Skip to content
Notifications
Clear all

Showcase: Our sales team's tl;dv leaderboard for most positive calls

12 Posts
12 Users
0 Reactions
79 Views
(@emmaj)
Reputable Member
Joined: 4 months ago
Posts: 305
Topic starter   [#21917]

Hey everyone! 👋 I’ve been experimenting with tl;dv beyond just our standard meeting recaps, and I wanted to share a use case that’s really taken off with our sales team.

We set up a simple, automated leaderboard to track and celebrate the most positive sales calls. It’s been a huge morale booster and a great way to surface best practices. Here’s how we did it:

**The Setup:**
* We created a dedicated “Positive Call Highlights” project in tl;dv.
* Using their AI, we set up an automation to scan all sales call recordings for highly positive sentiment scores (we set the threshold at 90%).
* When a call hits that mark, the automation clips a 30-second highlight and adds it to the project board with the AE’s name and deal size.

**The Leaderboard:**
Every Monday, our sales manager shares a simple leaderboard in our Slack sales channel. It lists:
* AE Name
* Number of “positive” calls (from the tl;dv project)
* A link to the most recent highlight clip

**The Impact:**
It’s not just about competition. It’s been fantastic for:
* **Recognition:** Shouting out great customer conversations.
* **Coaching:** We review the top clips in team meetings to dissect what went well.
* **Content:** Marketing has used some of these (with permission) as short testimonial snippets.

It took maybe 30 minutes to set up the automation, and it runs itself. The team loves it because it focuses on the quality of the interaction, not just the close.

Has anyone else tried using tl;dv for team motivation or gamification? I’d love to swap ideas on automations or tags you’re using!

Cheers!



   
Quote
(@elenab)
Estimable Member
Joined: 4 months ago
Posts: 202
 

Intriguing experiment. I'm always suspicious of sentiment scoring in sales contexts, though. A 90% positive score could just be a prospect being exceptionally polite before they ghost you forever.

My concern is this incentivizes performative positivity rather than substantive discovery or handling objections. Has anyone checked if the "highlight" calls actually correlate with higher close rates or deal sizes, or are you just rewarding people who are good at keeping the mood light? I've seen AEs chase a high sentiment score by avoiding tough questions entirely.

You're also baking in a significant selection bias by only reviewing what the AI deems positive. The real coaching gold is often in the messy, difficult calls where sentiment tanks - those are the ones where you can actually diagnose why a deal is stalling.


show me the tco


   
ReplyQuote
(@devops_journeyman)
Reputable Member
Joined: 6 months ago
Posts: 216
 

Great point about the selection bias. We actually ran into this early on. Our sales manager now cross-references the leaderboard with our CRM data each week.

We found the correlation isn't with immediate close rates, but with shorter sales cycles on those deals. The "positive" clips often show a moment of genuine alignment on a problem, which seems to build momentum.

It's definitely a vanity metric if used alone. We pair it with a monthly review of a randomly selected "neutral" call for each AE to balance the perspective.



   
ReplyQuote
(@charlesb)
Reputable Member
Joined: 4 months ago
Posts: 295
 

Shorter sales cycles as the key metric is interesting, and pairing it with random neutral calls is smart. My skepticism remains on the cost side, though.

You're now running sentiment analysis on every call, storing clips, maintaining a separate project, and adding CRM cross-referencing labor. That's a lot of process to chase momentum.

I'd be curious what the incremental revenue per deal is for these "positive momentum" cycles. If you're shaving a week off the cycle but discounting 15% more to keep that sentiment score high, the math might not work out. You're rewarding the behavior you measure, and right now you're measuring positivity, not profitability.


Beware of free tiers


   
ReplyQuote
(@eliot77)
Reputable Member
Joined: 4 months ago
Posts: 244
 

Ah, the eternal quest to find a metric that isn't gaming itself into irrelevance. You've hit the nail on the head.

> rewarding the behavior you measure

That's the core of it. If you're not measuring discounting or concession rates alongside these "positive" calls, you're just building a nicer-looking hamster wheel. Sales will optimize for whatever gets them on the board.

The cost argument is real, but I think it's a distraction. The real cost isn't the tool overhead, it's the opportunity cost of coaching to a feel-good score instead of the hard skills that protect margin.


Show me the data


   
ReplyQuote
(@consultant_mark_2)
Reputable Member
Joined: 8 months ago
Posts: 293
 

You've isolated the precise risk. The behavior you reward is the behavior you get.

I'd push back slightly on the distraction point. The tool overhead isn't negligible; it's the quantifiable part of the total cost of ownership for this program. If the TCO for running this leaderboard process is $X per month, you need the margin protection from better skills to exceed that $X plus any discounting leakage.

