
I make people comment "Claude" on LinkedIn │Head of Content @Gojiberryai
San Francisco
Joined December 2022
- Tweets1.5K
- Following585
- Followers1.6K
- Likes1.9K
Pinned Tweet
Gojiberry (YC) went from 0 to 4,000+ paying customers in under a year by doing the opposite of what most people recommend.
We only ran one channel at a time. We didn’t add the next one until the current one was clearly working.
0 → 100: Pure outbound.
Cold email + LinkedIn using an early version of our own product. No brand, audience, or content. Just finding people showing intent and starting conversations. Ugly, manual, and it worked.
100 → 250: Reddit.
Educational posts in SaaS subreddits. 10M+ organic views. Traffic quality was mixed, but the volume filled the trial funnel for months. Cost: basically zero.
250 → 750: Content + free blueprints.
LinkedIn, YouTube, motion videos, and giving away our internal systems as free blueprints. Content got the reach. Blueprints built the trust. Founder-led content started compounding here.
750 → 1,500: Partnerships + X.
B2B influencers, sponsored newsletters, X, and a lifetime affiliate program that became a real lever. YC happened around this point and everything accelerated.
1,500+: Paid + hiring.
Meta, Google, influencer agencies, and the first real hires across growth, sales, engg, and product.
Every previous channel still runs. We just never started the next one until the current one was working.
Teams usually try to run 4–5 channels at once and wonder why none of them work.
What’s the one channel you’d actually double down on right now instead of spreading yourself thin?
🤖 Made with AI
We met an a16z founder who gave us his playbook to get customers BEFORE starting cold outreach.
It’s something most sales teams and founders completely overlook.
Before spending time and money finding new prospects, make sure you’ve exhausted the opportunities sitting right in front of you.
Your CRM and LinkedIn 1st-degree network are goldmines.
The playbook:
1. Start with your CRM
Look at the contacts and companies you already have.
Find:
→ Old leads that never converted.
→ Prospects who went cold.
→ Lost opportunities.
→ Past customers.
→ Contacts who match your ICP but were never properly approached.
2. Mine your LinkedIn 1st-degree network
You probably have thousands of existing connections you’re not using like you should.
Find people who:
→ Match your ICP.
→ Work at companies you’re targeting.
→ Recently changed jobs or companies.
→ Could become customers, partners or referral sources.
3. Enrich and segment everything
Don’t just dump everyone into an outreach sequence.
Filter by:
→ ICP fit.
→ Job title and seniority.
→ Company size.
→ Industry.
→ Company relevance.
4. Prioritize warm opportunities
Start with people who already know you, have interacted with your company or have previously shown interest.
You don’t need to introduce yourself from scratch.
5. Only THEN start cold outbound
Once you’ve properly worked through your existing network, start looking for completely new prospects.
Why this works:
→ Less friction. You’re reaching people with whom you already have some connection.
→ Better personalization. You have context before sending the first message.
→ Lower acquisition costs. You’re leveraging data and relationships you’ve already paid to acquire.
→ Faster opportunities. Some of your next customers might already be sitting in your CRM.
You can pull your CRM contacts and your entire LinkedIn 1st-degree network directly into @GojiBerryai
Import, enrich, filter against your ICP and build your next outbound campaigns.
Good luck!
Try gojiberry.ai here (free for 7 days)
1
17
Prerona Basu retweeted
This is cool
my honest experience using @gojiberryai
READ TILL END...
DAY 0:
I put in my product info and the pain points as detailed as possible.
I found 0 leads that are actually useful or relevant
~300 leads it gave are all about people who raised or switched jobs recently
ZERO ICP fit
DAY 1:
Went back briefly but didn’t find anything interesting.
But kept it going as it's still on free week trial
DAY 2:
Got an email with 18 new leads
All look at least ICP adjacent
Even if one good ICP lands a week, that's a win.
I will keep testing and report back. I see the hype now. Well done @romanbuildsaas and team
We’re sick of the term “AI agents.”
Everyone’s building one and everyone’s selling one.
There’s an AI agent for literally everything.
Finance. Marketing. SEO. Sales.
But has anyone stopped to think about the little guys doing all the work?
Meet Goji.
He’s our tiny, extremely hardworking sales buddy.
He finds high-intent prospects, starts conversations, handles the follow-ups, and keeps going until demos land on YOUR calendar.
Some mornings, nobody replies. Sometimes people say no.
Goji doesn’t take it personally. He just keeps working.
And he’ll work for you FREE for 7 days 👇
Give the little guy a chance.
He might just become your favourite teammate. 🧡
Prerona Basu retweeted
We’re sick of the term “AI agents.”
Everyone’s building one and everyone’s selling one.
There’s an AI agent for literally everything.
Finance. Marketing. SEO. Sales.
But has anyone stopped to think about the little guys doing all the work?
Meet Goji.
