Amanda Zhu

Co-Founder

❗Your engineers are wasting 6+ months building meeting integrations.❗

Why? Because every platform (Zoom, Google Meet, MS Teams) has different APIs, constant updates, and hidden edge cases.

❌ Building in-house = wasted engineering time.
❌ Maintaining it = ongoing distractions from core product development.
❌ Scaling it = reliability nightmares that break at the worst time.

That’s why we built Recall.ai.

✅ One API to handle every meeting integration for you.
✅ Ship in days, not months.
✅ Powering 1200+ companies, from early-stage startups to billion-dollar SaaS teams.

🚀 See how Recall.ai can save your engineers months of work

✔︎ Who gets results using Recall
∙ Product teams at fast-growing startups who want to go to market in days instead of months. 
∙ Engineering teams that don't want to burn engineering hours maintaining their meeting bot integrations.

✔︎ Here’s what we do
We provide one API for meeting bots on every video conferencing platform. We abstract over differences between meeting platforms (even for platforms that don’t have an APIs). We run, monitor, and scale the complex real-time video infrastructure.

Interested in learning more? Go to: https://recall.ai 🌐

Recruiters offer our sales reps $400K roles. None have left.


Our reps are having a career year.

Our AE Liz closed $3M.

Every rep is over 100% quota.
And they do it while helping each other win.

In June, we launched self-serve.
And changed our pricing structure.

But our sales team didn't complain.
They leaned in.

John closed $200K+ in one quarter after our product launch.
Sam stepped up to close HubSpot when our VP of Sales got stranded in Nebraska.
Paul closed two deals while vacationing in Italy.
Marcus booked 20% of our outbound meetings.
Steve brought in 230+ new customers.

As a founder, there are a lot of moments that make the job hard.
Getting to work with a sales team like this isn't one of them.

The API for meeting recording | Co-founder at Recall.ai

I landed six $1M deals last year. Here’s when/how I rescue 7 figure deals by flying to prospects: 

I fly out when a deal hits a roadblock,
and I don’t know why.

This is the playbook:

1/ Make it easy to say yes

I never ask:
“Can I fly out to meet you?” 

Instead, I’m specific: 
“I’m going to be in Atlanta on Friday from 2-7pm. 
Could we connect for dinner?”

Why this works:
- it signals effort without desperation
- a tight time window is easier to say yes to
- it reduces coordination work for a busy exec

-----

2/  Fly out to uncover, not to pitch

In one deal, I learned that a decision maker was about to say no. 

That surprised me.
On Zoom, she never raised an objection.

I flew out to meet her,
and that’s when she told me the truth: 
she was more committed to her internal solution than I realized.

That information let me walk through:
- the edge cases she hadn’t considered
- the ceiling she’d hit
- the tradeoffs she was about to own

Why this works:
- you’re meeting them on their turf
- you have the exec’s full attention
- they don’t need to perform in front of their team

-----

3/ Be picky about where you meet

The setting determines the conversation you get. 

If I need to meet with multiple stakeholders,
I meet at their office.

It’s easier to loop people in
and align everyone at once.

If I need one exec’s honest take,
I meet outside HQ.

Why this works:
- offices optimize for alignment
- out-of-office settings optimize for candor
- the room shapes what people feel safe saying

-----

Flying out isn’t just about pushing the deal forward.
It’s about finding out what’s actually blocking it.

No alternative text description for this image

I wanted to sponsor a dev conference for $100K. My marketer said no. He spent $12,500 on lunch instead. $2.6M return.

We’d just raised our Series B.

I wanted to get the word out. 
Fill top of funnel.
Make some noise.

A dev conference seemed like the obvious move.
$100K for a booth, some swag, maybe a speaking slot.

But my marketer pushed back.

“We don’t have the bandwidth to staff a booth for three days.”

He proposed something simpler.
“What if we just bought developers lunch?”

