Tom Hunt

$3.6m ARR. $0 raised. I have no idea what I'm doing.

Jan 2014.

I'm walking home from a depressing New Year's trip to Oslo with an ex. girlfriend.

It's raining in East London.

I made a decision:

I would replace my management consulting salary (£42,000) by the end of the year.

I had no idea from where, but that was the goal...

I just about hit it and left the corporate world.

And I haven't been back since.

Fast forward a decade and I've:

- Tried and failed with 17 businesses
- Invested in 11
- Learnt a LOT

The one that is finally working is called Fame:

- We have a weird culture (in a good way)
- We start and grow podcasts for B2B businesses
- We are fully remote and have 75 full-time team members

I live for our awesome team at Fame, building businesses and great marketing.

I post about them here 🤗

Founder & CEO

We start and grow the world's most profitable B2B podcasts.

Clients include:

👉 IDEO
👉 Workday
👉 HighRadius
👉 People.ai
👉 Zendesk
👉 Bevy
👉 Circle
👉 Suzy
👉 Mazars
👉 LeanData
👉 CXL
👉 Ramp
👉 Alibaba
👉 Signify
👉 eToro

On my last day of work, the Sr. director of marketing told me that I "was a pain in the ass"

Why? I did things differently. I challenged the norm. I found my ways to do things.

And he really appreciated it.


I've worked with all kinds of companies and even built my own-

Early-stage bootstrapped startups, well-funded startups,  Unicorns, Enterprise companies


➖


I started my B2B SaaS inbound marketing agency in 2021.

It was... strange.

Getting clients was too easy.

At first, I was proud.

Then as I fell in love with the industry, I got concerned.

To put it lightly... we hardly have any good B2B SaaS marketers.

My team and I will change this by 2030.

You'll see B2C marketers learning from B2B SaaS much more commonly.

Watch us do it. (Or join the cause?)


➖


Spear Growth helps B2B SaaS companies with Ads & SEO.


📈For Ads,
we go deep into every aspect that can drive pipeline & SQLs.

▪️Tested 12+ paid channels
▪️Optimize the entire funnel
▪️Built the most advanced Demand Gen framework in the world
(and more)


📈For SEO,
we've built the most sustainable SEO service model in the world.

▪️Focus on SQLs, not vanity metrics
▪️No retainers - Only 3 month projects
▪️No BS audits, only actionable roadmaps.


Let's talk B2B growth 💬

https://speargrowth.com/30-min-call


➖


Here is me bragging about myself:

🎓 I've trained 2500+ marketers so I understand my stuff - but more importantly, I know how to communicate it.

🚀 I led SEO & analytics for a Unicorn startup.

📖 I've contributed to 2 internationally best-selling marketing books

💻 I've helped amazing B2B SaaS companies: Seed, Series A-E, Unicorns.

➖

My agency: speargrowth.com

My email: ishaan [@] speargrowth [dot] com

Ishaan Shakunt

My birthday gift was my wife measuring the macros of my lunch.

And the weird part? That felt completely normal to me.

Looking back at my birthday on Thursday, it wasn’t a highlight reel. It was just… a really good day.

Started quietly. Stayed in the night before. Home date night. A Korean series. One of the best chocolate cakes I’ve ever had. Wine I actually liked. (Neither of us enjoy drinking, so this felt noteworthy.)

Woke up on time. Which shouldn’t be a win, but fixing my sleep schedule might be the most underrated achievement of the year.

My brother Aarav came over. We did a lower body workout. I hated it. He hated it. Watching him suffer made it better.

My wife cooked lunch. As a birthday gift, she measured the macros. Because that’s what I asked for.

50 calories. 4g protein. Per hara bhara kebab. Elite.

Work was one of those days that quietly compounds.

Two solid meetings with managers. Big things in motion.

Harsh figured out how to create product launch videos using Claude Code. Had one of those “wait… that’s possible?” moments.

Finished an important custom proposal for an enterprise client. An investor call got pushed by a day. Theek hai.

The Chosenly & Spear Growth team sent me a cake just to roast me. That’s the picture.

Ended work with a post on Exit Five.

Lots of calls from family, friends, relatives.

Ended the day with a date and long conversations with my wife.

Also: I wore something other than my Uniqlo t‑shirts. Once‑a‑year character development.

No lesson here. No takeaway.

Just grateful for: People who care. Work that’s moving. Systems that are slowly getting better.

Back to building.

