TLDR: For Swiss SaaS SMEs, a repeatable win/loss analysis program — interviewing the buyers who signed and the ones who walked — turns scattered deal losses into the sharpest go-to-market signal a small revenue team owns.
Swiss SaaS win rates halved in a year, and CRM notes cannot explain it
The economics of Swiss software selling shifted hard in a single year. In the 2025 go-to-market benchmark from Ebsta and Pavilion, the average business-to-business (B2B) win rate fell to 19 per cent, down from 29 per cent the year before — a read drawn from 655,000 opportunities and 48 billion dollars of pipeline. For a software-as-a-service (SaaS) small or mid-sized enterprise (SME) selling out of Zurich, Geneva or Lausanne, that shift is existential. A lean revenue operations (RevOps) function cannot absorb a ten-point drop in close rate the way a venture-funded American competitor can, because every lost deal in a small domestic market is harder and slower to replace.
The instinct, when the number drops, is to open the customer relationship management (CRM) system and read the loss reasons. That instinct is misplaced. Clozd, which runs win/loss programs at scale, reports that around 85 per cent of the loss reasons logged in a CRM are wrong, and that the reason a buyer gives for walking away matches the reason the seller recorded only about 15 per cent of the time. A rep marks a deal “lost on price” because price was the last objection voiced in the room. The buyer actually left because onboarding looked risky, or because an internal champion changed jobs mid-cycle. The field the forecast leans on is fiction, and no dashboard built on it can locate the real leak.
The decline is not spread evenly across deal sizes either. Median win rates fall as contract value climbs, because larger deals pull in more stakeholders and longer scrutiny — a pattern visible in benchmark data compiled across hundreds of SaaS companies. A Swiss founder reading only a blended win rate misses this entirely, and misreads a structural stakeholder problem as a pricing one.
None of these figures tell a specific Swiss company why its own deals die. They set the stakes. The one instrument that converts a falling benchmark into a decision a team can act on is a structured conversation with the people who actually made the buying call — which is exactly what a win/loss program is.
Nearly half of lost deals never chose a rival — they chose nothing
The most expensive losses are not the ones a competitor takes off the table. They are the deals that dissolve into indecision, and they are systematically miscoded. Matt Dixon and Ted McKenna analysed 2.5 million recorded sales conversations for The JOLT Effect and found that 40 to 60 per cent of deals a rep forecasts to close end in no decision at all — the buyer stays with the status quo rather than picking any vendor. In the roughly 30 per cent of calls carrying high buyer indecision, win rates collapsed to 6 per cent.
This matters more in Switzerland than the raw percentage suggests, because Swiss B2B buying skews conservative and consensus-driven. A cantonal institution, a private bank in Geneva, or a family-owned industrial firm in Vaud rarely lets one person sign; the decision travels through committees, legal review, and a cultural preference for not being wrong. Indecision is not a pricing objection or a feature gap — it is a fear of making a costly mistake — and it hides in the CRM under labels like “no budget” or “bad timing” that a busy rep reaches for when a deal simply goes quiet.
A win/loss interview is the only reliable way to separate the two. When a neutral interviewer asks a lost Swiss buyer what actually stopped them, the answer is frequently a risk the seller never surfaced: an unproven migration path, an unclear internal owner, a board that wanted one more reference. Those are fixable with proof, phased onboarding, and risk reversal — but only once a team knows indecision, not competition, is what is bleeding the pipeline.
The deals that closed carry more intelligence than the ones that died
Win/loss analysis is misnamed if it is read as loss analysis. The won deals are half the signal, and often the more actionable half, because they reveal what the team should do more of on purpose rather than by accident. Buyers are unusually clear about what moved them, and the research is consistent about what that is.
In a RAIN Group study of 528 buyers and sellers, 71 per cent of buyers said a thorough discovery of their needs was highly influential, 67 per cent were swayed by a clear return-on-investment case, and 64 per cent by a seller who educated them with new ideas. The gap sits on the sell side: only 16 per cent of sellers, by buyers’ account, actually build a convincing ROI case, and only 21 per cent genuinely differentiate themselves even though half of buyers reward it. A Swiss SME that interviews its won deals usually finds one or two reps who do these things instinctively — and can then turn a personal habit into a repeatable play.
Pattern-matching across won and lost interviews is where the program pays back. A theme that appears in three lost deals and is absent from every won one is a diagnosis, not an anecdote. It points at a specific fix — a discovery question that was skipped, an ROI model that was never built, a proof point the champion needed and never got — and it tells the founder which fix will move the next quarter’s number rather than the last one’s.
A win/loss program is four repeatable steps, not a yearly survey
The reason most SMEs never run this is that they imagine an expensive annual research project. It is closer to a light operational loop, and it works because it is small and constant. Four steps make it repeatable.