I posted a short note on LinkedIn:
“Lunch is on us. We just raised our Series B and want to celebrate with SF devs.”

We linked a Typeform.
We manually reviewed every submission and sent out $25 DoorDash credits to 500 developers.

Total cost: $12,500.

No follow-up email.
No pitch.
No “book a demo” link.
We didn’t even send them to our website.

Over the next four months, the ROI started to show.
Those 500 developers remembered the gesture.
Some signed up for the product.
Others brought their entire teams or sent us referrals.

$2.6M in revenue traced back to that one day.

It came from 500 people who had a positive, friction-less interaction with our brand because we didn’t try to sell them anything.

Lost a $815K deal. 6 months later, their CTO Slacked: “Read your HN article. Are you free today?”

Three months of conversations.
Clear use case.
Detailed technical validation.

Then the CTO reached out with a short message:
“We’ve decided to build this ourselves.”

No further explanation.
The deal was dead.

Six months later, my Slack lit up. Same CTO.
“Read your HN article. Are you free today?”

On the call, he was candid:

They’d spent 6 months building. Five engineers.
They got it working for one meeting platform. Stable-ish.
Nowhere near production-ready.

Then someone on their team sent the CTO our HackerNews article about Postgres scaling issues.
He read it and realized: we were solving problems they hadn’t hit yet, but he could see them coming.

They needed to ship their product. Not become an infrastructure company.

We signed two days later.
They were live in two weeks.

Sometimes the fastest way to close a deal is to let the customer try to build it.
When they see you’re operating one layer deeper than they ever want to go,
they’ll come back on their own.

I track WFH as PTO. Gray areas kill company culture faster than bad hires.


We’re a five-days-a-week, in-office company.

The hardest part of making that work is avoiding ambiguity.

The gray area looks like this:

Someone flies to Boston to attend their sister’s graduation.
They say they’ll work remotely.
They’re “technically” working.
Kind of available.
Half present.

In practice, that creates confusion.
Are they off?
Are they on?
What should others expect?

I chose clarity instead.

If you’re away, you’re off.
WFH is marked as PTO, and PTO doesn’t count against you.

If you’re traveling, enjoy it.
If you’re in the office, be fully here.

I’d rather someone disconnect completely than try to split attention.

Clear expectations create trust.
Ambiguity erodes it.

Cluely is a $120M company. They became a customer because we didn’t treat them like everyone else.

2:04 AM. Self-serve signup.
Cluely.

We went back and forth a bit.
Then we realized they were a 15-minute Waymo ride away.
So instead of adding another call, we showed up.

I’d been watching Cluely for a while.

They taken bigger swings with a higher hit rate than companies 10x their size by being:
intentional,
opinionated,
completely uninterested in generic playbooks.

That set the bar.

So we brought three things:

1/ a Teenage Engineering Recorder, engraved with their name

I picked the recorder because I personally think it’s cool, 
and because it actually maps to what we do.

I chose to engrave Cluely, not Recall.ai, to make it irreversible.

If this went nowhere, the effort was still spent.

The gift was one of one.
Swag without risk is meaningless. 

-----

2/ 10 lbs of chicken

We offered pizza. Their engineer said they wanted protein.
So we brought 10 lbs of chicken.

It seems trivial, but because we showed up exactly how they asked, 
the tone for the rest of the visit was set.

We’re here to understand what you’re actually building.

-----

3/ Tucker V., our engineer

Normally, this would’ve been a sales meeting.
Engineering would’ve come in later.

Instead, we brought Tucker on the first visit.
He worked with their engineers to start implementing Recall.ai live.

Instead of hypotheticals, now the discussion was grounded in what worked and what didn’t.

-----

By the end of the visit, there wasn’t much left to debate.

We’d already done the work.
We’d already made the tradeoffs visible.
And both sides knew exactly what this would look like in practice.

That’s how Cluely became a customer.

Wanting to work at OpenAI is a startup red flag.


