Traditional sales organizations used the concept of a sales “funnel” to describe the process through which potential customers move, ending up with sales at the end. Winners today have abandoned that way of thinking in favor of building flywheels — business models in which every element reinforces every other.

Ah, the marketing funnel…

Prospective clients go through a predictable set of experiences, students learn in business school marketing classes. It looks like this:

Understanding the funnel helps evaluate sales success indicators. Gail Goodwin, former CEO of small business direct mail provider Constant Contact, said managing the pipeline was key to escaping the sluggish SaaS ramp of death.

Like the funnel concept. To predict how well your business will do, measure how many potential clients are aware of it (awareness) and how many take the next step. If 1,000 people heard about your offering and 10% showed interest, you'd have 100 at that point. If 50% of these people made buyer-like noises, you'd know how many were, etc. It helped model buying trends.

TV, magazine, and radio advertising are pricey for B2C enterprises. Traditional B2B marketing involved armies of sales reps, which was expensive and a barrier to entry.

Cracks in the funnel model

Digital has exposed the funnel's limitations. Hubspot was born at a time when buyers and sellers had huge knowledge asymmetries, according to co-founder Brian Halligan. Those selling a product could use the buyer's lack of information to become a trusted partner.

As the world went digital, getting information and comparing offerings became faster, easier, and cheaper. Buyers didn't need a seller to move through a funnel. Interactions replaced transactions, and the relationship didn't end with a sale.

Instead, buyers and sellers interacted in a constant flow. In many modern models, the sale is midway through the process (particularly true with subscription and software-as-a-service models). Example:

You're creating a winding journey with many touch points, not a funnel (and lots of opportunities for customers to get lost).

From winding journey to flywheel

Beyond this revised view of an interactive customer journey, a company can create what Jim Collins famously called a flywheel. Imagine rolling a heavy disc on its axis. The first few times you roll it, you put in a lot of effort for a small response. The same effort yields faster turns as it gains speed. Over time, the flywheel gains momentum and turns without your help.

Modern digital organizations have created flywheel business models, in which any additional force multiplies throughout the business. The flywheel becomes a force multiplier, according to Collins.

Amazon is a famous flywheel example. Collins explained the concept to Amazon CEO Jeff Bezos at a corporate retreat in 2001. In The Everything Store, Brad Stone describes in his book The Everything Store how he immediately understood Amazon's levers.

The result (drawn on a napkin):

Low prices and a large selection of products attracted customers, while a focus on customer service kept them coming back, increasing traffic. Third-party sellers then increased selection. Low-cost structure supports low-price commitment. It's brilliant! Every wheel turn creates acceleration.

Where from here?

Flywheel over sales funnel! Consider these business terms.

This resulted to obimy's new audience.

Hi! obimy's official account. Here, we'll teach app developers and marketers. In 2022, our downloads increased dramatically, so we'll share what we learned.

obimy is what we call a ‘senseger’. It's a new method to communicate digitally. Instead of text, obimy users connect through senses and moods. Feeling playful? Flirt with your partner, pat a pal, or dump water on a classmate. Each feeling is an interactive animation with vibration. It's a wordless app. App Store and Google Play have obimy.

We had 20,000 users in 2022. Two to five thousand of them opened the app monthly. Our DAU metric was 500.

We have 6 million users after 6 months. 500,000 individuals use obimy daily. obimy was the top lifestyle app this week in the U.S.

And TikTok helped.

TikTok fuels obimys' growth. It's why our app exploded. How and what did we learn? Our Head of Marketing, Anastasia Avramenko, knows.

our actions prior to TikTok

We wanted to achieve product-market fit through organic expansion. Quora, Reddit, Facebook Groups, Facebook Ads, Google Ads, Apple Search Ads, and social media activity were tested. Nothing worked. Our CPI was sometimes $4, so unit economics didn't work.

We studied our markets and made audience hypotheses. We promoted our goods and studied our audience through social media quizzes. Our target demographic was Americans in long-distance relationships. I designed quizzes like Test the Strength of Your Relationship to better understand the user base. After each quiz, we encouraged users to download the app to enhance their connection and bridge the distance.

We got 1,000 responses for $50. This helped us comprehend the audience's grief and coping strategies (aka our rivals). I based action items on answers given. If you can't embrace a loved one, use obimy.

We also tried Facebook and Google ads. From the start, we knew it wouldn't work.

We were desperate to discover a free way to get more users.

Our journey to TikTok

TikTok is a great venue for emerging creators. It also helped reach people. Before obimy, my TikTok videos garnered 12 million views without sponsored promotion.

We had to act. TikTok was required.