First, trigger on outcome: every deal above a set contract value that closes won or lost enters the queue automatically, so the program never depends on a rep remembering. Second, interview independently and fast — within two weeks, while memory is fresh, and crucially by someone other than the rep who owned the deal, because a buyer will not tell the seller who lost the deal the real reason to their face. Third, code the conversations into themes and route each theme to a named owner: pricing to the founder, product gaps to engineering, messaging to marketing. Fourth, feed the themes into the next go-to-market cycle and measure whether the win rate on that theme moves.
The payback is documented, not theoretical. Clozd reports that 63 per cent of companies with a win/loss program increased their win rates, and that 84 per cent of programs running for two or more years reported sustained win-rate growth — the compounding comes from the loop, not from any single interview. The discipline mirrors how a strong RevOps team already works: find the one funnel stage that is actually leaking, fix it, confirm the rate moved, then move to the next constraint.
Switzerland’s thin market and the nLPD shape how the loop runs
The same program needs Swiss calibration to work. The domestic market is small — roughly nine million people split across German-, French- and Italian-speaking regions — so a Vaud or Zug SaaS seller cannot burn through leads to average its way past a bad quarter the way a US firm can. Every lost deal is a larger share of a finite total addressable market, which makes the intelligence from each one compound harder, and makes a repeatable learning loop more valuable in Zurich or Geneva than it would be in Austin.
Language is the second adjustment. A win/loss interview only surfaces the truth in the buyer’s own language, which means running them in French for a Geneva or Lausanne account and in German for a Zurich or Zug one, with an interviewer fluent enough to catch hesitation and nuance rather than a translated summary. Outsourcing the interview to a neutral third party solves both the language coverage and the candour problem at once.
The third is legal, and it is not optional. The revised Federal Act on Data Protection (nLPD, or nFADP in its French and English forms) entered into force on 1 September 2023 and governs how a company records interviews and stores the personal data of the buyers it speaks to. Recording a win/loss call and retaining a named buyer’s feedback requires informed consent and a lawful basis, so consent belongs in the interview script from the first sentence. Handled well, that opening is not friction — it signals to a Swiss buyer that the firm takes their data as seriously as their business.
Turning lost Swiss deals into next quarter’s playbook
The through-line is simple: the benchmark decline is structural, the real loss reasons are invisible in the CRM, and the cheapest way to see them is a small, repeatable interview loop that treats won and lost deals as one dataset. What to do with that depends on the seat. A founder or chief executive should mandate the program and protect it from being quietly dropped when a quarter gets busy. A sales leader should separate the interviewer from the closer, so the feedback is honest. A RevOps owner should code every interview to a theme, route it to the person who can fix it, and report whether the win rate on that theme actually moved.
Pupsic builds this kind of RevOps loop for Swiss SMEs and startups — the plumbing that turns scattered deal outcomes into a go-to-market decision, in the buyer’s language and inside Swiss data rules. A team that wants to stop guessing why its deals close or die can start that conversation with Pupsic and run the first loop on the deals it has already lost this quarter.
Win/loss analysis in Switzerland: common questions
What is win/loss analysis for a SaaS company? It is a structured program that interviews the buyers behind recently closed deals — both won and lost — to learn the real reason each decision went the way it did, then feeds those reasons back into pricing, messaging, product and sales. It exists because the loss reasons recorded in a CRM are wrong roughly 85 per cent of the time, so the truth has to come from the buyer directly.
How many deals does an SME need to interview to get value? There is no fixed threshold, because the value comes from repetition rather than volume: a theme that recurs across a handful of lost deals and is absent from the won ones is already a diagnosis. A small Swiss team running interviews continuously on every deal above a set contract value will see actionable patterns within a single quarter.
Does the nLPD allow recording win/loss interviews in Switzerland? Yes, provided the program follows the revised Federal Act on Data Protection: obtain informed consent before recording, tell the buyer why their feedback is collected and how long it is kept, and store the personal data on a lawful basis. Building that consent into the interview script keeps the program compliant and reinforces buyer trust.
References
- Ebsta & Pavilion, 2025 GTM Benchmarks. https://benchmarks.ebsta.com/2025-gtm-benchmarks
- Matt Dixon & Ted McKenna / Challenger, The JOLT Effect: Losing to Customer Indecision. https://challengerinc.com/losing-to-customer-indecision/
- RAIN Group, 9 Ways to Influence Buyer Purchase Decisions (study of 528 buyers and sellers). https://www.rainsalestraining.com/blog/9-ways-to-influence-buyer-purchase-decisions
- Clozd, The Ultimate Guide to Win-Loss Analysis and 2025 State of Win-Loss Analysis Report. https://www.clozd.com/guides/win-loss-analysis
- Salesmotion, Sales Win Rate Benchmarks 2026 (compiling HubSpot and Optifai data). https://salesmotion.io/blog/sales-win-rate-benchmarks-2026
- Sidley Austin, The New Swiss Data Protection Act Enters Into Force on September 1, 2023. https://www.sidley.com/en/insights/publications/2022/04/a-wakeup-call-the-new-swiss-data-protection-act-enters-into-force-on-september-1-2023