Startups can’t and shouldn’t compete with OpenAI on salary or prestige. 
Especially at an early-stage.

So the people who actually fit startups are those who are skilled, but don’t want to work at OpenAI.

Because at OpenAI’s size,
- you’re rarely working on the thing that moves the business
- you don’t get to see how the whole system works end to end

It’s a tradeoff.

Early-stage startups need people who want exposure to consequence.
People who want to see how decisions ripple through the entire company.

I don’t want the people who want OpenAI now.
I want the people who wanted to join OpenAI before it was cool.

I’m not the reason our $250M company exists. 
If you’ve talked to my cofounder, you already know why.

One thing David is unusually good at is getting to the actual problem.

If I’m struggling to fill a role and bring it to him, the conversation doesn’t stop at “that’s hard” or “we’ll figure it out.”

It turns into:
- ”What have you tried so far?”
- “Who have you reached out to?”
- ”Did you test X? What about Y? Why not Z?”

I always leave those conversations with more work.
And that’s exactly what I need in the moment.

-----

That way of thinking sets the tone for how we operate.

It’s uncomfortable to be explicit about working long hours.
A lot of companies dance around it.

David doesn’t.

He’s clear about what this phase demands and steps up alongside everyone else. 
That clarity matters more than pretending otherwise.

-----

You can see his influence everywhere at Recall.ai.

A lot of teams try to do too many things at once. Parallelize everything.
David pushes the opposite.

One thing at a time. 
Finish it completely.

Doing things sequentially feels slower, but it consistently produces better outcomes.

“Done” doesn’t mean shipped.
Done means it actually works.
Done means people adopt it.

That mindset spreads.

------

Problems don’t get smaller around David.
They get clearer.
And you leave knowing exactly what to do next.

Which, in my experience, is why this works.

The expensive habit I had to unlearn managing a $250M company:
answering my team’s questions too quickly.

Before, these conversations were automatic.
Direct report: “Hey, I’m having trouble with this.”
Me: “Okay, then do X, Y, Z.”

But as the company grew, so did the problems.
Scopes got larger.
Situations got more nuanced.
The people asking often had teams of their own.

Every time I jumped in, 
I’d have to reconstruct context I didn’t fully have,
solve a problem I wasn’t closest to,
and walk away thinking: “that’s another thing on my plate.”

Now, when a direct report brings me a problem, 
I don’t respond with a solution.

I ask one question:
“What’s your plan?”

That does three things immediately:
- it keeps ownership with them
- it gives them the opportunity to figure it out 
- it lets me evaluate their abilities

I can still give feedback.
I can still pressure-test the plan. 
But I’m no longer carrying the problem for them. 

Teaching judgement takes more patience upfront.
But solving everyone’s problems costs you time and hides where your team actually needs to grow.

$623K deal. Two weeks to close. Then my champion went dark.


We’d been working this deal for 6 weeks with their VP of Engineering.
Kickoff was on the calendar.
I sent a final check-in.

My email to him bounced.
Out of office. Cabo. Back in 10 days.

I wasn’t worried.
Until our AE pinged at 4pm:
”Someone from product just killed the deal.”

A new stakeholder I’d never met had stepped in.
She reviewed everything and said no.

Her objections were familiar:
- risk of migrating
- fear of being on the hook if this broke in production

Every one of those had already been worked through with the VP of Engineering.
She just hadn’t been part of those conversations.

With 2 weeks until kickoff and no champion, the deal was effectively dead.
So I got on a plane to NYC that night.

The next morning, I grabbed breakfast with her.
And just listened.

She walked through her concerns. I brought her up to speed on everything she’d missed.
The 6 weeks of discovery. The plan. The commitments we’d made.

We signed 10 days after I landed in NYC.

When a new buyer says no, they’re rarely blocking the deal.
They’re protecting themselves from owning risk they didn’t sign up for.

