We dogfooded our own product. That means we used Sandbox to run Sandbox's GTM — prospecting, outreach, follow-up sequences, content distribution. Everything we sell, we ran against our own ICP.
After 700+ outreach contacts, 119 blog posts, and two active email campaigns running simultaneously, we have real data on what resonates with the operators we're trying to reach: founders running small businesses, consultants managing multiple clients, serial entrepreneurs who have done this before and are running it again.
Here's what we actually learned — including the things that didn't work the way we expected.
The 58–63% open rate is the headline stat. Industry average for cold outreach is 20–25%. We're running at 2.5–3x that with no manual personalization per contact — all of it generated through Sandbox's agent execution layer from a single prompt per campaign.
We ran two distinct campaigns simultaneously with different positioning:
| Campaign | Angle | Open Rate | Unique Opens |
|---|---|---|---|
| Sandbox Operator Outreach | Execution overhead / capacity | 54% | 39 of 63 |
| Friday Pipeline Panic | Pipeline urgency / follow-up | 63% | 40 of 63 |
The Friday Pipeline Panic campaign — which leads with pipeline urgency and the emotional context of end-of-week stress — opens at nearly 10 points higher. The difference: execution overhead is a rational problem, pipeline panic is an emotional one. Operators open emails about things they feel, not just things they think about.
The highest-performing subject lines across both campaigns share three characteristics:
A 60%+ open rate means the problem framing is right. These operators recognize themselves in the subject lines. They are bandwidth-constrained, they do feel pipeline urgency, they have thought about GTM hires and found the math difficult. The opens confirm the ICP is correct — we're reaching the right people with the right framing.
What the 0 click rate tells us is that the conversion path from awareness to action is still broken. High opens mean strong problem-market fit for the content. Zero clicks means the offer or the CTA doesn't feel like the natural next step for where these operators are in their decision process.
Looking at which emails got the highest open rates, and what content drove the most index page visits, a pattern emerges:
If we were starting the outreach motion over with this data, here's what we'd change:
The biggest learning: email converts to reply better than email converts to click. An operator who replies "yes, we've been dealing with this" is more likely to eventually book a call than one who clicks a link and reads a blog post. Replies lower the friction of the first real interaction.
We built this entire GTM motion — 700+ prospects, 119 blogs, two active campaigns — without a single SDR, without a content team, without a marketing hire. Everything runs from prompts. Not perfectly — we'd iterate several things — but consistently, and at a volume and quality that would take a team of 4–5 to produce manually.
That's the point of the dogfood. Not to show a perfect result, but to show that the execution layer actually runs. The 60%+ open rates aren't an artifact of a polished pitch — they're the result of well-targeted outreach produced at scale with no human bottleneck in the execution loop.
If you're a small business operator wondering whether this works for your type of business, the answer we have is: it's working for us, and we're the same kind of business we're selling to.
15 minutes. We'll walk through the exact prompts, campaign architecture, and how the execution layer actually works. No demo script — just the real setup.
Book a 15-Minute Walk-ThroughSandbox · rob@sandboxgtm.com · sandbox.co