I wasn't a TikTok user before obimy. Initially, I uploaded promotional content. Call-to-actions appear strange next to dancing challenges and my money don't jiggle jiggle. I learned TikTok. Watch TikTok for an hour was on my to-do list. What a dream job!

Our most popular movies presented the app alongside text outlining what it does. We started promoting them in Europe and the U.S. and got a 16% CTR and $1 CPI, an improvement over our previous efforts.

Somehow, we were expanding. So we came up with new hypotheses, calls to action, and content.

Four months passed, yet we saw no organic growth.

Russia attacked Ukraine.

Our app aimed to be helpful. For now, we're focusing on our Ukrainian audience. I posted sloppy TikToks illustrating how obimy can help during shelling or air raids.

In two hours, Kostia sent me our visitor count. Our servers crashed.

Initially, we had several thousand daily users. Over 200,000 users joined obimy in a week. They posted obimy videos on TikTok, drawing additional users. We've also resumed U.S. video promotion.

We gained 2,000,000 new members with less than $100 in ads, primarily in the U.S. and U.K.

TikTok helped.

The figures

We were confident we'd chosen the ideal tool for organic growth.

• Over 45 million people have viewed our own videos plus a ton of user-generated content with the hashtag #obimy.

• About 375 thousand people have liked all of our individual videos.

• The number of downloads and the virality of videos are directly correlated.

Where are we now?

TikTok fuels our organic growth. We post 56 videos every week and pay to promote viral content.

We use UGC and influencers. We worked with Universal Music Italy on Eurovision. They offered to promote us through their million-follower TikTok influencers. We thought their followers would improve our audience, but it didn't matter. Integration didn't help us. Users that share obimy videos with their followers can reach several million views, which affects our download rate.

After the dust settled, we determined our key audience was 13-18-year-olds. They want to express themselves, but it's sometimes difficult. We're searching for methods to better engage with our users. We opened a Discord server to discuss anime and video games and gather app and content feedback.

TikTok helps us test product updates and hypotheses. Example: I once thought we might raise MAU by prompting users to add strangers as friends. Instead of asking our team to construct it, I made a TikTok urging users to share invite URLs. Users share links under every video we upload, embracing people worldwide.

Key lessons

Don't direct-sell. TikTok isn't for Instagram, Facebook, or YouTube promo videos. Conventional advertisements don't fit. Most users will swipe up and watch humorous doggos.

More product videos are better. Finally. So what?

Encourage interaction. Tagging friends in comments or making videos with the app promotes it more than any marketing spend.

Be odd and risqué. A user mistakenly sent a French kiss to their mom in one of our most popular videos.

TikTok helps test hypotheses and build your user base. It also helps develop apps. In our upcoming blog, we'll guide you through obimy's design revisions based on TikTok. Follow us on Twitter, Instagram, and TikTok.

In this article, I'll describe a weekend project I started to make something. It was Product Hunt's #1 of the Day, #2 Weekly, and #4 Monthly product.

How did I make Landing Page Checklist so simple? Building and launching took 3 weeks. I worked 3 hours a day max. Weekends were busy.

It's sort of a long story, so scroll to the bottom of the page to see what tools I utilized to create Landing Page Checklist :x ‍

As a matter of fact, it all started with the startups-investments blog; Startup Bulletin, that I started writing in 2018. No, don’t worry, I won’t be going that far behind. The twitter account where I shared the blog posts of this newsletter was inactive for a looong time. I was holding this Twitter account since 2009, I couldn’t bear to destroy it. At the same time, I was thinking how to evaluate this account.

So I looked for a weekend assignment.

Weekend undertaking: Generate business names

Barash and I established a weekend effort to stay current. Building things helped us learn faster.

Simple. Startup Name Generator The utility generated random startup names. After market research for SEO purposes, we dubbed it Business Name Generator.

Backend developer Barash dislikes frontend work. He told me to write frontend code. Chakra UI and Tailwind CSS were recommended.

It was the first time I have heard about Tailwind CSS.

Before this project, I made mobile-web app designs in Sketch and shared them via Zeplin. I can read HTML-CSS or React code, but not write it. I didn't believe myself but followed Barash's advice.

My home page wasn't responsive when I started. Here it was:)

And then... Product Hunt had something I needed. Me-only! A website builder that gives you clean Tailwind CSS code and pre-made web components (like Elementor). Incredible.

I bought it right away because it was so easy to use. Best part: It's not just index.html. It includes all needed files. Like

• postcss.config.js

• README.md

• package.json

• among other things, tailwind.config.js

This is for non-techies.

Tailwind.build; which is Shuffle now, allows you to create and export projects for free (with limited features). You can try it by visiting their website.