“You’re here to replace me.” The lead engineer’s voice was shaking.
Two weeks later, he pushed our $800K deal through.

Up until that moment, Mike hated me.

Every sales call was a fight.

“Our current system already does this.”
“We don’t need another platform.”

I tried acknowledging his concerns.
I offered customer intros.
I tried logic.

Nothing worked.
The deal was dying.

So on a 1:1 call, I finally asked him:
“Why are you so against this?”

Long pause.

“You’re here to replace me.”

That’s when it clicked.
I was calling his baby ugly.

I was asking him to rip out months of his work.
To admit his system didn’t matter anymore.

I was trying to solve an emotional problem with a logical solution.
On the next call, I reframed the deal to make him the hero.

Instead of talking about what our product would replace,
I focused on what he would unlock.

He’d ship the new infrastructure in 2 weeks,
Get credit when it launches,
And finally get to spend time on interesting problems instead of babysitting plumbing.

Suddenly, the same engineer who was shutting us down
Became the one pulling the deal through procurement.

2 weeks later, I closed for $800K.

Most “hostile” stakeholders aren’t anti-you.

If you make their win obvious,
They’ll do more for the deal than you ever could.

I got 135 DMs last month asking “How on earth are companies paying you $1M+”


The short answer isn’t sales tactics or clever pricing.

It’s that LLMs can now automate most data entry and admin work.

Sales notes don’t need to be typed into a CRM.
Interview feedback doesn’t need to be copied into an ATS.
Standups don’t need to be summarized into an issue tracker.

Knowledge workers spend up to 70% of their time on admin work.

But for an LLM to do this well, it needs deep context.

Most of that context isn’t written down.
It’s in conversations.

And conversations are the largest human-generated dataset in the world.

There are ~0.88 quadrillion words on the internet across all of history.
Knowledge workers speak ~3.3 quadrillion words every year at work.

That’s why Recall.ai exists.

Recall.ai is the API for meeting recording.

Developers use Recall.ai to capture recordings, transcripts, and speaker attribution from meetings, desktops, and phone calls - without managing bots, integrations, or real-time infrastructure.

Our customers are building AI products that run on conversation data.

We price on usage.
Our new pricing starts at $0.50 per meeting hour and scales with volume.

The ROI shows up as:
- shipping months faster
- not staffing teams to maintaining brittle infra
- engineers spending time on product differentiation instead of plumbing

I’m deeply grateful to the teams and engineers who trusted us and are building on top of Recall.ai every day.

It’s still early, but it’s becoming clear every serious AI product will run on conversations.

I planned to hire 20 AEs to 3x revenue to $XXM. We did it with 5.


One rep brought in 230+ new logos.
Another closed a 7-figure deal on their 2nd month.

We passed on hundreds of candidates to find these people.
They’re incredible.

Now demand is outpacing what 5 people can handle.
So we’re hiring number 6.

If you’ve closed enterprise deals and want to see what you can do here, check out the JD below.

After growing 3x and raising our Series B, my co-founder is paying it forward to infra founders in SF.

The last 5 years, we learned how hard infra actually is:
finding PMF, earning trust, and executing at scale.

We only got here because other founders were generous with their time and advice.

David Gu wants to do the same.

If you’re building (or thinking about building) an infra company in SF, he’s opening up office hours for coffee and conversations.

Let me know in the comments, and I’ll connect you.

If I were to go back to $10M ARR, I would tell myself to walk away from that multi-million dollar deal. 

At the time, it felt insane not to chase.

Huge logo.
Real usage.
Their engineers were excited.

On paper, it looked like a perfect deal.
But I missed something in discovery.

The prospect’s problem was real, but it wasn’t on fire.
No one’s job depended on fixing it now.

So the deal stalled.

Looking back, harder discovery would have made this obvious:
- “Who feels the pain personally?”
- “What breaks if this doesn’t get solved this quarter?”
- “What internal initiative is this competing with?”