– Ishaan "remembers being curious about how founders went about their day. lol" Shakunt

AI Search Optimisation for B2B SaaS | 2x Founder @ Chosenly.com & Speargrowth.com | DM me for an SEO + GEO audit

"I don't have time to go to the gym and exercise"


Gets dumbbells at home.

"I don't have time to exercise"

AI does my work while I exercise.

"I.....Sigh..."

- Ishaan "professionally out of excuses" Shakunt

No one wants to own listicle placements.
But everyone wants the outcome.

- “Mentioned in the list.”
- “Featured on the page.”
- “Included in the roundup.”

Listicle placements have 3 types of value:

1/ GEO value
You get added to pages that LLMs already cite.

2/ SEO value
You earn backlinks.

3/ Referral traffic value
Many of these pages already rank for keywords we’re targeting. Getting a link from them drives traffic.

And yet… no one wants to run the machine. So we built Automated Listicle Placements inside Chosenly.com.

Just press one button.
Chosenly does the outreach for you, end-to-end.

It shows you a log of what it’s doing, and only asks you for approvals like:
- Should we ask for X placement in Y list?
- Should we pay $200 to get placed in Z? (We negotiated it down from $400.)

This system is built off the last 12 months of us doing placements manually for clients, learning what works, what gets ignored, what gets a reply, what turns into a yes.

It’s not a shiny feature.
It’s the boring job no one wants to own, productized.

If you’ve ever said “we should do more listicles” and then watched it die in a Notion doc… you’ll get why we built this.

P.S. This post is part of our Chosenly 3.0 launch.
Every B2B SaaS company that engages with this post gets our Reddit automation feature for free.
Also, 10 companies to engage get a 1-month Chosenly subscription.

– Ishaan “outreach feature new UI is my new dopamine loop” Shakunt

🎉Giveaway 🎉
Chosenly now has a Reddit module, and it’s completely free (for you).

Reddit now ranks high on Google search, and influences AI search. So a lot of teams want to work on it.

But it’s hard!

Either too many threads get started in your industry. So you can’t keep up with them.

Or, too few threads get started so you miss them all the time.

So, we automated this for you.

Chosenly.com finds relevant submissions and comment threads automatically. 
Then it clusters them by intent (what the OP is asking), so you know which ones to reply to.

Every B2B company engaging with any of our launch posts (including this one) gets this feature for free.

Also, 10 companies get a 1-month Chosenly subscription.


So, comment “REDDIT” and I’ll send you your login credentials.

– Ishaan “Feeling like Santa Claus in January” Shakunt


Psst: In the coming weeks, Chosenly will also draft the replies and post them, on-brand, while respecting Reddit culture.

When you ask ChatGPT a question, it doesn’t just search once. It runs a bunch of searches in the background, then answers you.

That’s query fan-out.

Most teams either ignore it because it looks like a messy pile of searches or take a lazy takeaway like “add 2026 to the title”.

Query fan-out is full of hints about what AI search is trying to find.

It does 3 things that matter. It tells you:

1/ How it chooses software 
AI search literally adds criteria into the searches.

Example: When we searched for "sales tax software", it added criteria like pricing and features into the searches.

Action: Address those criteria in your positioning, but also in your competitor alternatives pages and listicles.

2/ The specific features it believes matter
Sometimes it gets more specific than criteria. It starts searching for very particular features.

Example: For the keyword “skill management software”, it fans out features like skill tracking, development, career planning, and gap analysis.

Action: This is not a nice-to-have checklist. If your product does these things, your page should say it clearly.

3/ Who you are competing with
Sometimes the model starts adding company names into the search itself.

Example: For “AWS cost optimization tools”, it literally starts naming tools like CloudHealth, Cloudability, Spot in the searches.

This means AI search has already formed a shortlist before it reads your page.

Action: If you are not on that shortlist, you need to earn your way in by positioning against the names it already trusts.

If you are on it, congrats, now you know exactly who you are being compared to.

That is why we are finally making query fan-out usable inside Chosenly.com.

Chosenly will show Query fan-out by default.
And over the next month, the insights from it will start showing up where you actually need them: in positioning and in content strategy, already sliced and grouped so you can act on it.

Giveaway: This post is part of our Chosenly 3.0 launch. Every B2B SaaS company that engages with this post gets our Reddit automation feature for free.
Also, 10 companies to engage get a 1-month Chosenly subscription.

