You burn your TAM when dial volume grows faster than the reasons to call. The fix is not to dial less. It is to measure coverage, attempts, negative dispositions, number reputation and signal share every week, and to throttle volume on the segments where those numbers go red.

Here is how we think about it after analysing 1,000,000 B2B cold calls in France, and the thresholds we use at Sonik.

What "burning your TAM" actually means

A total addressable market is finite. In French B2B SaaS, a realistic target segment is often 3,000 to 15,000 callable contacts, not millions. A team of four SDRs on a parallel dialer can touch 2,000 contacts a week. Without a plan, they cycle the whole segment every month, and by the third cycle every connect starts with "you called me already".

Burn shows up in three ways:

• Fatigue. Connect rates on the segment drop because prospects recognise the number or the company and stop answering.
• Reputation. Carriers and call-blocking apps flag your numbers after too many short, unanswered calls. Every number you own gets worse.
• Brand. Prospects who said no twice tell colleagues. In a market like France, where decision-makers in a sector know each other, that spreads.

The paradox: a dialer that gives you 5x more conversations also gives you 5x faster burn if the only thing that changed is speed.

The 5 metrics we track every week

1. Segment coverage

Contacts touched by phone this week, divided by callable contacts in the segment. Above 20% a week with a static list means you will have cycled the segment in five weeks. Either widen the segment, slow the cadence, or stop and rebuild the list from signals.

2. Attempts per contact

In our experience, most connects happen within the first three attempts. Attempt six on a contact who never picked up is almost pure burn. We cap at five attempts over three weeks, then park the contact for 60 days or move it to email.

3. Negative disposition rate

Share of connects ending in "not interested", "not the right person" or "do not call again". A rising rate on a segment is the clearest early warning. If negatives cross 40% of connects on a list, the list is wrong, not the reps.

4. Connect rate per number

Track connect rate for each outbound caller ID, week on week. A number whose connect rate drops by a third against the team average is probably flagged. Retire it, rotate in a fresh one, and keep volume per number moderate. In our experience, flagged numbers rarely recover quickly.

5. Signal share

Share of calls with a CRM trigger behind them: website visit, job change, closed-lost, dormant lead, social intent. This is the metric that makes the other four better. Trigger-based openers reach a 7 to 10% meeting rate against 2 to 3% for pitch-first openers in our benchmark. Every point of signal share you add is a point of volume you do not need.

Throttling: where to slow down and where to push

Volume should be a dial you turn per segment, not a team-wide setting.

• Push on segments with high signal share and low coverage. Dormant leads are a good example: a "How have you been?" opener to a known contact reaches around 10% success against 1.5% for a standard cold open.
• Push in the slots that convert. Across personas, 4pm gives a 15.8% connect rate and 10am 14%, against 9.8% at 8am. Same list, same reps, 60% more conversations per dial by moving the block.
• Throttle on segments where negatives are rising or coverage is already above 20% a week.
• Stop on contacts past five attempts. Let the sequencer take them.

The line count on the dialer follows the same logic: 4 to 5 lines on fresh cold lists, 1 to 2 on signal lists where every pickup deserves a prepared opener.

Rebuilding the list before it runs dry

Reps use about 20% of the signals already in their CRM. The other 80% is the cheapest TAM extension there is, because it is the same TAM called for a better reason. Half of Sonik's own pipeline comes from signals that were already in the CRM when we looked.

A monthly list rebuild we recommend:

1. Export closed-lost deals older than nine months. The contact, the pain and the budget cycle are known.
2. Pull champions who changed company in the last 90 days. They know your product and need to look good in a new job.
3. Match website visitors to target accounts and add the decision-maker, not just the visitor.
4. Add inbound leads that went quiet after a demo.
5. Only then top up with a fresh persona list, sized so weekly coverage stays under 20%.

How Sonik approaches TAM burn

Sonik's dialer logs every disposition automatically, so coverage, attempts and negative rates are computed rather than estimated. Teams set the line count and the attempt cap per list, and connect rate is reported per outbound number so a flagged caller ID is visible within days.

Signals from HubSpot or Salesforce (website visitors, job changes, closed-lost, dormant leads, social intent) feed the call queue directly, which keeps signal share high without manual list building. European numbers are provisioned and rotated inside the platform.

Our north-star funnel is 20 conversations per SDR per day. That target only works if the conversations are with the right people, which is why the mission reads 2x more meetings through 5x more conversations with the right contacts, not just more calls. See sonik.ai/signals and sonik.ai/dialer.

Conclusion

Dial volume is safe as long as someone is watching coverage, attempts, negatives, number reputation and signal share. Set thresholds, review them on Monday, and throttle per segment rather than per team. This week, calculate weekly coverage for your main segment: contacts called divided by contacts callable. If it is above 20%, your next sprint is a list rebuild from CRM signals, not a bigger call block.