Your CRM is lying to you

Read time: 5 minutes

Welcome to the Succession newsletter where 1,000+ life science sales reps improve their skills in 5 minutes per week. If you’re getting value from these newsletters, we'd love it if you could forward it along to your sales colleagues. If you’re new here, subscribe below.

It’s that time of year when you’re probably frantically chasing for POs before Q4 ends, dealing with the Christmas party hangover (whilst wondering whether you said something you shouldn’t have to your boss after a few drinks), and receiving a barrage of ooo replies. 

Between that, many will be reviewing the year’s win/loss data.

Here's a question that should make you uncomfortable:

When was the last time you actually knew why you lost a deal?

Not the sanitised version that was typed into your CRM. Not the "budget constraints" or "went with a competitor", “Bad timing” checkbox. The actual reason.

Most sales teams run quarterly win/loss reviews, pull reports from their CRM, spot trends, and make decisions based on what they find. 

"We're losing too many deals to price. Let's build better ROI calculators and train the team on negotiation."

"Competitors keep beating us on delivery timelines. We need to speed up fulfilment."

"Technical objections are killing us. Product needs to prioritise these feature gaps."

Then they spend the next quarter fixing problems that might not actually exist, and win rate barely changes.

Most loss reasons logged in your CRM are educated guesses at best, and complete fiction at worst.

Your AE isn't lying to you, often they just genuinely don't know. The deal went quiet after the demo, or the champion stopped responding. Procurement said they'd "circle back" and never did. So they pick the most logical-sounding reason from the dropdown and move on.

But what seemed most logical and what is actually true are very different things.

If your entire coaching strategy, product roadmap, and competitive positioning are built on bad data, you're optimising for the wrong things.

Let's fix that.

The real cost of guessing

Bad win/loss data cascades through your entire organisation.

Sales leadership trains on the wrong skills.
If your CRM says you're losing to price, you'll spend Q1 teaching negotiation tactics and building ROI calculators. Meanwhile, the real problem is that your AEs can't articulate value in discovery, so prospects never see the point of your solution in the first place.

Product builds the wrong features.
Your product team pulls closed/lost feedback and sees "lacks integration with X platform" showing up repeatedly. They spend six months building that integration. Deals still don't close. Turns out the integration was never the blocker; it was just the easiest thing for the prospect to point to when they didn't want to admit their internal champion lost political capital.

Marketing targets the wrong messaging.
If the data says you're losing on "lack of validation data," marketing creates ten new case studies. But if the actual issue is that prospects don't trust your team to deliver on time based on past experiences, those case studies won't move the needle.

You end up in this cycle where everyone's working hard, making changes, iterating, but nothing improves because you're solving for symptoms, not root causes.

Why AEs don't actually know

Let's be clear, this isn't about AEs being lazy or incompetent. It's that deal loss is messy, multi-threaded, and often invisible to the person running it. 

The prospect goes dark.
You had three great calls and they seemed engaged, then nothing, radio silence. Your AE follows up twice, maybe three times, then marks it "timing wasn't right" and moves on. But what actually happened? Did the champion leave? Did funding get pulled? Did a competitor get to the CFO first? No idea.

The real objection never surfaces.
In life sciences, decisions are rarely made by one person, on average there are 7 people involved. You might have a scientist champion who loves your tech, but procurement hates the contract terms, finance thinks the ROI timeline is too long, and the VP doesn't trust your company's ability to scale with them. Your AE only hears "we need to table this for now." They log "budget" and move on.

The loss reason is political, not technical.
Your solution is better, your pricing is competitive, and your data is solid. But the decision-maker has a 15-year relationship with your competitor's sales director. Or there's an internal mandate to consolidate vendors. Or the previous project with your company went poorly, and someone high up still remembers. None of that makes it into your CRM.

Time erases memory.
By the time your AE is filling out the loss reason, it's been three weeks since the last real conversation. They vaguely remember something about delivery timelines, so they pick that. But if you went back and listened to the actual calls, you might hear the prospect was way more concerned about your company's regulatory compliance track record.

Lack of experience.
How many closed-lost reason codes never blame the seller? But sometimes experience is the difference between closing a tricky deal and losing it. We’ve all been there when we know we messed up or missed something, and the deal falls through. This is totally normal, but it’s important to recognise why we lost the deal, what areas we need to work and what training we need to get it over the line next time.

This isn't about blame. It's about recognising that subjective memory plus CRM dropdowns equals unreliable data. Then putting a plan in place to fix it.

How AI can help

What AI is genuinely good at is pattern recognition across large amounts of unstructured data. The kind of data your AEs can't possibly hold in their heads: dozens of calls, hundreds of emails, meeting notes, proposal comments, demo feedback.

We’re not replacing human judgment here, we’re just giving people a way to do an objective reality-check.

Here's the process:

Step 1: Gather the full story

For every closed deal (won or lost), pull together everything that happened:

• Call recordings and transcripts

• Email threads with the prospect

• Internal Slack conversations about the deal

• Proposal documents and any redlines

• Demo feedback forms or technical evaluation notes

Step 2: Let AI analyse the real story

Feed all of that context into Claude, ChatGPT, or whatever LLM you prefer. The key is giving it enough data to spot patterns you might have missed.

Example prompt:

I'm sharing transcripts, emails, and notes from a deal we lost. 

The AE logged this as "lost to competitor on price."

Please analyse all the materials and tell me:

1. What was the actual primary reason this deal didn't close?

2. What signals or patterns suggest this reason?

3. Were there any secondary factors that contributed?

4. At what stage did we likely lose control of this deal?

5. What could we have done differently?

Be specific. Quote exact language from calls or emails where relevant.