He’s our tiny, extremely hardworking sales buddy.
He finds high-intent prospects, starts conversations, handles the follow-ups, and keeps going until demos land on YOUR calendar.
Some mornings, nobody replies. Sometimes people say no.
Goji doesn’t take it personally. He just keeps working.
And he’ll work for you FREE for 7 days 👇
Give the little guy a chance.
He might just become your favourite teammate. 🧡
We just launched a Chrome extension that turns any company website into a prospecting opportunity.
Here's how it works:
1/ Visit any company's website.
2/ Launch the GojiberryAI extension.
3/ See the employees and filter them by job title.
4/ Import the right prospects into your CRM or outreach campaign.
5/ Reach out and turn them into potential customers.
No more manually searching for employees or guessing who to contact.
Find your next buyer while browsing the company's website.
Prerona Basu retweeted
We just launched a Chrome extension that turns any company website into a prospecting opportunity.
Here's how it works:
1/ Visit any company's website.
2/ Launch the GojiberryAI extension.
3/ See the employees and filter them by job title.
4/ Import the right prospects into your CRM or outreach campaign.
5/ Reach out and turn them into potential customers.
No more manually searching for employees or guessing who to contact.
Find your next buyer while browsing the company's website.
Legacy lead databases are cooked 🪦
🚨 Introducing the world’s first High-Intent Lead Database.
800M+ leads. Unique data points. One prompt.
Your AI agents find the right people, uncover signals that matter, and identify the right moment to reach out.
Available now via MCP or API. and within Gojiberry.
While your competitors are all using the same old database, this is your chance to outperform them.
7
392
Prerona Basu retweeted
Promise me you won’t use this for evil!
We’re giving every founder a full sales team of 6 AI agents that:
- Find buyers showing intent
- Score and enrich them
- Start conversations for you
They know your business, and nothing goes out until you say so.
Try it for free right here: gojiberry.ai/?utm_source=x&u…
Also - reply “LEADS” + your site for a free GTM audit.
Promise me you won’t use this for evil!
We’re giving every founder a full sales team of 6 AI agents that:
- Find buyers showing intent
- Score and enrich them
- Start conversations for you
They know your business, and nothing goes out until you say so.
Try it for free right here: gojiberry.ai/?utm_source=x&u…
Also - reply “LEADS” + your site for a free GTM audit.
Prerona Basu retweeted
There’s a lesson taught at Y Combinator that can turn 4 weeks of GTM research into just 3 hours.
And no, it’s not typing “research the market for X” into Claude.
You start by giving it raw materials:
→ 25-40 competitor websites, fully scraped (pricing pages, case studies, docs)
→ Recent earnings calls from the 3-5 biggest players in the market
→ 100+ customer reviews (G2, Capterra, Trustpilot, app stores)
→ Reddit threads and forum discussions where customers openly complain about the category
Then ask questions that make it dig deeper instead of summarizing what it finds:
“What do the winners in this market understand that their customers would never openly admit?”
“List the assumptions this entire industry operates on. For each one, what would have to happen for it to be completely wrong?”
Within 20 minutes, you'll start spotting patterns.
Where competitors overlap.
What everyone is overlooking.
And which customer problems are still waiting for a real solution.
Now turn the tables.
Take your strongest angle and ask Claude to challenge it:
“Steelman this thesis like a top-tier VC would. Then show me exactly where it falls apart.”
Run a few rounds of this.
Three hours later, you'll have a first GTM hypothesis backed by research that would normally take weeks to pull together.
The difference is in using Claude as a thinking partner. One that can challenge your assumptions, explore different angles, and keep pushing your ideas further.
Of course, a hypothesis is only a starting point. You still need to validate it with actual buyers.
To get there fast, use Gojiberry to find and reach the right people to test it with.
Customer conversations are the only thing that matters.
There’s a lesson taught at Y Combinator that can turn 4 weeks of GTM research into just 3 hours.
And no, it’s not typing “research the market for X” into Claude.
You start by giving it raw materials:
→ 25-40 competitor websites, fully scraped (pricing pages, case studies, docs)
→ Recent earnings calls from the 3-5 biggest players in the market
→ 100+ customer reviews (G2, Capterra, Trustpilot, app stores)
→ Reddit threads and forum discussions where customers openly complain about the category
Then ask questions that make it dig deeper instead of summarizing what it finds:
“What do the winners in this market understand that their customers would never openly admit?”
“List the assumptions this entire industry operates on. For each one, what would have to happen for it to be completely wrong?”
Within 20 minutes, you'll start spotting patterns.
Where competitors overlap.
What everyone is overlooking.
And which customer problems are still waiting for a real solution.
Now turn the tables.
Take your strongest angle and ask Claude to challenge it:
“Steelman this thesis like a top-tier VC would. Then show me exactly where it falls apart.”
Run a few rounds of this.
