Founder - https://teapaw.org/t.co/yQW7vVNn7B || Ex YC Founder - W20 || GTM, Growth & Influencer Marketing expert || Prev - Emergent | Runable | Tars | Unacademy
Hopping between SF & BLR
Joined October 2022
- Tweets2K
- Following430
- Followers551
- Likes2.4K
Gaurav Bisen retweeted
Today, we’re introducing Ideogram's first vector model.
Generate clean, editable SVGs with smooth gradients and sharp typography from a text prompt, or turn any image into a detailed, scalable vector.
Live in Ideogram and the API: ideogram.ai/models/vector
building a brand is a moat. and you don’t need a big budget to start.
the fastest way to do that at an early-stage startup is creators (not necessarily the big shots).
every partnership adds to brand equity that compounds over time.
start on day 1. don’t wait until you’ve “earned” a brand budget!
"How do you compete with the labs?"
Last Thursday, I attended a panel by my colleague @thepmfguy re influencer marketing, followed by a keynote by @thisisgrantlee , the founder of Gamma.
Someone in the audience asked Grant straight up: Claude is already really good at making slides, so how will Gamma compete with the labs?
Grant's answer had three parts:
1) If the labs go general purpose, go purpose-built. Pick a persona and a use case where the nitty-gritty details actually matter.
2) Ask what the most ambitious version of your company is. Grant said Gamma isn't an AI slide-making company. Its mission is to help individuals and businesses tell their story, and he's actively thinking about how Gamma can become a company's entire autonomous storytelling function, something a model can't do (yet)
3) Brand is a moat. If software is a commodity, brand matters more: how you look, what you stand for, what people feel when they think of your product. And he's been investing in this since day 1.
Kudos to @aryangulzar_ , AKG Media for organizing.
Gaurav Bisen retweeted
Behind @GammaApp’s incredible growth story at #SFTechWeek.
CEO @thisisgrantlee joins @aryangulzar of @akg_media_uk for a fireside chat. Plus, a creator marketing panel with Fiona Turko (@GammaApp), Eduardo Sanchez-Ubanell (@WisprFlow) and Gaurav Singh Bisen (@thepmfguy)
what's next?
instinct will slide into dms and answer on my behalf as well? 🤔
Instinct in group chats
Starting today, early access users can add Instinct to group chats. A new Instinct joins and works for the whole group.
Making plans with friends usually turns into a frustrating back-and-forth over times and places. With Instinct in the group, you can explore options together, agree on a plan and get it done, all in one thread. Your friends don't even need Instinct to join in.
A few things it’s good at:
- Planning a weekend trip with friends, including dates and arrival times
- Coordinating logistics with a roommate
- Getting tickets as soon as they go on sale, then splitting the cost and sending one to everyone
- Running a fantasy league, keeping standings, settling rules disputes, and reminding everyone to set their lineups
- Sorting out carpool schedules - works out who’s driving each day from everyone’s schedules
- Organizing Thanksgiving, keeping track of who’s bringing what and ordering the shopping list
Group chats are new territory for Instinct, and we put a lot of care into getting the permissions right. Here’s how it works:
- Your personal Instinct asks before connecting with the group’s Instinct. You choose which groups to trust and can remove trust at any time.
- The group’s Instinct has no direct access to your personal accounts. Your Instinct checks your permissions before sharing information or taking action.
- If someone new joins the group, all pending replies from your Instinct are held until you grant it permission to share with the expanded group.
Group chats are available to all of our early access users, and will be rolling out to everyone soon. To join early access for this release, ask your Instinct to put you on the list.
Great work with @GajanNagaraj @anuda_w and @saagarwashere
Gaurav Bisen retweeted
Most positioning problems show up in 5 places:
1. Clarity
Can a buyer explain what you do after one sentence?
2. Market
Can they tell why you win instead of comparing you to the wrong competitor?
3. Value
Do your features connect to an outcome they already want?
4. Sales
Can a champion explain your value internally without calling you?