After downloading the project, you can edit the text and graphics in Visual Studio (or another text editor). This HTML file can be hosted whenever.

Github is an easy way to host a landing page.

1. your project via Shuffle for export

2. your website's content, edit

3. Create a Gitlab, Github, or Bitbucket account.

4. to Github, upload your project folder.

5. Integrate Vercel with your Github account (or another platform below)

6. Allow them to guide you in steps.

Finally. If you push your code to Github using Github Desktop, you'll do it quickly and easily.

Speaking of; here are some hosting and serverless backend services for web applications and static websites for you host your landing pages for FREE!

I host landingpage.fyi on Vercel but all is fine. You can choose any platform below with peace in mind.

• Vercel

• Render

• Netlify

After connecting your project/repo to Vercel, you don’t have to do anything on Vercel. Vercel updates your live website when you update Github Desktop. Wow!

Tails came out while I was using tailwind.build. Although it's prettier, tailwind.build is more mobile-friendly. I couldn't resist their lovely parts. Tails :)

Tails have several well-designed parts. Some components looked awful on mobile, but this bug helped me understand Tailwind CSS.

Unlike Shuffle, Tails does not include files when you export such as config.js, main.js, README.md. It just gives you the HTML code. Suffle.dev is a bit ahead in this regard and with mobile-friendly blocks if you ask me. Of course, I took advantage of both.

creativebusinessnames.co is inactive, but I'll leave a deployment link :)

Adam Wathan's YouTube videos and Tailwind's official literature helped me, but I couldn't have done it without Tails and Shuffle. These tools helped me make landing pages. I shouldn't have started over.

So began my Tailwind CSS adventure. I didn't build landingpage. I didn't plan it to be this long; sorry.

I learnt a lot while I was playing around with Shuffle and Tails Builders.

Long story short I built landingpage.fyi with the help of these tools;

Learning, building, and distribution

• tailwindcss.com/docs

• Shuffle (Started with a Shuffle Template)

• Tails (Used components from here)

• Sketch (to handle icons, logos, and .svg’s)

• metatags.io (Auto Generator Meta Tags)

• Vercel (Hosting)

• Github Desktop (Pushing code to Github -super easy-)

• Visual Studio Code (Edit my code)

• Mailerlite (Capture Emails)

• Google Analytics

• Jarvis / Conversion.ai (%90 of the text on website written by AI 😇 )

• CookieHub (Consent Management)

That's all. A few things:

The Outcome

.fyi Domain: Why?

I'm often asked this.

I don't know, but I wanted to include the landing page term. Popular TLDs are gone. I saw my alternatives. brief and catchy.

CSS Tailwind Resources

I'll share project resources like Tails and Shuffle.

• Wicked Templates

• TailwindCSS Buttons

• Blocks | Wicked Templates

• nine4

• TUK — Tailwind UI KIT

• HyperUI

• Lofi UI

• Beginner Tailwind (I lately enrolled in this course but haven’t completed it yet.)

Thanks for reading my blog's first post. Please share if you like it.

Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.

James Howells lost 8,000 bitcoins. He has $11 million to get them back.

His life altered when he threw out an iPhone-sized hard drive.

Howells, from the city of Newport in southern Wales, had two identical laptop hard drives squirreled away in a drawer in 2013. One was blank; the other had 8,000 bitcoins, currently worth around $181 million.

He wanted to toss out the blank one, but the drive containing the Bitcoin went to the dump.

He's determined to reclaim his 2009 stash.

Howells, 36, wants to arrange a high-tech treasure hunt for bitcoins. He can't enter the landfill.

Newport's city council has rebuffed Howells' requests to dig for his hard drive for almost a decade, stating it would be expensive and environmentally destructive.

I got an early look at his $11 million idea to search 110,000 tons of trash. He expects submitting it to the council would convince it to let him recover the hard disk.

110,000 tons of trash, 1 hard drive

Finding a hard disk among heaps of trash may seem Herculean.

Former IT worker Howells claims it's possible with human sorters, robot dogs, and an AI-powered computer taught to find hard drives on a conveyor belt.

His idea has two versions, depending on how much of the landfill he can search.

His most elaborate solution would take three years and cost $11 million to sort 100,000 metric tons of waste. Scaled-down version costs $6 million and takes 18 months.

He's created a team of eight professionals in AI-powered sorting, landfill excavation, garbage management, and data extraction, including one who recovered Columbia's black box data.

The specialists and their companies would be paid a bonus if they successfully recovered the bitcoin stash.

Howells: "We're trying to commercialize this project."

Howells claimed rubbish would be dug up by machines and sorted near the landfill.

Human pickers and a Max-AI machine would sort it. The machine resembles a scanner on a conveyor belt.

