A good outbound cadence for European B2B is 10 to 12 touches over 15 working days, call-first, with the phone carrying six of those touches at the persona's best slot, email and LinkedIn setting up the next call rather than replacing it, and a rule that any live signal interrupts the sequence immediately. Below is the cadence we run, why the phone goes first, and the European rhythm most templates ignore.

Call-first or email-first: what the connect data says

The argument for email-first is that it warms the call. The argument for call-first is arithmetic.

Across our benchmark of 1,000,000 B2B cold calls in France, the average connect rate is 11.3%. More usefully, in our experience most connects happen within the first three attempts. Three calls in the first five days reach most of the people you will ever reach. Three emails in the first five days reach an inbox.

In our experience the sequences that start with a call book more meetings per contact, for a simple reason: the first call either produces a conversation, or it produces a reason for the first email ("I tried to reach you this morning about X"). Email-first produces neither until someone replies, and most do not.

The exception is a known contact. A dormant lead or a past champion gets a short email first, then a call opening with "How have you been?", which reaches about 10% success in our data against about 1.5% for a standard cold open.

The 15-day cadence, touch by touch

Eleven touches, six by phone. Each call sits in the slot where the persona picks up: founders around 10am, marketing around 4pm, HR around 1pm. Across personas, 4pm connects at 15.8% and 8am at 9.8%, so never put the first call at 8am.

1. Day 1, call at the persona's slot, trigger-based opener. No voicemail yet.
2. Day 1, email one hour later. Three lines: the trigger, one question, a mention that you called.
3. Day 2, LinkedIn connection request, no note or a one-line note naming the trigger.
4. Day 3, call, different slot than day 1. Leave a 12-second voicemail naming the trigger.
5. Day 5, call, back to the best slot. This is the third attempt; most of the connects you will get have now happened.
6. Day 5, email reply to the first thread, one new piece of context (a peer example, a number).
7. Day 8, LinkedIn comment or message on something they posted. Skip if they have posted nothing.
8. Day 9, call, mid-morning or 4pm.
9. Day 11, email, short, direct ask for 15 minutes with two time options.
10. Day 12, call, voicemail that references the email.
11. Day 15, call, the last one. Then park the contact with a return date, not a "closed" status.

Each touch names the previous one. A cadence where every email could be the first email is a mailing list.

The European rhythm: when not to send or call

Most cadence templates were built for a market that works through lunch and takes two weeks off in summer. European B2B does not.

• Lunch. In France the 12pm to 2pm window is dead for most personas except HR, who answer around 1pm in our data.
• Wednesday afternoon. Many managers with school-age children are unreachable. Put the day-3 call on Thursday.
• Friday after 3pm. Low pickup and low attention. Use it for LinkedIn touches and list building.
• August and the week between Christmas and New Year. Pause cadences, do not let them run into out-of-office auto-replies. A sequence that fires three emails into a closed office starts over in September with a worse reputation.
• Bank holidays and bridge days. May in France is half holidays. The dialer should know the calendar.

The legal frame is simple for B2B in France and most of the EU: cold calls and emails to professional contacts are lawful under legitimate interest, as long as you identify yourself, keep a working opt-out and honour it on the spot.

Signals should break the sequence, not wait for it

The biggest flaw in a linear cadence is that it treats day 9 as day 9 regardless of what the prospect did on day 8.

If the contact visits your pricing page, opens the day-5 email three times, changes jobs, or their company posts a job for the role you sell to, the next touch is a call within the hour, whatever the calendar says. Reps use only about 20% of the signals already in their CRM, and 50% of our pipeline came from signals that were already there. A cadence that cannot be interrupted by a signal wastes most of them.

The practical rule: the sequence is the plan for silence. A signal is a louder instruction, and it wins.

Where cadences fail, and how to tell which part is broken

Three diagnostics, in order.

Connect rate below 8%. The problem is the list or the numbers, not the cadence. Check mobile coverage and the slot before you touch the copy.

Connects but no meetings. The opener. 65% of reps open by explaining what they sell. Trigger-based openers book 7-10% of conversations against 2-3% for pitch-first. Pull ten recorded calls and count.

Meetings but low volume. Not enough conversations. A manual dialer caps a rep at a handful of conversations a day. Parallel dialing gives about 5x more, which is the difference between completing the cadence for 50 contacts a week or 200.

How Sonik approaches outbound cadence

Sonik runs the phone side of the cadence as timed dialing sessions: the day-1, day-3 and day-5 calls for each contact land in the block where that persona picks up, with the AI call prep brief and the trigger on screen at connect.

Signals override the schedule. A website visit, a job change or a reopened thread moves the contact to the top of the current session. After each call, the AI logs the outcome and drafts the follow-up email that references it, so day 1 email and day 5 email actually mention the calls.

Our north star is 20 conversations per SDR per day. The cadence exists to produce them; the dialer is what makes the number reachable.

Conclusion

Call first, three times in five days, at the persona's slot. Let email and LinkedIn earn the next call. Respect the European calendar and let any signal jump the queue. This week, take your current sequence and move the first call to day 1 at 4pm for one segment. Compare conversations after five days with the segment that still starts with an email. That test settles the call-first debate faster than any article.