Forem: Igor Boky

This is how I built an analytic tool for Digital Products

Igor Boky — Wed, 23 Oct 2024 11:34:26 +0000

Hey guys,

Recently, I’ve fallen in love with data analytics of all kinds.

As I was looking for a side hustle, I explored different options and stumbled upon the world of digital products. Platforms like Gumroad, Ko-fi, and others dominate this market.

But I hit a roadblock right at the beginning: there are hundreds of products.

So, which one should I sell?

After researching various articles and platforms, I realized that there wasn't enough solid data to make a well-informed decision.

This led me to start tracking trends on one of the biggest digital product marketplaces: Gumroad.

v0.1

I started with a simple JavaScript script that performs basic API requests to the publicly available Gumroad API. You can easily see these requests in the browser's Networking tab:
https://gumroad.com/products/search?&tags%5B%5D=vrchat&from=10

It was a good start. I decided to track the 10 most popular tags and fetch data daily.

Tech stack I used:

Node.js
Axios + axios-retry + axios-rate-limit
FS to write to the file system
AI tools to speed up coding

It worked out well. I collected data for several days in a row and found it interesting enough to expand further. I also received support from the community on X, so I decided to take the next step.

v0.2

I realized that knowing the most popular tags wouldn't be enough — the real potential lies in identifying the tags that are growing.

In my next iteration, I collected sub-tags from the most popular niches and gathered over 5,000 tags this way.

This generated more than 20,000 rows of data daily, so I needed a storage solution.

Since it's still possible to host a free cluster on MongoDB Atlas and MongoDB pairs well with JavaScript, I chose it as my storage option. Now, the data is stored, and I can run various analytics.

Another challenge was the time required to collect the data.
It initially took five hours. That's when I learned about proxies. After setting up several proxy servers to process the data in parallel, the collection time dropped from five hours to just 30 minutes.

v0.2.1

I was running all the scripts manually, so I decided to automate the process.

I hosted the source code on a remote server on Hetzner and set up cron jobs via crontab.

0 18 * * * /usr/bin/node index.js collect-data
0 21 * * * /usr/bin/node index.js collect-totals

The first job collects the data, and the second one aggregates it for a summary. It removes duplicates and calculates the growth compared to the previous day.

const [mode = 'collect-data'] = process.argv.slice(2);
if (mode === 'collect-data') {
    return collectData();
}
if (mode === 'collect-totals') {
    return collectTotals();
}

v0.3

I decided to create a simple UI for this project and share it publicly. This is how gumroadtrends.com was launched.

I used:

VueJS for the UI
Bootstrap for the CSS framework
Chart.js for visualizing the data
GA4 for usage analytics
I also used Hetzner and PM2 for hosting since I already had a server for hosting all my products.

Here’s what the UI looks like:

v0.4

It’s hard to understand trends with only seven days of data, so I plan to continue collecting data on autopilot and revisit it after a few weeks for deeper insights.

Final words

I'm happy to share anything I used to build this product. This is my small contribution to the #buildinpublic community that I’m part of.

Good luck with your projects!

Feel free to get in touch if you’re interested about the topic. Follow me on X

How to build AI solution capable to build a e-commerce store

Igor Boky — Tue, 11 Jul 2023 18:32:14 +0000

AI today is everywhere, while big corporates fight for a place under the shining sun, we can use AI for quite a practical purpose.

One issue every person faces once is building something from scratch. The same happens for code. If you join an existing project you can just repeat and change. You don't need to build a thing from scratch quite often.

The same happens every day in other areas users act: would it be a marketplace product description, or bio for your Twitter profile, it's all could be generated via AI today and it would be in the best English ever and it would be in the tone you want it to be.

What is the goal of this post?

This post is an evolution of another one I did a month ago. Now we do the next step and talk more about system prompts, which we usually provide to AI as a general context. We will talk about ChatGPT API, but the idea is the same for other APIs.

So this time we will go deeper and make a small node.js app that allows us to build a full e-commerce store!

Ready to go?

The plan

It's pretty simple, we need to prepare all the information, collect it together and share it with the world.