Three hours later, you'll have a first GTM hypothesis backed by research that would normally take weeks to pull together.
The difference is in using Claude as a thinking partner. One that can challenge your assumptions, explore different angles, and keep pushing your ideas further.
Of course, a hypothesis is only a starting point. You still need to validate it with actual buyers.
To get there fast, use Gojiberry to find and reach the right people to test it with.
Customer conversations are the only thing that matters.
Try Gojiberry.ai for free 👇
1
106
Grab this free course right now! Just RT and comment "SAAS" for access 🙌🏼
Gojiberry AI just crossed $5M ARR.
To celebrate, I put together a free course breaking down everything we learned on the journey from $0 to $5M ARR.
No generic startup advice. Just the actual strategies, experiments, mistakes, and growth tactics that moved the needle.
Want the playbook?
RT + comment "SAAS" and I'll send it to you as soon as it's live.
5
201
Prerona Basu retweeted
You'll read it here first, before the rest of the market catches on.
There's an absolutely insane way to book demos right now. And the conversion rates are through the roof.
It combines two powerful things:
1. Finding high-intent leads and reaching out to them through LinkedIn + email.
2. Creating personalized blueprints at scale.
Here's why this works.
Most of your competitors are sending generic blueprints, reports, or resources to their lead lists.
In enterprise sales, reps have the time to personalize resources for individual prospects.
In SMB and mid-market? Almost never.
And that's where you can create a HUGE competitive advantage.
Thanks to AI, delivering a premium, highly personalized experience to every prospect has never been easier.
Here's exactly how I'd approach it:
1/ Build a list of qualified leads.
You can do this manually or use gojiberry(dot)ai to identify high-intent prospects based on buying signals that indicate they're more likely to convert.
For example, someone following one of your competitors is an intent signal worth acting on.
2/ Make a personalized blueprint part of your outreach offer.
Instead of simply asking for a demo, offer prospects a PERSONALIZED blueprint showing exactly how you could help them achieve [X] using [Y].
3/ Build an incredible resource for your market.
Create a blueprint with customizable sections that can be automatically populated for each prospect.
I built mine in Lovable using a simple interface.
All I have to do is enter a prospect's email address.
It automatically:
- Adds their name throughout the report.
- Uses their company domain to gather relevant business information.
- Personalizes the blueprint based on their company's context.
4/ When someone expresses interest, generate their personalized blueprint in just two clicks.
Then send it straight back as part of your follow-up.
5/ Give prospects an experience your competitors aren't offering.
Instead of receiving yet another generic PDF, they get something that includes their name, their company, and, most importantly, recommendations tailored to THEIR specific situation.
And here's another use case:
You can generate the same type of personalized report after every sales call or whenever someone requests a resource on your website.
Your prospects and customers LOVE that level of attention to detail.
What used to take 20 minutes now takes 2.
Your prospect gets a premium experience. Your sales team barely spends any extra time.
That's the real power of AI in sales.
Welcome to the new era :)
Look at the example below.
I start with an intent signal: someone following one of my competitors.
Then, I send them a personalized blueprint showing exactly how we can help their business.
I've shared an example of what this personalized blueprint looks like in the first comment.
This is the highest-ROI outbound move we’ve seen in 5+ years.
Instead of recording voice notes one by one, do this:
Record your voice once
Pull prospects already showing buying signals
Let AI clone your voice and write a custom message for each one
Send them to everyone
And the demos basically book themselves
Because every message goes to someone already showing intent, every one sounds like you, and the grand total of your effort is about 4 minutes of talking into a mic.
Name any other outreach method that has THIS much potential for THIS less effort.
I'll wait😁
Prerona Basu retweeted
We found a LinkedIn message format that keeps booking demos.
And this is how we found it:
We analysed thousands of LinkedIn accounts that collectively sent 500,000+ messages through @gojiberryai
We looked for patterns across the strongest-performing campaigns.
One message structure kept appearing among the accounts booking the most demos.
So we turned it into a ready-to-use template.
I’ve shared it in the image👇🏼
Steal it. Test it. Book more demos.
Let me know if it works 👊🏼
We found a LinkedIn message format that keeps booking demos.
And this is how we found it:
We analysed thousands of LinkedIn accounts that collectively sent 500,000+ messages through @gojiberryai
We looked for patterns across the strongest-performing campaigns.
One message structure kept appearing among the accounts booking the most demos.
So we turned it into a ready-to-use template.
I’ve shared it in the image👇🏼
Steal it. Test it. Book more demos.
Let me know if it works 👊🏼
Find high-intent leads and reach out with personalised messages at gojiberry.ai.
Try it free for 7 days.
Reaching out to high-intent leads with personalised messages can drive a 40% connection acceptance rate and a 35%+ reply rate.
The key: reach the right people when they’re already showing interest.
90






