Sellestial

AI-driven sales assistant that delivers accurate and autonomous CRM updating. That’s because unlike traditional CRM software, we offer unparalleled reliability and context-aware automation.

Co-Founder, Chief Revenue Officer

Meetings.bio

Commercial Advisor and Co-Founder

Meetings.bio is a life-science-oriented lead generation service provider. We prospect, engage, and book sales meetings with relevant decision-makers in life science industries, so you can focus on closing deals.

It's completely risk-free: no retainers, no hidden fees. You only pay for meetings that actually happen.

Making life science decision-makers excited about a product or service can be challenging. But we cracked it—we booked over 2,700 sales meetings for 70+ happy life science vendors just in the last 3 years.

I am happy to show you how we’ve done it—simply book a strategy call here - https://calendly.com/meetings_bio/30min

Chief Commercial Officer and Co-Founder

RevOps Co-op

The RevOps Co-Op is a community built to serve its members through on-demand expertise, best practice content, educational opportunities, and networking events. All go-to-market operations professionals are welcome - from marketing to customer success and everything in between. Join us and grow as a RevOps Pro

Member

RevGenius

RevGenius is a group of revenue-generating sales and marketing professionals brought together to learn, share, support, and grow with each other.

HYPCCCYCL

Genialis

Genialis is a computational precision medicines company unravelling complex biology to find new ways to address disease. ResponderID™, Genialis’ clinical biomarker discovery platform, defines, models, and validates actionable biomarkers and optimally positions novel drugs to accelerate translational research and clinical development. Genialis is trusted by biopharma and big pharma alike, and together, we bring precision to medicine.

Senior Account Executive and Co-Founder

AD Mass Ljubljana

Pole Vault Coach

TechPeaks - The People Accelerator

Participant

Študentska organizacija Univerze v Ljubljani (Student organisation of University of Ljubljana)

Member of the Student Assembly

President of the Student Organization of Faculty of Pharmacy

Društvo študentov farmacije Slovenije

Event organization, Finances, Project management

Event organizer

How Founder-led sales can be done right? • Founder at Sellestial • 3rd-time Founder • Don't be a stranger, follow me and reach out

We are all guilty of sending boring follow-ups to our potential customers. 

Even with all the data we gather on prospects, it seems tough to personalize and truly contextualize our communication. We often default to generic messages that might not hit the mark.

We’re working on something that could help craft more personalized, relevant communications at scale, whether it's for hot deals and leads, webinar attendees, form submissions, sign-ups, or any other inbound leads, within HubSpot. 

Interested in a deeper dive into Sellestial? Let's chat over a virtual coffee - my treat! Feel free to reach out directly here on LinkedIn or pick a slot here: https://meetings-eu1.hubspot.com/klemen-hrovat

🤖 Meet our new team member: The Contact Cleaner

We're excited to introduce an AI agent that's already making a huge impact on data cleanliness in many HubSpot portals.

What does it do?

The Contact Cleaner works 24/7 in your HubSpot portal,
automatically identifying and fixing data quality issues that used to take hours. 

It's like having a meticulous data analyst who never sleeps.

And a team member on whom you can rely that you will never send a marketing email like this.
"Hey KLEMEN,"
"Hey Klemen Hrovat, PhD,"
"Hey Klemn"
"Hey klemenh@sellestial.com"

The capabilities are impressive:
✅ Fixes typos and formatting inconsistencies across thousands of contacts
✅ Validates phone numbers and postal codes using real-time verification
✅ Removes role-based emails (goodbye, info@ and sales@ addresses)
✅ Standardizes names, job titles, and locations
✅ Flags suspicious entries for human review

But here's what makes it special:
It's smart enough to know when NOT to make changes.
The agent uses confidence scoring and a conservative approach. 
If it's not certain about a correction, it flags it for our team to review rather than guessing.
It respects data integrity above all else.

The best part?

RevOps and Marketing teams now spend little to no time fighting bad data before sending the new campaign. 

Curious about cleaning up your HubSpot portal? Drop a comment or DM me, happy to share what we've learned and how we are helping others.

#HubSpot #DataQuality #SalesOps #MarketingOps #DataHygiene

Klemen Hrovat

Why should you not treat a HubSpot contact as a person?

I'm explaining this to HubSpot admins, RevOps experts, and HubSpot solution partners daily.

People change jobs all the time.

You shouldn’t lose visibility, or break reporting, just because someone changed their job.

Our rule is simple:
- Create a new contact for each new employment.
- Keep the old record as the historical truth for the prior company.