Generate Products + Images (done in prev post)
Generate Categories
Distribute products among categories
Present the result as a single JSON.

Preparations

Please start with this guide and repeat all steps there

On top of it, there is one more function to handle ChatGPT API responses in a better way from this post. It allows to retry requests to API in case it responses with not a pure JSON.

async function createChatCompletition(prompt, retryAttempts = 2, currentAttempt = 1) {
  const completion = await openai.createChatCompletion({
    model: 'gpt-3.5-turbo',
    messages: [{ "role": "user", "content": prompt }]
  });

  const output = completion.data.choices[0].message.content;
  try {
    return JSON.parse(output);
  }
  catch (err) {
    if (retryAttempts > currentAttempt) {
      return createChatCompletition(prompt, retryAttempts, currentAttempt + 1)
    }

    return throw 'Not a json from API';
  }
}

Every Product Needs a Category

Let's start by creating a list of categories and then distributing products generated last time between them. To create categories we need a simple method

function getCategories(info, count) {
  const propmt = `Generate array of ${count} objects where each object is a unique product category of marketplace which is described as ${info},
  use following structure  "{name: string}",
  respond only with JSON`;

  return createChatCompletition(propmt);
}

That's it, now we have to distribute existing products. We will write a simple method working like this

  products.forEach((product, index) => {
    product.category = categories[index] ? categories[index]["name"] : categories[0];
  })

It doesn't guarantee that all categories would be connected to at least one product, you can experiment here to make it better, and you can also as ChatGPT to do that distribution for you.

Let's collect things together

So far we have products and categories, what is a missing puzzle here? Let's think about a company that could sell this. Let's ask API to create a nice name for a company that could sell these beautiful products

function getCompany(info) {
  const propmt = `Use ${info} to generate information about the company: name, slug, slogan, description 2 sentences. Please respond with JSON in format {name, slug, description}. Send only JSON`;
  return createChatCompletition(propmt);
}

So the final result would look like this.

async function execute() {
   const topic = 'Toy for kids';
   const productNumbers = 4;

   const promises = [];
   for (let i = 0; i < productNumbers; ++i) {
     promises.push(getProduct(topic))
   }

   const products = await Promise.all(promises);
   console.log('PRODUCTS', products);
   const categories = await getCategories(topic, 3);
   console.log('CATEGORIES', categories);
   addCategoriesToProducts(products, categories);
   const companyInfo = await getCompanyInfo(topic);

   console.log({
    ...companyInfo,
    products,
    categories,
   });
}

execute();

What could be the next step or many?!

If we can imagine that we continue in that direction we can come up with a nice tool, we can call AI Marketplace Builder. Fleexy team just released one to test this out.

At the end

We did one more step ahead to build something new, just imagine we didn't have this possibility a year ago, now it's time for such creative solutions to become real. I would continue experimenting with AI on such practical aspects to post further.

Gift me your likes/stars, it will help me to stay motivated!

Follow me on Twitter
Connect on Linkedin

How to Ask AI to return JSON in the Right Way

Igor Boky — Thu, 06 Jul 2023 12:53:48 +0000

Quite often when you request AI to send you a JSON, it responds with anything but not the JSON, or JSON is just a part of it.

There are some ways to solve this

1. Ask to return JSON at the end

For example your prompt looks like

Generate products that could be described as "Headphone", use the following structure "[{name: string, price: number}]

To make AI answer with JSON with a higher chance add

.... Respond with an array of a flat objects in JSON

2. Split prompts and requests to API

Let's say you need to generate products for your marketplace solution. Instead of asking AI to generate the whole info like in the example above. It could be better to do it in 2+ steps.