If I had pushed on those questions,
I would have found out before the deal died that 
the prospect’s problem wasn’t urgent enough yet. 

Big deals are expensive when they’re mistimed.
They pull you away from deals that can close,
and customers who can actually succeed with your product.

Learning when to walk away from a large deal is one of the most important sales skills you can develop.

As a co-founder of a $250M startup, here’s how I get direct reports to own things outside of their job description.

First, a misconception.

Most people don’t avoid ownership.
They avoid owning things that feel optional.

That’s a communication and hiring problem.

Here’s how I address that:

1/ Make ownership explicit by tying it to consequence

When I want someone to own a project,
I don’t assume they understand why it matters. 

I make the expectation and the consequence clear.

For example, I’ll say: 
“This is a top priority.
This came directly from a customer who’s blocked on it.
If we don’t solve this, we risk losing them.”

That does two things at once:
- it clarifies why the work matters
- it sets expectations

-----

2/ Hire for people who optimize for impact

Some candidates are drawn to the work itself.
A specific programming language.
A tool.
A narrow type of work they enjoy.

Others care about impact.
It’s less about the how, more about
what are we doing, and why are we doing it.

In interviews, I listen for the difference.

When someone describes their past work, I’ll ask:
- “Why did you choose to do that?”
- “Why did that matter to the business?”
- “What changed because of it?”

People who are impact-driven can explain:
- the business reason behind the work
- the tradeoffs they made
- the outcome they were responsible for

Impact-driven people are the ones who adapt when the job keeps changing.

-----

Startups don’t stay static long enough for rigid roles.

To manage through constant change,
you have to make outcomes explicit and hire people who can own them.

Shortly after raising our $38M Series B, we realized we had a problem: we had no system for surfacing what our reps were learning from the market.

Reps talk to prospects every day.
They hear people are saying.
They see objections evolving.
They notice when the buying committee changes.

But that intelligence was disappearing.

Yes, you can watch sales calls. That takes hours.
Yes, you can ask people. But it’s random and incomplete.

We needed a way to systematize it.

That’s when I started running situation rooms.

Here’s how they work:
- First 5 minutes: everyone fills out a form together (blockers, customer intel, priorities)
- Next 5 minutes: everyone reads
- Rest of the hour: we talk through what’s surfacing

The form is public. Anyone can read it.
Engineering sees what sales is hearing.
Marketing sees the patterns across deals.

The repetition matters.
Every week, the same format. The same questions.
It keeps critical information top of mind.

And it creates space for knowledge transfer that doesn’t happen anywhere else.

0.4%. That’s the difference that won a $1.1M deal in our AEs second month.


Liz was selling to the CEO of a company that was recording 1M meeting hours per month.
They’d built their own meeting recording infrastructure.

99.5% uptime.
We offered 99.9%.

Leadership wasn’t impressed.
A 0.4% difference doesn’t justify ripping out something you already built.

This is where most ramping AEs lose the room.

Liz didn’t argue the number.
She translated it.

“At your scale, that’s 4,000 meetings per month that don’t get recorded. What happens, when that happens?”

That’s when it connected:
- Customer conversations that can’t be recovered.
- Customers lose trust fast when the product “misses” what they said.
- Engineering gets dragged into firefighting instead of shipping product differentiators.

They signed two weeks later.

This was Liz’s second month…

At enterprise scale, decimals aren’t details.
They’re thousands of conversations that either get captured or disappear.

p.s. we’re hiring for Strategic Account Executives. JD in the comments

One of the scariest parts of leading a $250M company is taking visible bets. Last year, I made one knowing short-term revenue would drop.

Not eventually.
Not hypothetically.

Immediately.

This wasn’t a small optimization or a local decision
It affected the entire company.

When a decision changes pricing, packaging, or how customers buy, there’s no hiding it.
Everyone feels it at once.

That’s what makes these bets hard.

