1
2 Comments

Month 1 of my productized design subscription: 49 cold outreach messages, 0 replies. What I changed.

I started my design subscription from scratch. No audience, no reviews yet, no warm list to lean on. So in the first month I did what everyone tells you to do. I found founders who looked like a fit, connected with them, and sent messages. I sent 49 of them. I got zero replies. Not a slow trickle. Zero.

Here is the message I was sending, more or less every time.

"Thanks for connecting. Curious, for you is the harder part getting the design right, or getting it actually built?"

On paper it looks fine. It is short, it is a question, it is about them. That is exactly why I trusted it and kept sending it. But 49 sends and no replies is not bad luck. It is feedback. So I stopped and actually read my own message the way a stranger would.

Three things were wrong with it.

The first is that it sounded like AI wrote it. That exact "is the harder part A or B" shape is everywhere now. The moment someone recognizes the template, they stop reading, because the template tells them a real person did not think about them specifically.

The second is that it asked a stranger to do work. "Which part is harder for you" is a question that makes the other person stop, reflect, and type out an answer, all to help someone they do not know sell them something. Nobody owes you that, especially not in a cold message.

The third is that it gave nothing. There was no reason for them to reply beyond being polite, and busy people are not polite to cold DMs. I was asking to take before I had given anything at all.

So I threw the whole approach out and rebuilt it around one rule. Give something real before you ask for anything.

Now, before I message anyone, I actually open their product. I sign up, I click through their onboarding, I read their pricing page. Then I find one specific, true thing I would change, and I lead with it. Not a category truism like "your onboarding could be smoother." Something I could only know by looking, like a placeholder still sitting live on the homepage, or two buttons competing for the same click on the pricing page.

The new message looks more like this.

"I was looking at your onboarding and noticed the dashboard lands empty on the first run, which is usually where trials go quiet. I put together two specific ideas for that first screen. Want me to send them?"

It is specific, so it proves I actually looked. It offers something useful, the free notes, before asking for anything. And it closes with a question they can answer in one second, yes or no, instead of an essay. I also stopped reusing the exact wording, because the whole point is that it should not feel like a formula. Different person, different screen, different opening line.

I am not going to pretend this turned into a flood overnight. It did not. But the difference between "zero replies" and "actual conversations with real people" was not a bigger audience or a cleverer hook. It was the order of operations. Give first, ask second, and make the ask tiny.

If you are early and cold outreach is going nowhere, before you blame the channel, read your own message as if a stranger sent it to you. Would you reply? If the honest answer is no, the problem is not the number of messages. It is the message.

on August 17, 2026
  1. 1

    The empty-dashboard example works because the recipient can check the observation themselves.

    I think outreach becomes less convincing when it starts with something true, then jumps straight to an impact the sender can’t prove. For example, noticing a slow page and immediately claiming it is costing the company customers.

    Did anyone challenge your interpretation of what you found, or were the conversations mostly positive once you changed the opening?

  2. 1

    This is a good lesson: cold outreach isn't really a volume game if the message gives the other person no reason to care.

    I’ve found the same thing when reaching out for product feedback. A specific observation about someone’s product creates a much better conversation than a generic “would you like to test this?”

    The “give something useful before asking” part is probably the biggest takeaway here. It also makes the outreach feel like a human interaction rather than a sales sequence.