Generate Names:
Generate Names Generate ${count} different real product names for a product category "${info}", respond with a JSON {names: string[]}
Generate Single Product for each product name from the previous request: Generate a product description for a product that could be described as "${productName}", use the following structure "{name: string, description: string, price: number }", respond with a single product **as a flat** JSON

3. Even now you have a problem? Let's retry.

If you have an issue even if you ask specifically and it happens then the only solution I know is repositioning on request. This is the method I use for this.



async function createChatCompletition(prompt, retryAttempts = 2, currentAttempt = 1) {
  const completion = await openai.createChatCompletion({
    model: 'gpt-3.5-turbo',
    messages: [{ "role": "user", "content": prompt }]
  });

  const output = completion.data.choices[0].message.content;
  try {
    return JSON.parse(output);
  }
  catch (err) {
    if (retryAttempts > currentAttempt) {
      return createChatCompletition(prompt, retryAttempts, currentAttempt + 1)
    }

    return {};
  }
}

Some line-by-line explanation

Create a completion and wait for response

const completion = await openai.createChatCompletion
Try to parse



  try {
    return JSON.parse(output);
  }

Repeat if the response is not a JSON



catch (err) {

    if (retryAttempts > currentAttempt) {

      return createChatCompletition(prompt, retryAttempts, currentAttempt + 1)

    }

return {};



}

Here we are

Now you can fight with an AI's random responses which sometimes happens. Use it!

Please share in the comments your solutions to solve this issue.

Follow me on Twitter
Connect on Linkedin

Launch Fast to Learn Fast

Igor Boky — Wed, 05 Jul 2023 13:30:43 +0000

Launching a startup and moving quickly in business are essential for success. The speed at which you operate can determine your fate in a competitive market. Here are four reasons why speed is crucial in business and why you should consider launching your startup faster.

Stay Ahead of Competitors

In a fast-paced business world, keeping up with the competition is a constant challenge. The market landscape is ever-changing, and if you can't move quickly, your competitors will leave you behind. Take the example of Blockbuster, which failed to adapt to the rise of streaming services like Netflix.

While Blockbuster was hesitant to embrace the digital revolution, Netflix revolutionized the industry and gained a significant competitive advantage. To stay at the forefront of your industry, you need to be faster than anyone else. As Jeff Lerner, CEO of Xurli, wisely advises,

"Accelerate until you're at the front and move fast to stay there."

Meet Customer Expectations

In today's society, people have developed a culture of instant gratification. They expect immediate access to products, services, and information.

By launching your product or service quickly, you can satisfy these expectations and capitalize on the market demand. Look at smartphone releases as an example—they come out every year to meet consumer demand for the latest technology. If you can't deliver promptly, customers won't hesitate to switch to a competitor who can meet their needs. As Luke Kanies, CEO of Puppet Labs, warns,

"You need to leverage technology to innovate and evolve, or your competitors will crush you in the market."

Launching faster allows you to be responsive to customer expectations and maintain a competitive edge.

Evolve Faster Through Learning

Moving quickly as a company leads to rapid evolution and growth. When you embrace speed and agility, you create an environment where learning becomes an integral part of your business strategy.

The more you practice speed, the easier it becomes to maintain it.

By constantly pushing boundaries, experimenting, and learning from both successes and failures, you can adapt to changing market dynamics and customer preferences. This adaptability enables you to make informed decisions, optimize processes, and deliver innovative solutions. As Savina Singh, CEO of Adog, aptly points out,

"The speed with which an entrepreneur can move determines their ability to balance scalable infrastructure needs and low overhead costs."

By embracing speed, you can surpass expectations, stand out in the market, and drive continuous improvement.

Foster a Culture of Speed

Once you learn to move quickly, a culture of speed develops within your company. This culture sets high standards for efficiency, innovation, and adaptability. When speed becomes ingrained in your organization's DNA, it permeates every aspect of your operations. Employees become more agile, proactive, and solution-oriented.

Decision-making processes become streamlined, enabling faster responses to market opportunities and challenges. With everyone and everything moving faster, you gain a competitive edge and delight your customers with swift and effective solutions. In entrepreneurship, a slow culture hinders progress, stifles innovation, and puts your business at risk. On the other hand, a culture of speed drives innovation forward, positions you as an industry leader, and ensures your long-term success.

Achieve Perfection

Launching your startup faster can actually help you achieve perfection in your product or service. Many founders hesitate to launch due to perfectionism, wanting to ensure everything is flawless before presenting it to the market.