5. Alignment
Would your founder, marketer, and salesperson describe the product the same way?
If one answer is "no," more content will not fix the problem.
Fix the message first.
Good stuff @AyushkWadhwa 🙌
Have put both muse and instinct to try for a complimentary flight upgrade! 99.99% chance both will lose, but you never know! lfg!
Gaurav Bisen retweeted
I'm building a video editor by making Cursor edit my video
This is a Cloud Agent with the editor loaded
Cursor is using Grok 4.7 to edit the video via MCP, then testing the UI
At each failure, the agent reflects then fixes the product. The goal is a clean end-to-end edit
With the right feedback loops, you can autonomously build software
a year ago this was a $20k agency invoice and two weeks of back-and-forth. now it's one person with taste and an afternoon
Gaurav Bisen retweeted
Sooo you're telling me that:
- You can research winning ads with the Parker or Trendtrack MCPs.
- You can turn that research into concepts, scripts and briefs with Claude.
- You can generate statics and AI video with the Higgsfield MCP.
- You can source thousands of creators with Apify or Trybe.
- You can review their footage against the brief with the Google Drive MCP.
- You can QA the final ads and leave timestamped feedback with the Frame MCP.
- You can edit the footage and create hook variations with Astra.
And the whole process can be orchestrated agentically through Claude Code or Codex.
We are seeing the emergence of the one-person creative team in real time.
What a time to be alive.
Gaurav Bisen retweeted
We hit $18M ARR at @heyreach_io
Every milestone looks great from the outside. But there's always good and bad behind it. So here's both, plus what I learned along the way. if you're building a SaaS in 2026, hopefully some of it saves you time:
THE GOOD:
1. We launched Voice Notes, the most requested feature
For a year and a half, on the top of our most requested features was Voice notes.
Our competitors had this feature long before us, but IMO, none of them executed it the right way. We lost a ton of deals because of that, and we still didn't prioritize it.
Why?
The technology wasn't there yet, and it would have been a clunky and messy execution. Every feature that's out in HeyReach, is really well-thought, and it's solving a problem.
Now when the voice models became affordable and much better, we decided to do this feature.
Bottomline...you don't have to build something just because your competitors have it. The main question you have to ask yourself is: "can we solve this for the user in a really good way right now?" if not, wait (even if you lose some deals).
2. We finally hired a product person
One of the biggest bottlenecks at HeyReach was... me.
I was the guy owning product since day one. At the start, that's normal. You're product, sales, marketing, ops and support, and we couldn't even get a single senior person to join full-time before $1M ARR.
Then HeyReach kept growing, and I kept owning product. At some point, I started being the main problem.
This summer, I did something about it. I brought in someone who's been doing product for 15+ years.
Probably the best decision I could have done this summer.
3. We are slowly repositioning HeyReach toward being AI-first
The industry is moving forward, and GTM teams are already becoming agentic.
At one point our roadmap had an AI SDR, AI hyper-personalization, AI profile optimization... I'm super glad we didn't blindly follow that route. We wanted to do it the right way.
A year ago we shipped our MCP, and customers started using agents with HeyReach in ways we hadn't even thought of.
We are changing our positioning in the next few months, and considering the amount of data and know-how we have about the LinkedIn outreach space, I'm sure we'll nail it.
4. Our EBITDA is 73%+ (I don't think this needs any more explaining haha)
5. We have an exciting launch coming up next month (and a potential new product)
As HeyReach is becoming more and more autonomous (we have an amazing team), I started feeling a bit obsolete.
Then, in March this year I had an opportunity to meet a founder with a great product who was down to sell it to us. We acquired it, and changed the business model.
Planning to launch it publicly mid-October. I think it has a potential to be 5-10x bigger than HeyReach.
THE BAD:
1. Growth slowed down (seasonality + exhausting the TAM)
We got used to 10-15% MoM growth last year, but this year, we are in the single digits (7% MoM being the average).
The product is better than ever, the support is better than ever, more and more people are hopping into LinkedIn outreach, so what's the problem?
