Sales pipeline optimisation is the work of making deals move through your pipeline at a steady, predictable pace, rather than the work of adding more deals to it. The levers that matter most sit at the top of the funnel: what enters, how fast it gets a conversation, and whether anything stalls. Here are the nine strategies we see working across European B2B SaaS teams, with benchmark numbers from our analysis of 1,000,000 B2B cold calls in France.

Pipeline optimisation is about flow, not size

A pipeline with 300 open deals and a 20% close rate is less predictable than one with 120 deals and a 35% close rate, because the first one hides 150 deals nobody has touched in a month. Size makes a forecast look good. Flow makes it come true.

Flow has three components you can measure: what enters each week and why, how long deals spend in each stage, and how many leave each stage in each direction. Optimisation means improving those three numbers, in that order.

Our position is simple. Connect rate is an input, not an outcome, and the same goes for pipeline coverage. If you want a predictable quarter, work on the inputs that happen eight weeks before it.

Three numbers that tell you flow is broken

• Entry mix: the share of new opportunities that came from a signal (website visit, job change, dormant lead, closed-lost) versus a cold list. At Sonik 50% of our pipeline comes from signals that were already in the CRM; most teams use about 20% of those signals.
• Days since last buyer action: the age of each deal measured from the last thing the buyer did, not the last thing the rep logged.
• Conversations per SDR per day: our north star is 20. Below 10, nothing downstream is predictable because the sample is too small to forecast from.

9 sales pipeline optimisation strategies

1. Gate entry with a signal

An opportunity enters the pipeline when there is a reason the account should buy now, not when a rep got a polite reply. Write the trigger on the deal record. Deals without one are leads, not pipeline.

2. Call signals the same day

A website visit or a job change is a timer. In our experience a signal is worth a call the same day and worth very little after a week. Build the daily call list from signals first and cold lists second.

3. Dial when people pick up

Connect rate in our French data is 11.3% on average, 15.8% at 4pm, 14% at 10am and 9.8% at 8am. Founders answer around 10am, marketing around 4pm, HR around 1pm. Moving dial blocks costs nothing and is the fastest conversation increase most teams will ever get.

4. Make callbacks a stage event

A prospect who calls back after a missed call or a voicemail has done something no cold list does: they chose to talk. Treat the callback as a conversation-stage event, route it to the rep who left the message within minutes, and keep the trigger on screen. In our experience, most connects happen within the first three attempts, so a three-attempt rule inside five days captures most of the value; after that, the number goes back to the signal list with a date rather than sitting in limbo.

5. Open with the trigger, not the pitch

Trigger-based openers book meetings on 7-10% of conversations; pitch-first openers get 2-3%. Yet 65% of reps still open by explaining what they sell. This one change roughly triples the top of your pipeline without a single extra dial.

6. Multithread before the proposal

Single-threaded deals are the main source of late-stage surprises. The rule we recommend: no proposal goes out until a second person on the buyer side has joined a call. Objections surface earlier, and the deal survives when your champion goes on holiday.

7. Age out stalled deals every week

Set a maximum number of days per stage from your own history, then move anything older back a stage or out of the pipeline. A deal that leaves is not lost; it goes back to the signal list with a date. What hurts a forecast is not a lost deal but a deal that is neither alive nor dead.

8. Forecast from conversations, not from stage labels

Run the funnel backwards. If you need 10 new opportunities a week and your meeting-to-opportunity rate is 50%, you need 20 meetings held. At a 7-10% meeting rate per conversation that is 200-285 conversations, or 10-14 per day for a team of two SDRs. Those are activity numbers you can check on a Wednesday, not a stage label you discover was wrong in week twelve.

9. Close the loop on lost and dormant deals

Closed-lost and no-decision deals are your warmest list next year. A call to a known contact that opens with "How have you been?" reaches about 10% success in our data, against 1.5% for a standard cold open. Schedule the re-entry date at the moment you mark the deal as lost.

Which strategies first, by team

• Small team, high volume (SMB SaaS): strategies 3, 5 and 8. Slots, openers and a conversation target fix most of the variance.
• Enterprise, long cycles: strategies 1, 6 and 7. Entry quality and multithreading matter more than volume, and ageing rules stop the forecast from drifting.
• New team or new reps: strategies 2, 4 and 9. Signals and callbacks give new reps warmer conversations while they learn the cold ones.

How Sonik approaches pipeline optimisation

Sonik works on the inputs. Signals turn website visits, job changes, closed-lost anniversaries and dormant leads into a daily call list, so entry quality and same-day calling happen by default rather than by discipline.

The parallel dialer gets reps to 20 conversations a day with about 5x more conversations per hour than single-line dialing, and callbacks are routed to the rep who left the message with the trigger on screen.

The dashboard reports conversations, meeting rate per conversation and entry mix per rep and per slot, which are the three numbers a weekly pipeline review needs. Call coaching flags pitch-first openers, so strategy 5 does not depend on a manager listening to every call.

Conclusion

Predictable growth comes from a pipeline that moves, not a pipeline that is full. Gate entry with signals, dial when people answer, open with the trigger, and age out what has stopped moving. This week, add one field to your CRM, "days since last buyer action", sort your open deals by it, and have each rep decide what to do with their top five.