However, delaying your launch indefinitely can be detrimental to your business. By launching quickly, you can gain valuable feedback from early adopters and users. This feedback serves as a valuable source of information that can help you identify areas for improvement and make necessary adjustments.

The iterative process of receiving feedback, implementing changes, and re-launching allows you to refine your product or service gradually, moving closer to perfection over time. As Paul Graham, co-founder of Y Combinator, wisely says,

"It takes an effort of will to push through and get something released to users."

Launching quickly allows you to learn and adapt based on real-world user experiences, leading to a more refined and successful offering.

Understanding the Value of Your Time

As a founder, your time is one of your most valuable resources. Time is limited, and you can only accomplish so much on your own.

Spending too long in the development phase of your startup may not be the best use of your time. By raising funds earlier in the startup journey, you can allocate resources to hire talented individuals who can contribute to the growth and development of your company.

Delegating tasks and responsibilities to a capable team allows you, as a founder, to focus on strategic decision-making, networking, and other high-level activities that drive the success of your startup.

By maximizing the value of your time through effective delegation and resource allocation, you can accelerate progress and achieve your goals more efficiently.

Valuations for Early-Stage Startups

Early-stage startups often face challenges when it comes to valuations, as they typically lack substantial financial data. Valuations for startups are based on negotiations, market trends, and heuristics rather than fixed formulas. Investors assess a startup's potential by considering various factors such as team composition, market size, traction, and growth prospects.

To increase your startup's valuation, you need to demonstrate progress and milestones achieved.

This progress can be accelerated through securing funding, which allows you to invest in product development, marketing, customer acquisition, and other growth initiatives.

By reaching key milestones and showing investors tangible progress, you can move your startup into a higher valuation range, attracting more attention and potential investment opportunities.

Stages of Development to Launch

Successfully launching your startup involves going through several stages of development.

It starts with a "bad" minimum viable product (MVP), which is a basic version of your offering that provides some value to users. It's important to note that while the initial MVP may not be perfect, it should still offer something meaningful to users to avoid harming your reputation. Gathering feedback from early users and iterating based on their input, you can then move on to a more refined MVP.

This second iteration of the product should offer even more value to your target users, incorporating improvements and addressing pain points identified during the initial launch. Finally, the goal is to reach the stage of a minimum desirable product (MDP).

At this stage, your product has a higher chance of meeting and exceeding customer expectations, gaining wider acceptance in the market. Following these stages allows you to incrementally improve your offering, validate market demand, and build relationships with investors as you progress towards a more polished and desirable product.

AI Era

We are gradually but actively entering the era of AI, where things work differently than before. With AI providing us ample freedom to create, we can revolutionize the traditional rules of product launch and make the process even faster. An exemplary initiative in this regard is the Marketsy app developed by the Fleexy team. They have created an e-commerce store builder that enables you to set up your own store in a matter of seconds. Best of all, it's currently available for free!

It shows that AI capabilities combined with nice technologies could deliver more and much faster.

Brief summary

In conclusion, launching a startup faster is crucial for success in today's competitive business landscape. By moving quickly, you can stay ahead of competitors, meet customer expectations, foster a culture of speed, and accelerate your company's evolution. Launching quickly allows you to gain valuable feedback, improve your product, and achieve perfection. It also helps you maximize the value of your time as a founder and increase your startup's valuation.

Remember, launching a startup faster is a combination of working smart and hard, making the best use of your time, and seizing opportunities for growth.

So don't hesitate, take action, and launch your startup faster to pave the way for success

Follow me on Twitter
Connect on Linkedin

Niche Marketplaces for Newbies

Igor Boky — Mon, 03 Jul 2023 20:04:34 +0000

When it comes to entrepreneurship, there are two paths you can take. You have the option to serve a wide audience or target a specific niche market. If you choose to cater to the masses, you might find yourself creating numerous products or serving multiple industries. However, it can be challenging to establish yourself as the top player in each area.