I think we are slowly exhausting the TAM with the old positioning. Our feature-set so far was directed mainly toward agencies, and just in the recent months we started shifting it toward sales & GTM teams.
But I'd still tell every early founder to niche down, it's literally how we got here (without having to raise).
2. I wanna turn HeyReach into a $50M ARR company, but I'm not sure the market is that big (or if it's gonna grow that fast)
3. My focus is torn between 2 products, and sometimes I feel like I'm not doing the best work
On a good track to surpass $20M ARR this year (2x than what we were in January)!
If you've been a user or a current customer, cheers for making it along the way. Watch out for our upcoming launch next month 💪🏻
Fin.
higgsfield just crossed $1b annualized. 18 months after launch.
most people will read that as an ai video company winning. i read it as proof that the way brands get content made is changing for real, faster than most teams are set up for.
two years ago i was telling brands the opposite. the quality they wanted could only come from human production. i spent enough time fighting the old models to know - they genuinely could not get there, and i never shipped work i wasn't proud of just because ai made it cheap. that is not true anymore. the models are good now. the bar is high, but it is reachable for people who obsess over the output.
however - this is not slop scaling. the winners will not be the ones generating 100 posts a day. they will be the ones who unlearn how they worked before and relearn with the new stack. i learned this in my unacademy days and it keeps proving itself: whatever you have done, whatever you have worked on, might not be true anymore, might not work the same way. the talent that wins is the talent willing to be a beginner again.
on the "models will eat the app layer" debate - i don't buy it, not today.
yes, the value lands on the model. the model keeps getting better. but the model by itself serves maybe 1% of people: the driven, hyper-focused ones who want to sit inside the problem and push it directly. the other 99% will never do that. someone has to carry the model to them in a way that feels simple. that is the application layer.
we have seen this before. google has been around for decades and most of the world still doesn't know the best way to search. entire companies got built on making that easier. same principle, new stack.
and yes, there is the usual pushback: "whatever agents can do, i can do it with claude code, i can do it with codex." sure, you can. but you are not the whole world. most people will never sit down and brainstorm with a raw model to get work done. there is a reason even the model companies invest so heavily in their own app layers - they know the model alone does not reach everyone.
so no, the app layer is not getting squeezed today. maybe 10 years from now the conversation changes. right now there is more room to build on top of these models than there has ever been. $1b in 18 months is what that room looks like.
gave a coach bag its own muse. no model, no celebrity, no influencer. just a tiny plush character carrying the whole campaign
Gaurav Bisen retweeted
Got bored rn, so here’s the exact onboarding flow I used to consistently get 12%+ download to paid…
Screen 1: Show a screen recording with a phone mockup of your app
Screen 2: Goal
Screens 3–4: Gender + age
Screen 5: Graph showing your app compared to competitors
Screen 6: Ask for their name
Screen 7: Pain point
Screen 8: Another pain point
Screen 9: Specific niche question
Screen 10: Push notification permission
Screen 11: “Building your custom plan” loading screen
Screen 12: Show their custom plan, referring back to all the questions they answered
Screen 13: Reviews / social proof
Screens 14–15: iPhone mockups with big titles selling the transformation
Paywall
The key is making the user feel like the app is being built specifically for THEM before they ever see the paywall. Use there name throughout onboarding and epically on the paywall title
the first $1k is mostly a distribution test. a lot of products stuck at $0 never got in front of enough real buyers to find out if anyone wanted it
I was looking at TrustMRR which tracks almost $1.9B in verified startup revenue. 67.7% of the startups there have made less than $1k.
What surprised me is that of the ones that got past $1k almost half are already above $10k.
So maybe the hardest milestone isn't $1M or even $100k. It's the first $1k. That seems to be where most founders get filtered out.
ngl - every brand i talk to wants generated content. 50 ai creators, 100 posts a day. then i ask what their best performing ad is and it's always a real person who actually used the thing
talking all things creator growth, gtm, and what's actually working for ai first companies in 2026
@aryangulzar_ excited for this! 🙌