– Ishaan “Re: Making query fan out usable, not just available” Shakunt

ChatGPT just now announced ads.
Here's what you need to know(with sources):

Source 1: Their announcement

1. They're focused on B2C ads. (makes sense that's a majority of their user base)
2. The answers won't get influenced by ads at all (yet)
3. Just like we don't have a GSC for ChatGPT, advertisers won't see the exact prompts bidding on.
4. They're rolling this out in the US first, then other regions.
5. It's pretty easy to spot what's an ad vs what's an answer. (for now)
6. It's just a product placement for now, but will evolve into conversational ads later.
7. In every mock up they shared, only a single company's products were showcased. This should definitely not be the case, maybe even from launch.

Source 2: Other ad platforms' historical behaviour
8. They'll be pretty cheap early on, will get more expensive with time.
9. We'll see high CTRs early on, when they're new then users will learn to ignore them.
10. Ads will get harder to distinguish from the answer itself.

Source 3: From their older communications
11. They'll want to encourage (ecomm) companies to sell their products via their portal. They'll probably reduce the ad cost for this, but charge a transaction fee.
12. These types of ads shouldn't be available to most B2B companies. We'll probably only get access to display style ads.

From their App Dev notes:
13. ChatGPT ads targeting won't look like LinkedIn(audience), nor Google(keywords). It'll be intent-based.
14. A lot of "ads" will effectively be ChatGPT apps. Think- your data plugged into a conversation to help the user.
15. I'm hoping they'll at least let us see the ROI per "intent group" or something. Not ask us to just trust them with something like the pMax campaigns Google is pushing.

All of this is interesting.

But it’s not the part I’m most focused on.

What matters isn’t what this does *this quarter*.

It’s what this unlocks *next year*.

ChatGPT is testing new patterns.

Perplexity, Claude, and Google will follow.

Google is already testing ads inside AI Mode.

When Google fully replaces their traditional search fully with an AI‑first experience in the next 12–24 months, ads stop being a channel tweak.

They become part of the answer layer itself.

That’s when this actually changes how marketing works.

Not because there’s a new ad unit.

But because intent, trust, and narrative all converge in one place.

And teams built to run isolated channels will feel that shift first.

How are you thinking about this beyond “new inventory = new tests.”

– Ishaan "Luckily, has access to ad experiments by Spear Growth & LLM data from Chosenly" Shakunt

We launched the TLDR Podcast exactly 2 years back.


A lot of podcasts drift into life stories, vague lessons, and broad takes.
I don’t even enjoy listening to most podcasts for that reason.

I wanted the opposite. I wanted a repository of tactics anyone can steal. The kind of episodes where you finish listening and go:

“Okay. That’s sooo specific. I can implement this next week.”

Which meant I needed to:

A. Interview experts who are in the weeds not just people that are famous.
B. Follow a format where they share the little details they wouldn’t share anywhere else, not just high level strategy.

If you want a sense of how tactical it gets, here are a few examples:

1/ Natalie Marcotullio broke down how they made pricing public and how that connected to their LinkedIn motion to improve MQL to Opportunity ratio.

2/ Mason Cosby talked about his “scrappy ABM” approach including sending popcorn to people and how that turned into serious pipeline.

3/ Gal Aga went deep into the nuances of increasing ACV from $50k to $150k. He shared the actual mechanics and trade-offs.

4/ Tara Robertson turned a $40k booth into $100k+ pipeline in 3 days by making “wanted posters” of 20 target accounts and incentivizing people to bring them to the booth..

5/ Cory Snyder did one of my favourite episodes on partnerships. It’s the clearest, most tactical breakdown I’ve seen on how to activate partners early and get to revenue faster.

So yeah, we don’t publish as often as we “should.” We’re just 44 episodes in.

But if you’ve ever wanted a podcast that skips the fluff and gets into:

- What did you do, exactly? 
- What worked? 
- What didn’t?

…that’s what we’ve tried to build.

P.S. Who should I interview this year? Something who’s in the weeds, is open to sharing numbers and has something unique to share.

Want to listen to the podcast? 

Search “TLDR B2B SaaS Podcast on Apple / Spotify!


– Ishaan "Looking to speak on other podcasts this year" Shakunt

Know anyone in Bangalore who'd be available to catch up next week?

Hardik, Prakriti and me will be there at Capillary Technologies' HSR office.

We’d probably be at 2x MRR if we didn’t make this decision. I don’t regret it one bit. 