This keeps:
- Activity history intact and relevant to the company where it happened.
- Lifecycle stage, attribution, and ICP fit correctly for the new employer.
- Work email/phone semantics honest (no more bounces or “ghost” consent).

In this article, I explained what we advocate, why, and how we suggest doing it in HubSpot.

People change jobs all the time. You shouldn’t lose visibility, or wreck reporting, just because someone changed their job. Our rule is simple: Create

Open article: Why should you not treat a HubSpot contact as a person? by Klemen Hrovat • 3 min read

Why should you not treat a HubSpot contact as a person?

Prejšnji teden sem imel čast govoriti na konferenci.
Naslov moje predstavitve je bil
“How an Alien got us into the HubSpot community”

Delil sem našo zgodbo o tem, kako smo si zgradili prepoznavnost v HubSpot skupnosti.

Prvič sem predstavil celotno zgodbo od začetka do konca.

Bil sem kar malo živčen, kako jo bo občinstvo sprejelo.
In dovolj pogumen (nor), da sem stopil na oder oblečen v vesoljca.

Občutek pred nastopom je bil enak kot prvih 15 minut na INBOUND24.

In tako kot na INBOUND-u sem se hitro sprostil.
Res sem užival na odru.
In glede na povratne informacije po predstavitvi,
je ljudem bila všeč – ne samo predstavitev, ampak predvsem pristop.

Verjetno ni bila všeč vsem, ampak blagovnih znamk ne gradimo za vse, ane?
Če želiš ugajati vsem, če ne podžgeš vsaj dela svojega občinstva, igraš preveč na varno.

Eno glavnih sporočil, ki sem jih želil predati, je bil kako pomembni so odnosi.
Vsi smo ljudje, zaupamo drug drugemu, v dobi AI bomo še bolj iskali človeške odnose, zaupali bomo svojim vrstnikom, zaupali bomo LJUDEM.

P. S. Če te zanima predstavitev, mi pošlji DM — z veseljem pošljem prezentacijo.

Sales reps don’t wake up in the morning and say, “I can’t wait to enter data into HubSpot!”

Data hygiene is NOT their job.
It’s RevOps’ or HubSpot Admin's job to design systems that:
✅ Capture the right data automatically
✅ Minimize manual entry
✅ Audit for duplicates and bad records
✅ Standardize and clean important properties

Blaming reps for bad data is like blaming drivers for potholes. Wrong target.

Agree or disagree?

I wish that were me ...
I will be OOO for most of next week
So I added that picture to my OOO email

Do you think it's ok or too witty for the B2B space?

“I think HubSpot is broken.”

I’ve heard this from many sales teams.

But HubSpot isn’t broken.
The data is.

👉 Lifecycle stages that are not properly updated.
👉 Contact properties that reps don’t trust.
👉 Companies with 5 versions of the same entity.
👉 Contacts sitting in HubSpot for 10 years, still at the same company.

Before you throw money at new tools, ask yourself ➡️ Do I trust my CRM data enough to bet next quarter’s revenue on it?

If not, the problem isn’t HubSpot—it’s data hygiene.

What is your reaction to "My HubSpot is broken"?

Merge or don't merge?
This is often the question in deduplication.
But it turns out it’s not an easy decision.

One thing we learned in the past month
Is that deduplication is the end goal
But you need to do the steps before.

Clean
Enrich
Merge

These 3 steps should be taken if you want to
deduplicate your HubSpot portal
In the optimal way.
And don’t make any mistakes.

Why?

Wrong enrichment
Overwritten correct values
Badyl configured integrations

These are just a few cases that we saw
causing the mess in portals,
which prevents you from doing easy dedup.

If you take an naive approach,
Define a few simple rules
And merge records

I can almost guarantee
you will be making mistakes
and merging records you shouldn’t.

Are you aware of data quality in your HubSpot
before you set the simple rules to merge “duplicates”?

Did you ever merge some records that you realized you shouldn’t?
And wasted a lot of time unmerging
Because there is no easy unmerge function in HubSpot?

Before you think about deduplication next time, let me know, happy to chat and share some of the steps we suggest to other HubSpot users around deduplication.

Why HubSpot data cleaning matters NOW.

Not last year, not next year

Pre-AI
Extra data was cheap.
The risk of losing something wasn't worth the cleanup effort.

Post-AI
Data is expensive to enrich.
And the rewards for great information are exponentially higher.

This fundamental economic shift changes everything.

You can't afford to run AI and AI-enrichment on dirty data anymore. You'll either:
→ Overspend your budget by 2-3X
→ Or spread your compute so thin you won't see AI's value

The technology finally arrived. 

It's cheap enough and good enough—FINALLY.