On the other hand, if you decide to focus on a niche market, things work a bit differently. Niche entrepreneurs specialize in creating unique products or catering to specific audiences. By narrowing your focus, you not only streamline your efforts but also build a strong reputation and become the trusted brand for your target customers.

Let's talk about niche marketplaces, shall we?

So, what exactly are niche marketplaces? Well, they're specialized markets that cater to a specific group of shoppers who have similar demographics, needs, or preferences. It's like having a marketplace that focuses on a particular category of products. Take Airbnb, for example. It's a niche platform that connects people looking for short-term rentals.

But here's the twist: a niche marketplace can also have a variety of product categories, but they all share a common attribute. A great example of this is Etsy, where you can find all sorts of unique and handmade goodies.

On the flip side, you have general marketplaces like Amazon that take a catch-all approach. They offer a wide range of products to appeal to as many different consumers as possible. And let me tell you, these general marketplaces are raking in customers! In 2021 alone, Amazon's annual net sales revenue hit a mind-boggling $469 billion. That's a lot of zeros!

But here's the thing: with the explosion of eCommerce, everyone and their grandma wants a piece of that pie. There's fierce competition among sellers on general marketplaces like Amazon. In fact, approximately 3,000 new sellers are joining the Amazon bandwagon every single day.

That's where niche platforms come in. Despite having a more focused scope, they're gaining popularity because they bring together specific groups of sellers and buyers who are looking for each other. It's all about finding your niche, my friend.

Unveiling the Advantages of Niche Marketplaces

Niche marketplaces bring some deep-rooted, long-term advantages to the table. They came to be because people faced a common problem in our digitally connected world: the general market was so vast that it became a challenge for consumers to find the specific products they were after. And even if they did stumble upon something on a general marketplace, there was no guarantee it would meet their quality standards or requirements.

That's where niche marketplaces stepped in to save the day. They cater to a specific group of customers and carefully vet sellers to ensure the products meet certain standards. This creates an opportunity for businesses to focus their efforts, fine-tune their marketing strategies, and build a loyal customer base.

You know what they say in the marketing world:

"If you are selling to everyone, you are selling to no one."

Wise words, indeed!

So, what are the benefits of using niche marketplaces? Let's break it down:

Limited competition: Unlike general marketplaces, niche platforms usually have fewer vendors to compete with. It's like a smaller, cosier marketplace.
Targeted audience access: The right niche marketplace will connect you with your desired audience. They've already done the hard work of building a strong following, which you can tap into.
Credibility: Niche marketplaces often have stricter screening processes compared to regular ones. This gives customers more confidence in the products sold there. Trust is key!
Reduced marketing costs: By focusing on a niche, you save precious time and money by specifically targeting your ideal customers. No more marketing to people who won't be interested.
Brand visibility and loyalty: You can leverage the existing brand of the marketplace itself to boost your business's visibility. Plus, once you build a customer base within that niche, their loyalty will be strong.

So, there you have it! Niche marketplaces bring a range of benefits that can give your business an edge.

Discover the Coolest Niche Online Marketplaces!

In this digital age, online marketplaces have taken over the scene, making shopping a breeze. You can browse and buy stuff from the comfort of your couch, anytime you want. No need to leave the house or deal with crowded stores!

Now, the internet is bursting with niche online marketplaces that cater to specific interests and needs. But hey, which ones are the absolute best? Let's find out!

1. Etsy

Etsy is like the ultimate treasure trove of unique and handmade goodies. It's not just your average online marketplace—it's a haven for arts, crafts, and all things creative. From stunning custom jewellery to printable digital art, Etsy has it all. What sets it apart is the community of talented sellers and the emphasis on quality and craftsmanship. Plus, it's super easy to set up your own store and start selling your creations. If you're on the hunt for something special and one-of-a-kind, Etsy is the place to be.

2. Chewy

Calling all pet parents! Chewy is here to make your life easier. This niche marketplace is dedicated to providing affordable pet food, treats, healthcare products, and everything else your furry friends need. Whether you're looking for dog food, cat toys, or even prescription medications, Chewy has got you covered. The best part? They understand the love we have for our pets and go the extra mile to offer top-notch customer service. With Chewy, taking care of your beloved pets has never been more convenient.