We launched Chosenly.com in March 2025. By July, we saw 2-3 AI search analytics tools launched per week.

While our competition kept adding graphs and data points, we kept deleting them.

They wanted to split the data in every way they could think of.

Which meant they got market share and revenue upfront.

Honestly, that was smart.

For us, sales conversations were harder because it felt like we had fewer features, the tool was less useful.

But we had a vision for our tool.

Every graph, table, feature needs to be “useful.”
Useful: If you cannot take action seeing the feature, delete it.

And this hurt revenue.

So why don’t I regret this?

Today, our competition is running around trying to put together recommendation engines.

We don’t need to.

They need recommendation engines because their tools are bloated messes impossible to take action on.

Their recommendation engines cannot save them from this.

We have a 9 month head start on building features and adding data points that actually help our users take action.

Every dashboard has been fine tuned based on the way our users used the tool.

There are two ways to make software actionable:

1. Build something not actionable, then try to make it actionable later
2. Build something actionable from the start (it takes longer)

The next 9–12 months will tell us which bet was right.

– Ishaan “refuses to ship nonsense” Shakunt

Writing good content is rarely the hard part.
The hard part is knowing what the hell to write.

In most teams, product knowledge is scattered across docs, demos, sales calls, and half-answered Slack threads.

By the time it reaches the person writing?
It’s been translated 5 times… and it shows in their writing.

Lashay Lewis launched BOFU.ai recently, and it immediately caught my attention.

Unlike all writing tools in the market, it tries to understand the product before writing anything.

A few design choices here are worth calling out for:

1) Going all in to web-based SaaS.

2) They use a QA-style backend to understand the product by interacting with it, which is a much smarter foundation for content than human explanations.

3) Screenshots as part of the explanation

4) The data sources have been picked really smartly.

5) Pulling from competitor sites and reviews mirrors how buyers compare, not how companies describe themselves.

The part I liked the most about it is that it creates a plan first and shows it to you.

Most bad writing is actually a planning failure.

That said, I have some questions too:

1) If a system has deep access to a product, how do you stop it from over-indexing on features instead of real-world flows?

2) Browser interaction explains screens well but buying decisions are shaped by intent, trade-offs, and edge cases. How does the system decide which journeys actually matter?

3) The objections and “why this feature matters” really live inside sales calls. Without that layer, content can be accurate and still miss what convinces buyers.

Curious how you’re thinking about these tradeoffs as BOFU.ai evolves, Lashay.

– Ishaan "has opinions about chairs too" Shakunt

Channel-based marketing let us get away with siloed teams for years. AI search will not.

Because your entire web presence is now one surface area.

PR, SEO, content, socials, reviews, listicles. All of it.

Now you’ll say the solution is simple: break silos

I challenge you. Try doing it.
It is a nightmare.

Put your PR team and SEO team in a room and ask them to create one strategy together.
They will never be able to coordinate on ONE narrative for AI search.

Why it's impossible:
- They don't understand each other's channels
- Their KPIs are different
- They report to different people
- How they measure success is completely different
- They have zero shared context

Everyone sits together, understands every channel, and stays aligned. 
➡ ️ Never happens.

OR

One superstar who understands everything aligns everyone. 
➡ ️ Work until they quit, get sick, or become a bottleneck.

The two utopias just don’t work.

So here’s the only realistic model I see:

You keep channel teams.
But you pull creation into one central “context team”.

PR writers move into it.
SEO writers move into it.
The people actually producing content sit together, with a product marketer who has high access to subject matter experts.

Channel teams still do channel strategy and distribution.
But you get one coherent story across everything that goes on the web.

It’s annoying. It’s painful.
But at least the pain is at the strategy level, and teams learn how to work with it over time.

I don’t see any other way to solve this.

If you’re doing GEO today,
How else would you do it?

– Ishaan "Building the tech stack for this new reality” Shakunt

“We have AI search data”
Most tools saying this are LYING.

I keep seeing vendors claim they have:

- AI search volumes
- Top ChatGPT prompts
- Buyer intent data from LLM

If you ask how, they’ll say:

1. "We buy it from user panels."

They claim people sold them their ChatGPT history. 
For how much? Probably less than $5. 

Show me one CXO who’d do.

2. "We collect it via browser extensions."

Google actively removes extensions that scrape behavior without consent.

If this were happening at scale, it would already be dead.

3. "We get it from ChatGPT API usage."

This one is technically possible.