Remi Le Grand of Max-AI told us it will train AI to recognize Howells-like hard drives. A robot arm would select candidates.

Howells has added security charges to his scheme because he fears people would steal the hard drive.

He's budgeted for 24-hour CCTV cameras and two robotic "Spot" canines from Boston Dynamics that would patrol at night and look for his hard drive by day.

Howells said his crew met in May at the Celtic Manor Resort outside Newport for a pitch rehearsal.

Richard Hammond's narrative swings from banal to epic.

Richard Hammond filmed the meeting and created a YouTube documentary on Howells.

Hammond said of Howells' squad, "They're committed and believe in him and the idea."

Hammond: "It goes from banal to gigantic." "If I were in his position, I wouldn't have the strength to answer the door."

Howells said trash would be cleaned and repurposed after excavation. Reburying the rest.

"We won't pollute," he declared. "We aim to make everything better."

After the project is finished, he hopes to develop a solar or wind farm on the dump site. The council is unlikely to accept his vision soon.

A council representative told us, "Mr. Howells can't convince us of anything." "His suggestions constitute a significant ecological danger, which we can't tolerate and are forbidden by our permit."

Will the recovered hard drive work?

The "platter" is a glass or metal disc that holds the hard drive's data. Howells estimates 80% to 90% of the data will be recoverable if the platter isn't damaged.

Phil Bridge, a data-recovery expert who consulted Howells, confirmed these numbers.

If the platter is broken, Bridge adds, data recovery is unlikely.

Bridge says he was intrigued by the proposal. "It's an intriguing case," he added. Helping him get it back and proving everyone incorrect would be a great success story.

Who'd pay?

Swiss and German venture investors Hanspeter Jaberg and Karl Wendeborn told us they would fund the project if Howells received council permission.

Jaberg: "It's a needle in a haystack and a high-risk investment."

Howells said he had no contract with potential backers but had discussed the proposal in Zoom meetings. "Until Newport City Council gives me something in writing, I can't commit," he added.

Suppose he finds the bitcoins.

Howells said he would keep 30% of the data, worth $54 million, if he could retrieve it.

A third would go to the recovery team, 30% to investors, and the remainder to local purposes, including gifting £50 ($61) in bitcoin to each of Newport's 150,000 citizens.

Howells said he opted to spend extra money on "professional firms" to help convince the council.

What if the council doesn't approve?

If Howells can't win the council's support, he'll sue, claiming its actions constitute a "illegal embargo" on the hard drive. "I've avoided that path because I didn't want to cause complications," he stated. I wanted to cooperate with Newport's council.

Howells never met with the council face-to-face. He mentioned he had a 20-minute Zoom meeting in May 2021 but thought his new business strategy would help.

He met with Jessica Morden on June 24. Morden's office confirmed meeting.

After telling the council about his proposal, he can only wait. "I've never been happier," he said. This is our most professional operation, with the best employees.

The "crypto proponent" buys bitcoin every month and sells it for cash.

Howells tries not to think about what he'd do with his part of the money if the hard disk is found functional. "Otherwise, you'll go mad," he added.

This post is a summary. Read the full article here.

SaaS spending is 40/40/20. 40% of operating expenses should be R&D, 40% sales and marketing, and 20% G&A. We wanted to see the statistics behind the rules of thumb. Since October 2017, 73 SaaS startups have gone public. Perhaps the rule of thumb should be 30/50/20. The data is below.

30/50/20. R&D accounts for 26% of opex, sales and marketing 48%, and G&A 22%. We think R&D/S&M/G&A should be 30/50/20.

There are outliers. There are exceptions to rules of thumb. Dropbox spent 45% on R&D whereas Zoom spent 13%. Zoom spent 73% on S&M, Dropbox 37%, and Bill.com 28%. Snowflake spent 130% of revenue on S&M, while their EBITDA margin is -192%.

G&A shouldn't stand out. Minimize G&A spending. Priorities should be product development and sales. Cloudflare, Sendgrid, Snowflake, and Palantir spend 36%, 34%, 37%, and 43% on G&A.

Another myth is that COGS is 20% of revenue. Median and averages are 29%.

Where is the profitability? Data-driven operating income calculations were simplified (Revenue COGS R&D S&M G&A). 20 of 73 IPO businesses reported operational income. Median and average operating income margins are -21% and -27%.

As long as you're growing fast, have outstanding retention, and marquee clients, you can burn cash since recurring income that doesn't churn is a valuable annuity.

The data was compelling overall. 30/50/20 is the new 40/40/20 for more established SaaS enterprises, unprofitability is alright as long as your business is expanding, and COGS can be somewhat more than 20% of revenue.