3. Teachable

If you have a passion for teaching and sharing your expertise, Teachable is your new best friend. This online marketplace is like a virtual school where you can create and sell your own online courses. Whether you're a business guru, a fitness enthusiast, or a cooking connoisseur, Teachable lets you monetize your knowledge and connect with eager learners. The platform takes care of the technical stuff, leaving you free to focus on creating engaging and informative courses. It's a win-win for both students and instructors!

4. Fiverr

Are you a freelancer looking for a big break? Fiverr is the place to be. This platform connects talented service providers with clients seeking their skills. From graphic design to web development, you can offer your specialized services and build a thriving freelance business. Fiverr helps you stand out from the crowd with a user-friendly profile and a secure payment system. Plus, they have a massive user base, so you'll have plenty of opportunities to showcase your talents and land exciting gigs.

5. Airbnb

When it comes to unique accommodations and unforgettable travel experiences, Airbnb takes the cake. It started as a platform for couch surfers, but now it's evolved into a worldwide network of hosts offering all sorts of accommodations. Whether you're looking for a cosy spare room or a luxurious vacation rental, Airbnb has got you covered. It's a fantastic way to connect with locals, immerse yourself in different cultures, and create lifelong memories. So, if you're planning your next adventure, why not give Airbnb a try?

6. Booking.com

Booking.com is the ultimate go-to when it comes to finding the perfect place to stay during your travels. Whether you're looking for a hotel room, a private home, or a hosted stay, Booking.com has an extensive selection to choose from. With its user-friendly platform and trusted rating system, you can easily book your ideal accommodation and have peace of mind knowing you're in good hands. It's the ultimate travel companion for anyone seeking comfort, convenience, and a memorable stay.

These niche marketplaces offer something unique and cater to specific needs and interests. So, the next time you're on the hunt for something special or planning your next adventure, give these awesome platforms a try!

7. Fleexy’s Marketsy

All-in-one tool for service or product marketplaces. It has the foundation of a strong feature to start a marketplace immediately. But if you need any kind of customization or adjustment, it could be done with ease by the micro apps foundation. Also, infinite updates will improve your marketplace and your user's experiences during the whole journey.

You can run a marketplace in any niche. Or even a few to check more than one idea. Also, Fleexy’s Marketsy offers an AI marketplace generator to simplify your takeoff.

Discover Your Niche and Thrive

In today's bustling world of eCommerce, finding your niche marketplaces can be the secret ingredient to scaling up your business. It's all about honing in on the right audience and delivering exactly what they're looking for.

Competition is fierce out there, but fear not! By focusing your efforts on a specific niche, you can stand out from the crowd and attract those loyal customers who will stick with you for the long haul.

Think of it as a recipe for success: combine the power of niche marketplaces, targeted marketing strategies, and a business model tailored to your niche. With this winning formula, you'll be well on your way to reaching new heights and achieving your business goals.

Follow me on Twitter
Connect on Linkedin

What is Wrong with Classical Custom Development

Igor Boky — Tue, 27 Jun 2023 19:06:13 +0000

It is best to test the waters before taking a dip.

The minimum viable product helps you do exactly that. It enables you to develop a product with basic features, test it on the ground, make changes according to the market analysis, and finally launch the final product.

Custom Development Costs Composition

The cost of custom development of MVP varies and starts usually at ~$15.000.

And it's just the start, so proper product development is much higher. But let's what that pricing includes and what it depends on:

Application Type & Technologies

First, the cost of developing an MVP is mostly determined by the type of app, followed by its complexity and features. In other words, the lower the complications, the lower the expense of creating an app, and vice versa.

At the same time it is super dependent on the platform you build the MVP for, would it be a mobile app or Web, I probably you do a desktop application

Designing

The design part may add up to a hefty portion of MVP app development. In truth, how you choose wireframes, UI, and hiring experts all contribute to the high app/website design cost. This is why you should plan and estimate the development cost ahead of time to know exactly how much it costs to build an MVP.