Companies that clean their data in 2025 will be 12-24 months ahead.

Companies that wait will watch their successors do it instead.

The question isn't "should we clean our data?"

The question is: "Do you want to be the person who owns this challenge?"

First Brian Halligan on Hot Ones with Sean Evans

Now Luka Dončić on Hot Ones with Sean Evans

First, a picture with Brian at INBOUND

....

P.s I'm proud to say on Zoom calls, when I'm asked where I'm calling from, that I'm from Slovenia, just like Luka 🇸🇮

Nejc and I will walk you through the workshop for RevOps, marketing and sales experts. With the workshop, we want to help you master AI fundamentals and build practical skills to transform your operations.

Nejc (Nates) Škoberne and 1 other attendee

Practical AI For RevOps Workshop Overview

Your HubSpot is only as smart as its dirt
If your data’s messy, automation will break and reports mislead.
Fix these five hygiene challenges first:

1) Deduplication (Contacts & Companies)

Set merge rules:
How do you decide if something should be merged or just a parent/child relationship?
Which record is primary - keep the most recent record’s property values or the oldest record, so you preserve the creation date?

Lock a “Source of Truth” order: Human edit > CRM form > Integration > Import.
Or a different version of that, if you trust other sources more. 

Figure out what’s causing new duplicates to minimize the creation and define processes to dedupe as soon as possible. 

2) Unclean First & Last Names

What to check today
Names in ALL CAPS, all lowercase, with emojis, or stuffed titles (e.g., “John (CEO) 🚀”).
First name = Company or Initial only (“J”).

Configure the system
Clean Title Case
Strip emojis, trailing spaces, honorifics (“Mr.” “Ms.”), and role junk (“CEO”).
Separate Name from Job Title—never store both in First or Last Name.
Remove Middle name from First Name


3) Outdated Contacts

What to check
No activity in 180+ days AND not in an active segment.
Hard bounce, unsubscribed, or deactivated owner.
Contact created 5 years ago


4) Missing Job Title or LinkedIn Profile

Check
Contacts without Job Title, generic (“Staff”), or nonsense ones.
Contact LinkedIn URL missing

Quick wins
Enrich contacts with LinkedIn URL so you have a link to check for potential promotions and job changes
Validate LinkedIn URL property - must contain linkedin.com/in/ or leave blank.

5) Companies with Wrong LinkedIn URL & Bad Firmographics

Validate
Company LinkedIn URLs that aren’t linkedin.com/company/
Domains that are not the same to the company domain on their LinkedIn page. If that is not the same, it indicates some challenges with some of the enrichment tools. I have seen some pretty weird enrichments happened because of native HubSpot enrichment which takes primary domain and runs with it, no questions asked. 


Minimal data hygiene
👉 Clean contacts: First/Last name, Job Title, LinkedIn URL
👉 Contact and company Duplicates
👉 Clean company names

What's your biggest struggle with data?

AI here, AI there, AI everywhere
Does your head feel like that - 🤯 ?
A foundational workshop about AI for RevOps might save your head from blowing away. 

What you'll learn:

How LLMs actually work – Clear up misconceptions and understand the fundamentals that matter for RevOps

Which models to use when – Navigate the vendor landscape and make smart cost vs. performance decisions

Four core LLM applications – Generate, extract, classify, and execute tasks with practical exercises for each

Prevent hallucinations and ensure reliability – Guardrailing techniques using XML tags, JSON schemas, and structured outputs

Multiple ways to invoke LLMs – From Google Sheets scripts to Model Context Protocol (MCP) implementations

When to use AI agents (and when not to) – Agentic workflows, memory management, and tool usage strategies

Stay compliant and secure – Data privacy and compliance essentials for handling customer information

Real RevOps use cases – Apply everything you've learned to actual processes you face daily (including your submitted use cases)

Ready to build a strong AI foundation? 
Sign up today, don't let others take your seat (seats are limited for every workshop so we can provide the learning experience you deserve)

OpenAI has an Academy.

No matter how skilled you already are, I'm sure we can all learn something.

A few interesting lessons, with short videos available at the links below. 

During our workshops, Practical AI For RevOps: The Foundation, we'll do a more thorough overview of these topics and dig deeper.

