When outbound campaigns underperform, the knee-jerk reaction is to check domain health, SPF/DKIM records, and technical setup. We spend days optimizing deliverability metrics.

But looking closely at open to reply ratios, the technical setup was fine. The issue was that the email sounded like an automated sales script disguised as a quick chat. Technical infrastructure gets your message to the inbox, but clear, humanized messaging is what keeps it out of the trash.

How do you audit your copy to ensure it sounds like a real human writing to another human, rather than a bot following a sequence?

In the early days without a proper HR setup it s really hard to tell who genuinely believes in your vision versus candidates who just want a temporary gig while job hunting.

We had to change our interview style from standard technical coding tests to giving candidates messy real world problems to solve alongside us.

How do you test for true startup alignment and risk tolerance in early interviews before making an offer?

Most early text features suffer from the same problem the generated blogs and copy lack any actual personality making users waste 20 minutes editing out AI buzzwords before publishing.

We found that the only way to get consistent output was building tight structural boundaries directly into the prompt framework rather than relying on creative instructions. Once we forced hard formatting rules on the model the draft quality spiked immediately.

Have you found any specific prompt constraints that reliably preserve a natural writing style?

For a long time, I used to dread getting cancellation feedback explaining why our tool didn't work out. But looking back at the analytics, the users who quietly slip away without saying a word are much scarier. At least when someone complains about a broken or a confusing menu enough to push back. Silence just means they've already moved on.

What s your preferred way to extract honest feedback from users who cancel without leaving a note?

When our product was free, people were wonderfully forgiving of bugs, clunky navigation, and missing documentation. The moment we introduced a credit card requirement, the entire relationship shifted overnight. Every tiny friction point turned into an immediate support ticket or a silent cancellation because paying customers expect an instant return on their mental energy.

How long did it take your product to feel "worth paying for" after you made the transition away from freemium?

Understandably, when you team up with someone who is more popular than you are, it can help you grow your audience, create awareness about your presence and possibly increase purchases.

Most of the time, we research our competition and benchmarks.

Announcing: Voice Agents course and online community ...

@swyx and I are hosting a month-long technical deep dive into Voice AI and Voice Agents, starting in May.
Our goals are to:
cover all the lessons we've learned over the last two years building realtime, conversational AI,
host fun sessions with all our favorite people who are doing related things,
and
build a long-term online community.
Sign-up link: https://lnkd.in/gnPuHyD4
We'll start announcing free credits for students on Monday. Sign up this weekend with promo code PHUNT for a super-secret Product Hunt community discount.
Last year I signed up for the LLM fine-tuning course taught by Hamel Husain and Dan Becker.
The experience was fantastic in every way. The material was great. The course expanded to cover way more than fine-tuning. It seemed like all of Twitter signed up. I met people in the course Discord that have become online and offline friends. Someone eventually dubbed the course "AI Woodstock." (I think credit for that goes to Swyx.)
We think this is the moment to try to create a similar thing for voice AI.
Voice interfaces are going to be a huge part of the near-future of computing. Voice agents are being deployed at scale today for a wide range of use cases.
collecting patient data prior to healthcare appointments
following up on inbound sales leads,
handling an increasing variety of call center tasks,
coordinating scheduling and logistics between companies, and
answering the phone for nearly every kind of small business.
I'm personally excited about voice interactions for games, realtime video, and voice-enabled programming environments.
https://lnkd.in/gnPuHyD4
Promo code: PHUNT

A few minutes ago, I read an overview of a report that more people take a source of news, not traditional outlets and media, but influencers. [I am attaching the screenshot from Reutersinstitute.]

TL;DR:

FB, YT, IG  More attention is dedicated to news media/journalists