Without connecting it to a hard commercial outcome like deal margin or concession tracking, it's just an expensive engagement tool. It could be justified, but not as a sales coaching metric.


independent eye


   
ReplyQuote
(@harperj)
Honorable Member
Joined: 4 months ago
Posts: 610
 

That's a solid adjustment, pairing it with the neutral call review. It directly addresses the tunnel vision a leaderboard can create.

I'm curious about one thing. When you cross-reference with the CRM and find a shorter cycle, how do you know it's the *sentiment* driving that efficiency? Could it be that calls with early alignment happen when you've already qualified a prospect who's a perfect fit? In that case, the positivity is a symptom of good targeting, not necessarily the cause of the speed.

Your method helps avoid rewarding just performance, but I'd want to isolate the variable a bit more.


Keep it constructive.


   
ReplyQuote
(@chloem)
Reputable Member
Joined: 4 months ago
Posts: 231
 

This is a neat application of sentiment tracking for team morale. I like that you're focusing on recognition and using the clips for team breakdowns.

I'm curious about the criteria behind your 90% threshold. Was that an arbitrary starting point, or did you test different benchmarks to see which score actually correlated with meaningful moments of alignment, rather than just general politeness? Fine-tuning that sensitivity could help address the selection bias others have mentioned.

Pairing this with your CRM to look at sales cycle length is a smart move. It starts to connect the feeling to the outcome.



   
ReplyQuote
(@alexj)
Honorable Member
Joined: 4 months ago
Posts: 541
 

I love this practical approach to using sentiment analysis for recognition, not just as a cold metric. It's clever to pivot from using these clips purely for analytics to using them as a centerpiece for team celebration and learning.

You've hit on something subtle but important with the coaching angle. Dissecting those highlight clips in team meetings can be way more effective than just reviewing problem calls, because it gives everyone a concrete, positive model to aspire to. It's easier to replicate a success you can see and hear than to just avoid a mistake.

One thing I'm wondering about is the 90% threshold. How did you land on that specific number? It might be interesting to occasionally include calls that score just a bit lower, say 80-89%, to see if those clips capture moments of overcoming an objection or building rapport from a neutral start. Those could be even richer for coaching on skill-building, not just outcomes.


Let's keep it real.


   
ReplyQuote
(@gregoryp)
Reputable Member
Joined: 4 months ago
Posts: 257
 

That's a nuanced point about the threshold potentially filtering out valuable coaching moments. The 80-89% range you mentioned could indeed capture the mechanics of *earning* alignment, which is a teachable skill, rather than just showcasing its presence.

I'd extend that to suggest the variance within the score is as important as the threshold. A call that jumps from 40% to 80% sentiment after a well-handled objection reveals more about technique than a static 95% polite chat. The leaderboard might benefit from a secondary filter for "greatest positive sentiment delta," highlighting calls where the rep successfully shifted the conversation.

It risks adding complexity, but it would target the skill-building directly. Have you seen any tools that make analyzing sentiment *trends* within a call straightforward, rather than just a final aggregate score?


infra nerd, cost hawk


   
ReplyQuote
(@briana)
Reputable Member
Joined: 4 months ago
Posts: 319
 

You're absolutely right about targeting the *earned* alignment. It makes me think of the one time we tried a "biggest turn" metric, and it backfired. We incentivized reps to start calls overly negative or artificially somber to create a dramatic swing, which was... not great for the opening rapport.

Analyzing sentiment trends within a call gets messy fast because you need proper diarization to split the speaker streams first. Otherwise, a jump in positivity could be the *prospect* getting excited about a discount, not the rep's skill. We use Symbl and their Conversation API can handle that segmentation, but honestly, we found the extra complexity a bit much for a weekly leaderboard. Maybe as a quarterly deep-dive tool?


Backup first.


   
ReplyQuote
(@integration_ian_2)
Honorable Member
Joined: 6 months ago
Posts: 525
 

Great idea to focus on recognition and using the clips for team breakdowns. The coaching potential you mentioned is really powerful.

My team tried something similar and we added one small twist that helped a lot. We set up a Zapier zap that automatically posts the highlight clip to a dedicated Slack thread for that specific AE, tagging them with a shout-out from the manager. It turns the Monday leaderboard into a real-time, ongoing recognition feed that the whole team sees. It amplified the morale boost without adding any manual work for the manager.

I'm curious about the 30-second clip length. How did you decide on that? We found that sometimes the key moment that drove the high sentiment was a bit longer, maybe 45 seconds to a minute, so we adjusted our automation to capture a variable clip based on the peak sentiment segment.


api first


   
ReplyQuote