AI doesn't have ego, so it doesn't care about looking good. It just tells you what it sees in the data.

Step 3: Compare AI analysis to AE assessment

Now you've got two versions:

• What the AE logged in your CRM

• What AI found in the actual conversations

Sometimes they match. Great, your AE probably nailed it.

But when they don't match, you know where you need to pay more attention to.

Example from a biotech tools company:

AE logged: "Lost to competitor, they had a faster delivery timeline"

AI analysis: "Primary loss reason appears to be lack of confidence in implementation support. In the October 12 call, the lab director said 'we've heard from other labs that onboarding is quite long and requires a lot of input from our team' (timestamp 23:15). The AE pivoted to talk about delivery speed instead of addressing the concern. In follow-up emails, the prospect asked twice about implementation resources and received general answers rather than specifics. Competitor likely won on post-sale support, not delivery speed."

See the difference?

The AE wasn't wrong that the delivery came up. But it wasn't the real blocker. The real issue was trust in the team's ability to support them post-sale, which is a completely different problem to solve.

Step 4: Make it a conversation, not a gotcha

This is critical: the goal isn't to catch your AE being wrong. It's to help them get better at reading deals.

When there's a gap between the logged reason and the AI analysis, sit down together and review it:

"Hey, I ran the Genomics Lab deal through an analysis. You logged it as delivery timeline, but when I look at the transcripts, it seems like they kept circling back to implementation concerns. What do you remember about that?"

Most of the time, your AE will say "Oh yeah, that did come up. I guess I didn't realise how much it bothered them." This creates a learning moment where you can stop and discuss better approaches for when a similar situation comes up again.

Over time, your AEs get better at spotting the real objections in the moment, asking better questions, and logging more accurate data. Which means your win/loss analysis actually becomes useful.

What to do with better data

Once you've got reliable loss reasons, everything downstream gets easier.

Coaching becomes targeted.
If you're consistently losing deals because AEs aren't establishing urgency early enough, you can build role-play scenarios around that specific skill. No more generic objection handling training that misses the actual problem.

Product prioritisation gets smarter.
When product asks "why are we losing deals?", you can say with confidence, "We're losing because prospects don't trust our regulatory compliance track record, not because we're missing feature X." 

Competitive intelligence sharpens.
Instead of "we're losing to Competitor Y," you can say "Competitor Y is winning by offering dedicated implementation support in the first 90 days, which we don't currently provide." Now you know exactly what to counter.

Marketing creates content that matters.
If the real issue is that scientists don't trust your data quality, marketing can focus on validation studies and QC processes, not generic brand awareness.

The weekly habit that compounds

Here's how to make this practical without drowning in process:

Every Friday, pick three closed deals from that week (wins and losses). Run them through the AI analysis. Compare to what got logged and share insights with the team.

This builds a habit of pressure-testing your assumptions.

After a month, you'll start seeing patterns. After a quarter, your team's self-awareness will be noticeably better. After six months, your win/loss data will actually be something you can trust.

Start small, prove it works

If this sounds like a lot, here's the smallest possible version:

Pick one deal you lost in the last two weeks. Pull the call transcripts. Feed them into Claude or ChatGPT with a simple prompt: "Why did we lose this deal?"

See what it tells you.

Then compare that to what got logged in your CRM.

If they match, great. If they don't, you've just found a gap worth exploring.

Most sales teams will never do this. They'll keep running quarterly win/loss reviews based on bad data, wondering why their coaching and product changes don't move the needle.

You don't have to be one of them.

Capturing more data for better decisions

This whole process only works if you've got decent data to analyse, and you can always try to capture more to make it more accurate.

You're probably already sitting on most of what you need. Here's how to capture it better:

• Brain dump after calls: Two-minute voice note with raw impressions before the subtext is forgotten

• Use AI note-takers: Fathom, Fireflies, Grain, etc capture tone, cadence, and who dominated the conversation

• Collect multiple perspectives: Each stakeholder on your side saw different parts of the deal and probably interacted with different stakeholders on the customer’s side. Get all their notes and ask them for their insights.

• Track digital breadcrumbs: Did they download validation data? Copy in a colleague? Stop replying to emails after a specific meeting?

• Notice what didn't get said: Budget dodged three times? Never introduced to the decision-maker? Identify where the gaps are and use them as additional data points.

• Document internal champion strength: Are they bringing others to calls? Asking smart questions? Or just being polite?

• Track timeline slippage: Every time a close date moves, note what reason was given (patterns emerge fast).

• Record internal deal reviews: The post-call debrief often surfaces insights that never make it to the CRM

• Log competitor mentions: When they come up, how they come up, and what specific features got compared.

The more you capture, the smarter your analysis gets. Start with what you've got, layer in more as the habit builds.

P.S. If you want help setting this up for your team (or building the prompts and workflows), we can help. Book a strategy call, and we'll walk you through it.

And last but not least, wishing you a very merry Christmas from the whole Succession team.

Apple Podcast | Spotify | YouTube

Episode 77: [Leadership] Career progression from the bench to sales leadership with Heather Brown

1. Lead Generation: We’ll build target lists, write scientifically relevant messaging, and send messages on your behalf to book qualified sales meetings with biotech and pharma companies.

2. Training for Reps: A skill development platform for life science sales reps who want to improve their sales skills, exceed their quota, and take the next step in their career.

3. Training for Teams: If you want to upskill your team around prospecting, driving to close, key account management, AI, or any other topic, we can put together a training plan specific to your organization’s needs.

4. Strategy Call: Need more than training? Want help implementing and executing your sales strategy? In a 30-minute call, we will assess your company’s current situation and identify growth opportunities.