Introduction to GPTs
https://lnkd.in/dY9MxizN

Introduction to Prompt Engineering
https://lnkd.in/dG4Uw_Tr

Advanced Prompt Engineering
https://lnkd.in/dhZRxDAh

ChatGPT 101: A Guide to Your AI Superassistant
https://lnkd.in/dUjkuees

ChatGPT & Reasoning
https://lnkd.in/difN2Syx

Multimodality Explained
https://lnkd.in/db2BDJaD

OpenAI, LLMs & ChatGPT
https://lnkd.in/dcNcbQ64

ChatGPT for Data Analysis
https://lnkd.in/dAiETafB


Oh, and this is our workshop I mentioned above 😊 
https://lnkd.in/dw7MW3GU

There is no such thing as perfectly clean CRM data

And it’s fine. Here’s why

Yes, accurate data is important, but chasing 100% purity is often a time sink with diminishing returns.

Consider that one B2B sales team had a 40% duplicate rate in their HubSpot data, causing reps to spend 6 hours a week on manual cleanup instead of calling customers.
(Imagine how many deals could close in those 6 hours!) 

It’s no surprise that studies find sales professionals spend a lot of their time on non-selling tasks like data entry.

Tools are helping with some of that, but we are still far from clean CRMs.

What did we do wrong to change ABC -  “always be closing” to ABC -  “always be cleaning”?
- Marketing emails bounce
- “Hi KLEMEN,” still happens
- Contacts are in the same position in your portal for 10 years. But do you believe they are still with that company?
- Champion customers move position without you noticing that
- Some contacts in your HubSpot get 10 emails in a week because they are enrolled in different workflows

Forward-thinking RevOps experts recognize that not all data in HubSpot needs to be perfectly clean.

Different companies and contacts have different potential value. So treat data quality accordingly.

For someone who signed up for your newsletter 3 years ago but never did any other action, according to your data in HubSpot, you don’t want to monitor monthly if they are still with that company. 

But a raving fan at one of your customers, who moved to another ICP matching company a week ago - you do want to catch that insight, right?

Feeling overwhelmed by AI possibilities in RevOps?

You're not alone.

That's Nejc (Nates) Škoberne designed this intensive 8-hour workshop.

We're covering everything from LLM fundamentals to agentic workflows, from reliability techniques to real RevOps use cases. 

No fluff—just practical, hands-on learning you can apply immediately.

I wrote a detailed breakdown of all 8 modules we'll cover in the article below.


Fair warning
- It's intense
- You'll be exhausted
- But you'll have the foundation to confidently use AI in your operations.

#RevOps #AI #GTM #SalesOps

If you're a RevOps, GTM, sales, or marketing leader who feels overwhelmed by AI possibilities, you're not alone. After countless conversations with pr

Open article: Practical AI for RevOps: A Foundation-Building Workshop by Klemen Hrovat • 3 min read

Practical AI for RevOps: A Foundation-Building Workshop

That's one of the most important questions around data hygiene.
I've stopped trying to convince skeptics about data cleaning.
Instead, I ask one question:

"Do you believe there's gold in the dirt of your CRM?"

If they say yes → We can talk.

If they say no → I move on.

Here's why:

A company has 30,000 contacts in their CRM:
5,000 are duplicates
8,000 are maybe duplicates
15,000 are low-grade and not interesting
10,000 are worth investing AI enrichment on

That 10,000? That's the gold.

But right now, you're paying to enrich ALL 30,000.

Your nurture campaigns are hitting dirt, not gold.

Your sales team sees 5 versions of the same company.

Your enrichment budget is 3X what it should be.

If you believe there's gold in there, let's separate it from the dirt.

If you don't believe it, that's okay too. 

But no amount of convincing will change your mind; only pain will.

HubSpot, any particular reason why it's not possible to change the name of a sequence via API?

🎙️ New to streaming or looking to level up? Check out StreamYard and get $10 discount! 😍 https://lnkd.in/dAaAJzce

One of our customers tried to fix formatting issues with Data Hub, but they faced challenges as the system didn't do the optimal job, so it helps them do it natively and not do it in a spreadsheet manually. 

So they asked us to help them, and this is how the system works to make a basic first and last name cleaning, right in your portal. 

No more
Hi KLEMEN,
Hi JOhn,
Hi Sales,
Hi 👽 Klemen

Do you use Hi {first name} in your marketing emails in HubSpot?

taplio

Klemen Hrovat 👽's Linkedin Analytics

Klemen Hrovat

Performance ScoreKlemen Hrovat's LinkedIn Performance

What is Klemen talking about?

Who is engaging with Klemen

Klemen Hrovat's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Klemen Hrovat 👽's Linkedin Analytics

Klemen Hrovat

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);}Klemen Hrovat's LinkedIn Performance

What is Klemen talking about?

Who is engaging with Klemen

Klemen Hrovat's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Klemen Hrovat 👽's Linkedin Analytics

Performance ScoreKlemen Hrovat's LinkedIn Performance

Klemen Hrovat's Best Posts (last 30 days)