But the data mostly comes from cheap ChatGPT wrappers, image tools, hobby workflows.

Serious B2B buyers do not make buying decisions inside sketchy API resellers.

Which leaves the most common method.

They take Google search trends, extrapolate, and call it:
AI SEARCH VOLUME 

That's fine as an assumption. It's NOT fine to lie about it.

Here's the bigger problem though: 

Even if this data were real, it still wouldn’t matter.

You can’t optimize AI search like Google. There is no stable list of queries, no consistent rankings, and no single “prompt” to win.

What actually matters is the surface area of how your company shows up when buyers are reasoning and making decisions.

If you work at a tool that claims real AI search data and I’ve misunderstood your approach, I’d genuinely love to be proven wrong.

– Ishaan "making enemies on his birthday" Shakunt

Getting your hands dirty is non negotiable in 2026. Here’s why:


Marketing used to reward playbooks. If you found something that worked, you could run it for a long time without re-learning the craft.

That era is over.

Over the holidays, my LinkedIn feed was full of CMOs shipping side projects.

Then last week, I saw the most extreme version of this on a sales call.

A Head of Marketing had vibe-coded an internal version of an AI search product before speaking to us.

Just to understand what a good tool should even do.

He didn’t ask me what does Chosenly.com do?

He asked:

- Here’s my thesis
- Here’s where most tools break
- Here’s what I need baked into the workflow

He asked better questions than 90% of people I meet.

Why? Because he had context. When you have context, you will:

- Hire better people
- Evaluate tools faster
- Ask sharper questions

Those differences look small. But they compound.
Until suddenly the same strategies work for you and mysteriously fail for everyone else.

That’s when people say: “Nothing works anymore.”

Usually, that’s not true.

What’s true is this:
The marketers who will struggle in 2026 are the ones who still believe they can lead marketing without first‑hand context.

These kinda marketers are waiting. Waiting for direction. Waiting for someone to explain the new game. Waiting for a guide. Waiting for a course.

If you are leading marketing, you don’t get that luxury anymore.

If you’re confused about where to start, here’s my advice: 
Pick anything. It doesn’t matter what.

Try a new tool, vibecode something, test a new campaign, build a side project, speak with experts in the weeds, anything.

Worry about direction later.

– Ishaan "Almost done with my Reddit automation experiment” Shakunt

Pick one: ChatGPT / Claude / Gemini / Something else?

Honestly, I don’t really care about the answer.

Think about why you picked it.

For some of you, it may be about features like MCPs, Projects, Temporary chat, custom GPTs, and skills.

But for most of us, it’s something different.
“I like how it structures sentences”
“The images feel professional”
“I like how it understands my prompts”

And by the end of 2026, the feature parity won’t matter, the only thing that will remain is this “different”.

What it shows:
How opinionated it is
How much control you give the user
What you assume your ICP wants

That’s taste.

And when your taste matches your ICP’s taste…
People don’t just buy your software. They feel like it was built for them.

Chosenly.com has a lot of features our competitors don’t.

But, recently it feels like we’ve been winning more deals not because of that, but because of how the UI feels.

– Ishaan "Will make sure we keep our feature lead too 🙂” Shakunt

Most marketers still treat cold calling like it’s not their job.
And that’s a problem that needs to be fixed.

In this episode of TLDR, Evan Dunn shares how he took an outbound program from 0 to 5 meetings/day - not by just hiring more SDRs, but by applying demand gen-level experimentation to outbound.

Why does this matter for marketers?

Because the fastest path to offer clarity isn’t waiting on demo form fills.

It’s picking up the phone & talking to real buyers.

Evan breaks down:
1/ How he used cold calls to find PMF gaps and rebuild positioning
2/ Why outbound conversations beat other tools for finding your real competitors
3/ How he rewired a marketing org to own outbound (without losing attribution)
4/ The cold call stack + strategy that was used

“Founder-led sales works not because founders are great sellers, but because they’re closest to the feedback loop. Marketers need to get that close again.”

If you’re a marketer struggling to differentiate, this episode is a must-listen.

👇 Full episode in comments.

 – Ishaan “got humbled by the phone” Shakunt

taplio

The Battle

Tom Hunt

Ishaan Shakunt

Followers

Posts

Engagements

Likes

Performance ScoreIshaan Shakunt's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Tom Hunt

Ishaan Shakunt

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);}Ishaan Shakunt's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreIshaan Shakunt's LinkedIn Performance