity upstream.

How long before I kill a signal source?

Enough qualified sends for a real rate, often two weeks at low volume. Do not wait a quarter out of hope.

Do opens matter at all?

As a deliverability health check. Not as a success metric.

How does this fit signal-based outbound?

Signal-based outbound is the motion. This post is the scoreboard for that motion.