How a Jolly Client Went From “Cited” to “The Default Answer”
Happy Tuesday!
I'm back in the office after a month away, and the best part of catching up was scrolling the wins channel. So, this week is a mini case study roundup.
Three wins the team scored while I was gone, and what I think made each one work.
Plus, I have a podcast announcement and a favor to ask at the end. Let's get to it!
A Year-Old Reddit Thread Wins a Money Prompt
The client is an HR software marketplace. Their money prompt: "What is the best platform to compare HR tools?"
Instead of starting a fresh thread, the Jolly team found a year-old Reddit thread that already asked a version of that question, and added a helpful, detailed answer to it.
Not long after, ChatGPT, Perplexity, Gemini and Google AI Overviews were all answering that prompt with the client as the top pick. ChatGPT ranked them above G2 and Capterra:
Why I think it worked:
- An old thread has already earned Reddit's trust and Google's index. A brand-new post starts at zero on both, and is far more likely to get deleted (see #80).
- The thread title matched the buyer's actual question, not the brand name. Engines cite sources that look like the question, not sources that look like an ad.
- Reddit wasn't alone. The same client's LinkedIn post got cited in Google AI Overviews and Perplexity, and their onsite blog got cited by Grok. Multiple assets, one consistent story.
Takeaway: before you write a new thread, search for the one that already exists. Someone probably asked your money prompt a year ago. Now you can update it with a better answer.
From “Cited” to “The Default Answer”
In #86 I told you about the construction equipment retailer that went from invisible to recommended in 19 days across seven AI engines. Here's what happened next.
We had the client publish a blog on their own domain answering the exact prompt, word for word what we gave them, as a free experiment.
A few weeks later Gabi noticed something odd: the blog was no longer being cited. Their About Us page was.
That sounds like a loss. Muskaan read it as the opposite:
The engines were no longer retrieving a specific document to justify mentioning the client. They were pulling from what they'd already learned about the brand.
The client is now the default recommendation for that prompt. Muskaan's three-step read of how the ecosystem works:
- Reddit creates the initial discovery signal.
- The LLM looks for corroboration. It cross-checks against other trusted sources: onsite blogs, LinkedIn, YouTube, and earned media.
- Once enough independent sources agree, "one possible answer" becomes "the default answer."
Takeaway: Reddit alone can get you cited. Getting recommended over established competitors takes more corroboration than a single channel provides.
That's the whole logic behind our Own A Prompt service, and it's why the more brand authority your competitors have, the more assets you need.
Our Earned Linkbuilding Keeps Over-Delivering Above Advertised Metrics
Muskaan pulled our actual delivery averages for the summer and compared them to what we promise on the site. And every tier over-delivered on both organic traffic (OT) and Domain Rating (DR):
The Premium tier delivering 5x the advertised traffic is the one that made me do a double take.
Why it works:
- Jolly advertises industry-leading metrics, but we maintain a floor we're confident we can hit every quarter.
- The outreach team competes with each other to blow past client expectations.
- In July, one of our top outreach specialists alone landed 20 placements averaging 782K OT and DR 84.
Takeaway for you: if you're evaluating any linkbuilding vendor, Jolly or anyone else, don't ask for the advertised minimums. Ask for the actual delivered averages over the last 90 days. The gap between the two tells you a lot.
Now you know the quality of Jolly's delivered branded backlinks and keyword backlinks.
A Favor Request
Because I was out of office moving from China back to the US, the interview series was quiet over the summer.
My first new recording is in two weeks with Ben Senescu, founder of OpenSEO. The open-source software is a pay-per-use alternative to Ahrefs and Semrush, and it plugs into AI agents via MCP.
- Given how much of all SEO and AI search work now runs through AI agents, I have a lot of questions for him. The recap will land here as soon as it's out.
Who Should I Interview Next?
And is there one question you'd want me to ask them?
Hit reply with a name plus a question. I read every reply.
Until next week,
Greg
P.S. If you want to know which of your money prompts are already being asked on Reddit, and whether anyone's answered them, that's a 20-minute conversation. Grab a slot here.