[Journey] Scaling POD automation

madacot233

Junior Member
Joined
Aug 20, 2021
Messages
157
Reaction score
162
I'm currently trying to hit 50k designs across the top 4 POD sites (RedBubble, Zazzle, Teepublic, Society6). My current uploaded design amount across all of the platforms is 2k designs each.

Generation Process
  • I have developed modular software to generate POD merch.
  • There is currently 4 different modules, I will be adding more and modifying existing ones based of sales preformance of each module
  • Each module can have either text content (GPT3 generated) and/or images (using licensed graphics OR AI generated images (Style Transfer)
Niche Finding / SEO Research
  • I don't have a SEO / Niche finding process, everything will be done manually for now
Upload Process
  • I have built an upload manager site which handles the following processes:
  • Account management (splits uploads between accounts to bypass upload limit / split accounts into niche)
  • Upload management
  • Merch automatic tagging
  • Merch automatic description (GPT3)
I will be updating this once a month, currently need to create more accounts for Redbubble and Teepublic as the upload limits are slowing down the process.
 
How do you deal to avoid suspensions, on this sites for dmca/copyright thing
Redbubble blocked my account twice just on uploading 4 designs

All designs are unique using licensed graphics, and I don't upload any designs that might infringe copyright.

How much of sales are you currently averaging

I uploaded 600 bad designs on February 2021, those brought in ~$100 in sales. Now the designs are of a much higher quality, with better tagging and more scale. So we shall see.
 
Are your all designs related to a specific niche or is it like a general store?
 
You're displaying quite an army catalogue.
With those numbers it's almost imposible to fail on sales.
 
Are your all designs related to a specific niche or is it like a general store?
All the stores have pretty random designs for now, in the future it might be worth separating stores into niches. When I find a niche, I can easily generate more than enough designs to dominate the search results.

Good luck, sounds like a cool system. Watch you don't get caught out with paypal payments to the same or a banned email address.
Thanks, using general stealth techniques for now. They are not too fussed about multi accounts unless you abuse copyright.

if you can tell, how much is the average cost per design and where from you procuring them
It's in the generation heading. The daily running costs at the moment are:

$0.50 for captcha solving
$0.50 AWS data transfer fees.

Due to how cheap it is to run/scale I would be happy to run at a loss for ~2-3 years before giving up
 
Really interesting idea.
Are you doing text designs mainly?
 
Do you also use AI for titles and descriptions? How do you check for trademarked words that the AI generates for your designs?

Can you show a few design samples that it generates?
 
There's a text only module, text + image, image and one more module for now.



They already have, we shall see in a year.
You will probably make some sales here and there, but in the long run I don't think the effort is worth it. The fashion industry is fierce and you won't be able to compete with the top designers if all you do is scrape random wording and make a design out of it.
 
Very interested in your journey, specifically your bot. I've been doing some POD myself for a few years and built some python bots that automatically upload them for me but never thought to go as far as you have with it. I run mine manually after getting some images together and it uploads them.
 
How do you plan on promoting your designs?
 
I'm curious how you created the design. Do you hire a designer or do you use a tool to clone and separate the background from another platform. Hope your project is successful.
 
You will probably make some sales here and there, but in the long run I don't think the effort is worth it. The fashion industry is fierce and you won't be able to compete with the top designers if all you do is scrape random wording and make a it design out of it.

It depends on perspective, people don't buy from POD sites because they strongly identify with the niche which they have the desire to show off.

Very interested in your journey, specifically your bot. I've been doing some POD myself for a few years and built some python bots that automatically upload them for me but never thought to go as far as you have with it. I run mine manually after getting some images together and it uploads them.
I still have to do niche research manually and review designs afterwards, but that's something to automate later
Good luck OP.
Thanks
good luck ,dope seeing another pod/ecom person ...hit them goals
Thanks :)
How do you plan on promoting your designs?
Current strategy requires no promotion, I'm targeting niches with none / low competition and relying on traffic from the platforms itself + traffic from search engines.
I'm curious how you created the design. Do you hire a designer or do you use a tool to clone and separate the background from another platform. Hope your project is successful.
Read above, I've described the process.

THE UPDATE
Not exactly a month, but I'm not going to review this project again until at least a month.

Achieved some sales, which is a good sign. Especially as the designs are under 1-month-old and have not had time to rank yet.

Upload progress:

yLvJJMf


Ignore the total designs vs total queued, I had to purge the design's database a while ago while testing.
 
Back
Top