Week by week, channel by channel. What we did, what broke, and what we learned.
2,750
visitors in 8 weeks
107
signups (3.9%)
40+
places sent them
$0
ad spend at 600 a week
| Week of | Visitors |
|---|---|
| 3 Aug | 0 |
| 10 Aug | 245 |
| 17 Aug | 382 |
| 24 Aug | 259 |
| 31 Aug | 380 |
| 7 Sep | 477 |
| 14 Sep | 600 |
The story
Chapter 1 · Late July
We believed Scout7 could grow a company without ads. So we set a test: run ads and organic side by side, and see which one really brings people. That week, our site had 0 visitors.
667,395
ad views
Ads need weeks of tuning before they find the right people. We didn’t have weeks.
Google, Meta, Reddit and LinkedIn. We kept changing the ads toward whatever brought more people. Clicks were cheap, about 13 cents. People came to look. Nobody signed up.
667,395
ad views
Ads need weeks of tuning before they find the right people. We didn’t have weeks.
OFF
no more paid ads
Ads rent attention. Stop paying and it’s gone.
The moment we switched them off, the ad traffic was gone. Nothing stayed. We had rented attention, not earned it. So we went organic only.
OFF
no more paid ads
Ads rent attention. Stop paying and it’s gone.
20–30
comments a day, by hand
A new account grows by commenting, not by posting.
We’d posted on social media since July, and it barely moved. So I started replying, 20 to 30 times a day, to people asking for a tool like ours. The account started getting noticed.
20–30
comments a day, by hand
A new account grows by commenting, not by posting.
100
comments a day, from one account
We went from 30 a day to 60, then 100. Then our own system, Scout7, took over the writing from how we did it by hand, every 30 minutes on each app. A person still checked every one.
100
comments a day, from one account
3
apps pushed back
Every app has its own rules. People reply to help, not to a pitch.
Hacker News hid our first comment. Reddit moderators removed others. Threads removed one that pitched our tool too soon. So we dropped those two apps, and changed how we talk: help first.
3
apps pushed back
Every app has its own rules. People reply to help, not to a pitch.
125
visitors in one day
On 3 August, 2 people visited. On 19 August, 125 did. Some replies turned straight into signups, and even customers.
125
visitors in one day
46%
fewer visits in 3 days
Comments were doing real work. Time to turn everything up.
A planned test, like switching off the ads. Visits dropped straight away. Switching off ads had cost us nothing we could measure. Switching off comments cost us almost half.
46%
fewer visits in 3 days
Comments were doing real work. Time to turn everything up.
346
posts in 8 days
Fix the quality first. Then turn up the volume.
Each of us posted from our own accounts on at least 3 apps. At first the posts were off-topic and the images were bad. We fixed the quality, then turned up the volume.
346
posts in 8 days
Fix the quality first. Then turn up the volume.
175+
personal reports
How you share it matters as much as what it is.
Each one showed a founder how AI tools like ChatGPT see their product. Nobody opened the first five. So we changed how we shared them, and kept going.
175+
personal reports
How you share it matters as much as what it is.
12,352
emails
Cold emails, and warm ones to people who had signed up before. We built our own email system as we went. Past users started coming back.
12,352
emails
34%
of one batch bounced
Check every address. Test before you send.
We had guessed some addresses, and 34% of one batch bounced. Three days later, our first batch went to spam. We fixed both. Now every address is checked and every email tested.
34%
of one batch bounced
Check every address. Test before you send.
360+
articles and blogs
We wrote for Google, and for ChatGPT and Claude. Slowly, search started sending us more and more people.
360+
articles and blogs
5 → 34
signups a week
We’d had MCP for a while: Scout7 working right inside your AI agent. We put it at the front of our homepage. Then signups started climbing.
5 → 34
signups a week
600
visitors that week
In the week of 20 July, it was 0. Comments, posts, reports, emails and articles, all at once, every day.
600
visitors that week
Every user
asked us for the same thing
“It would be great if someone handled all of that.”
Scout7 could do the work, but it took time to learn, and founders have no time to spare. Running it in bulk also costs far less than one founder alone. We talked to our users.
Every user
asked us for the same thing
“It would be great if someone handled all of that.”
6
guides for your AI agent
Founders don’t need more to learn. They need it done.
We worked with agencies, and built an Academy: guides your AI agent can follow. We wrote a whole library, then cut it to the 6 best. It still wasn’t enough. There was too much to learn.
6
guides for your AI agent
Founders don’t need more to learn. They need it done.
We do the work with Scout7, every day, and you approve what goes out. What we learn running it shapes the product, until one day you won’t need us.
Our first 8 weeks, doing it for ourselves
visitors
signups
places sent them
Channel by channel
Comments and replies
11,832on 8 platforms
About 1 in 8 threads replied. You need volume.
Social posts and reels
1,140posts on 5 platforms, plus 100+ reels
Team accounts on every app all sent visitors.
Articles, blogs and guides
360+articles and blogs, plus 172+ guides
More isn’t better. We kept the strong ones.
Personal reports
175+each made for one prospect
The first five went unopened. How you share matters.
Emails
12,352emails to 7,404 people
Check every address, and test before you send.
Directories and listings
~28listings that sent visitors
Each one sends a few people. Together, they add up.
Paid ads
$889667,395 views, 0 signups
Lots of views. Not one signup.
If you remember 5 things
No single channel did it
3 in 4 visitors came straight to our site after seeing us somewhere. 40+ places sent the rest.
Volume, every day, beats one big push
The best weeks came after weeks of steady work on every app at once.
Ads didn’t do it for us
$889 on ads brought 0 signups. The conversations did the work.
Every app has its own rules
Break one and you’re hidden or removed, often without being told.
Have a person check what goes out
We deleted one of our own comments because it read as AI-written.
Give us your website. Within 2 hours, we email you your Growth Ops Report: where your buyers look for a tool like yours, and if you show up there.
Takes 30 seconds. No call needed.
Growth Ops Report
yoursaas.com
1 of 3
AI tools name you
12
buyers asking this week
5
rivals named instead
Reddit · asked 2 hours ago
“Any tool like yours that actually works for a small team?”
You’re not in the replies
Our reply, ready to post
“What worked for teams your size was…”
Next step
Have us do all of this for you
3 plans, one price a month. You start with $500 and see the work before you pay the rest.
See plans and pricingRather talk first?
20 minutes with the people who did all this. Bring your site, and we’ll tell you where we’d start.
Book a 20-minute call