You worry about the obvious things:
- What if it fails?
- What if people lose confidence?
- What if teams feel like they’re paying the price for a bad call?

And you worry abut the less obvious ones:
- What signal does this send internally?
- What precedent does it set?
- What happens if the numbers dip before the upside shows up?

In our case, the tradeoff was clear.

Short-term revenue would go down.
Long-term velocity, leverage, and customer access would go up.

There’s no committee for bets like this.
No vote.
No A/B test.

At a certain level, leadership is about owning the blast radius.

That’s the part of leadership most people never see.

p.s. we closed the year with six seven-figure deals

The exec team you hire at $3M ARR will be wrong at $20M ARR.


At $3M ARR, most startups are still figuring out their motion.
What works feels fragile.
What doesn’t work isn’t obvious yet.

And yet, that’s often when the pressure to hire for execs shows up.

The mistake isn’t the leaders you hire.
It’s assuming the shape of the company is stable enough to lock in leadership roles.

In our case, I waited to $6M ARR to hire our first VP.

If we had hired earlier, we would’ve optimized for a version of the company that didn’t last.
I wouldn’t have understood the work well enough to know what “good” actually looked like.

By then, the motion had settled.
The tradeoffs were clear.

Hiring an exec team early feels like progress.
But it also freezes assumptions and those are expensive to unwind later.

Building the Zendesk Killer.

Pylon is the next-generation support platform, optimized for B2B companies.

Backed by a16z, General Catalyst, and Y Combinator.

Marty Kausas

We just launched a new onboarding program. 

It's cohort-style and introduces new hires to different parts of the company.

Some of the week 1 sessions...

- Pylon history (hosted by me)
- Product vision & team overview (Advith and Robert)
- How we demo Pylon (Mana, our head of sales)
- How we use Pylon for support (Jon, head of support)
- How we use Pylon for success (Dan, head of CS)
- Lunch with the founders

We're growing so quickly that team-level onboarding isn't enough anymore. 

So Advith came up with a way for everyone to meet leadership and cross-train on the product from day 1.

—

P.S. We're hiring!

We have 25+ roles open across all functions: 

- Software engineering
- Product design 
- Performance marketing
- Legal ops
- Sales
- Product support
- Customer success
- and a lot more

Apply here: https://lnkd.in/gmcYnSnd


📸 : Dan demo-ing how we use Pylon for CS

CEO @ Pylon | Building the Zendesk killer

B2B support teams used to have a "stack". 

Zendesk → ticketing
Fin → AI agent 
Slack, email → customer messaging
HubSpot → account data
Gainsight → health scoring
TheLoops → sentiment analysis
Assembled → workforce management
MaestroQA → QA
Qualtrics → CSAT 
Asana → project or task tracking
HelpDocs → knowledge base
BI Tools → analytics and reporting

But teams are increasingly getting rid of this setup. Manually connecting data and workflows across those platforms is a bottleneck. 

Instead of a stack, teams should use one system...

A customer center.

You still have different systems to talk to customers (Slack, email) or capture account data (Snowflake, CRM).

But they automatically connect to a single core platform where you...

→ Manage tickets
→ Respond to omnichannel conversations
→ Configure AI agents
→ Plan team staffing
→ QA tickets
→ Send and track surveys
→ Report on support data
→ Create account overviews
→ Update knowledge base
→ Collaborate with other post-sales teams
→ Track customer-facing tasks
→ Monitor health scores and retention
→ See churn risk and upsell potential

...and run all your customer-facing operations. 

This is exactly what we've been building for the last year. 

Pylon started as a way for B2B teams to manage support interactions, but we saw teams struggling to connect the rest of their operations as they scaled.

That's why we built tools like Account Intelligence, Product Intelligence, AI issue QA, and more. All in the same platform. 

We're becoming the central place for customer data, workflows, and intelligence. 

B2B support = Pylon ✨️

We made a post about Pylon replacing support stacks.

Within hours multiple companies adopted the format for their own products/markets. 

It's crazy how fast things spread.

Looking back YC was one of the simplest times for us.

All we had to do was sell, build, and think.  

After that we ran into way more challenges including...

𝟭/ 𝗛𝗶𝗿𝗶𝗻𝗴 𝗴𝗼𝗼𝗱 𝗽𝗲𝗼𝗽𝗹𝗲. For a long time we only recruited in-network, and even then it wasn't always easy to find people who fit our values. 

𝟮/ 𝗘𝘅𝗽𝗮𝗻𝗱𝗶𝗻𝗴 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗮𝗻𝗱 𝗺𝗮𝗿𝗸𝗲𝘁. Our initial product would not turn us into a $1B ARR company. We had to find the true market we could expand into and build a compound startup.  

𝟯/ 𝗟𝗲𝗮𝗱𝗶𝗻𝗴 𝘀𝗮𝗹𝗲𝘀 𝗮𝗻𝗱 𝗺𝗮𝗿𝗸𝗲𝘁𝗶𝗻𝗴. Advith, Robert, and I all came from engineering backgrounds. We figured out GTM entirely from scratch by starting from first principles. 

𝟰/ 𝗘𝘃𝗼𝗹𝘃𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗜𝗖 𝘄𝗼𝗿𝗸. As the company and product grows, it just isn't possible for founders to touch every project. We're moving from ICs → managers → leaders. 

𝟱/ 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗽𝗿𝗼𝗰𝗲𝘀𝘀 𝗮𝗻𝗱 𝘁𝗲𝗮𝗺 𝗲𝗻𝗮𝗯𝗹𝗲𝗺𝗲𝗻𝘁. At 30 people we had no recurring meetings and everyone could collaborate ad hoc. Now that we're much bigger we need clear structure for teams to work together. 


We have to keep increasing momentum on all fronts (team, product, GTM) to build a generational company. 

Onwards 🚀 

—

Had a great time sharing these reflections along with Selin Kocalar from Delve and Sam Stallings from Artisan last month.

Thank you to all the founders who came out to chat!

📸 : Speaking at our YC alumni meetup

INTRODUCING AI REPORTS [BIG]

Support leaders don't have time to read every ticket.

Now, AI summarizes daily/weekly/monthly customer interactions for you. 

You can make reports for whatever you want...

"Provide a high-level summary of support activity for the week."

"Which types of tickets take us the longest to reply to?"

"Give me an executive summary of customer health for the month. Include key wins, challenges, and areas of concern."

"List any tickets that weren't closed properly this week."

"What are common issues that keep coming up across customers?"

"Summarize any notable escalations or high-priority issues that required special attention."

"Highlight top moments where customers complimented our team."

"List the top 5 most common issue categories with a brief description of trends or patterns observed."

...and they automatically get sent to you in Slack. 

AI reports synthesize information across call recordings, support tickets, CRM data, and any other customer data you have in Pylon.  

Internally, we're even using them across non-support teams (sales, marketing, product) to get customer insights. 

This is a huge release and we can't wait to see what teams build.


—

Shoutout to Ajay Krishnamurthy and Ben S. who absolutely crushed this project.

Link to docs in the comments 👇️

We just launched something big. 

Find out what: https://lnkd.in/gv3evGSG

Happening tomorrow!

Join us at 10 AM to learn how to scale Slack support.

You'll get to chat with Advith and Norberto Lopes (VP Engineering @ incident.io).

Register here:
https://lnkd.in/gMSVxCMK

taplio

The Battle

Amanda Zhu

Marty Kausas

Followers

Posts

Engagements

Likes

Performance ScoreAmanda Zhu's LinkedIn Performance

Performance ScoreMarty Kausas's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Amanda Zhu

Marty Kausas

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Amanda Zhu's LinkedIn Performance

Performance ScoreMarty Kausas's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAmanda Zhu's LinkedIn Performance