The price of design depends based on the location, global rates vary from 25$/hour (in Asia region) to 150$/hour (USA)

Development team

Custom development you can do only with a team of skilled specialists. Depending on the technology you need a team of 2-4 specialists.

The standard setup for a team is:

Developer (single full-stack or two: frontend + backend developer), at least 25$/hour
Quality Assurance (manual), at least 30$/hour
Project Manager (Often playing the role of Business Analyst), at least 35$/hour
Optionally it could be a Business Analyst, usually joining the MVP part-time, at least 35$/hour

So for a team of 4 on outsourcing you would pay an average of $20,000/month, you can make this amount smaller if you go for freelancers, but it will be much higher if you build an in-house team.

Approximate Totals for 2 months MVP

It's possible to deliver an MVP faster, then it will be $15.000 - $20.000

Post MVP Release Activities

You will need to spend money on brand awareness and sales. Even if you don't invest money in paid ad campaigns, you'll need someone to handle your social media, online presence, and other promotional activities.

Why is marketing crucial? A robust marketing strategy can help you create buzz around your product, attract early users and first clients, and help the company become a name.

You also need to think about maintenance which is typically 20% of the initial MVP development budget. Remember, it is entirely dependent on the team’s location and product complexity.

All postrelease cost could be various from $1.000 to $15.000 per month...

Classical Custom Development Totals

As we can see marketing and sales are a bit more on top of the development costs, so it's important to consider this when you join the development process. The cheapest option is $15,000 and it could go easily go above $100,000

What is wrong with the custom development approach?

Well, it works, most startups today people developed that approach, but let's make a closer look at the process itself.

Cost of failure
There is a huge number of startups appearing every year, and quite a lot of them fail, which means that the cost of failure is a minimum of $15,000, if the startup goes for custom development, and it's just a minimum.

Cost of repetition
At the same time, quite a lot of startups reinvent the wheel, most of startups implement similar functionality and pay for them. For example, each startup needs a login page and user management, and every startup will pay for this. And we can continue, there are a lot of things like this that could be easily reused.

Longer timeline
At the same time: lack of re-usage means a longer timeline, more bugs to create and more dummy work to do, more dummy work.

How can we solve this?

Combine different SaaS

Today we have a variety of technologies we can combine to make an MVP is up and running without development! Sure, it depends on the startup we going to build, but we should start with a simple web-site builder and simple-wait list solution that allows us to research if there is any interest. For instance we can use

Wix
Squarespace
Unicornplatform
a bit more advanced options like
- Webflow
- Front

The cost of a typical lending page is 20$-50$ per month. You still need some marketers and designers to help, but it will be in total not more than $1.000 - $2.000 to validate the idea and estimate the interest.

If interest is growing we can go next and use a no-code solution

No-code solution

No-code apps and tools enable people without coding skills to create software using graphical user interfaces in place of writing code. Before no-code solutions, website and app development relied on programmers. Removing that barrier has opened the door to many more people, enabling them to build beautiful, functional websites and apps without writing code.

The Webflow team did a great overview of no-code apps here.

Let's try to do some pros and cons

Pros

Simplicity
Possibility of simple customisation
Super cheap, for instance, Buble.io starts at 29$/month
No deployment and infrastructure support

Cons

The complexity of the support
Complex customisation requires more time compared to custom development and full-time developer
Lack of data migration mechanisms
Issues in environments switching
Lack of monitoring

So in simple words, it could be an awesome and cheap start, but not an infinite success. You would not be able to pivot a lot in other to find your niche and/or scaling would be blocked by the development timeline or limited by the platform.

Low-code solutions

A low-code platform (or low-code development platform), besides providing a GUI-rich development environment, borrows heavily from visual programming principles to simplify, augment, and democratize traditional programming.

So Low code solutions try to keep the pros of no-code at the same time they try to be closer to a custom development approach.

The following platforms have high ratings amongst various software development circles:

1. Appian

2. Quickbase

3. Zoho Creator

4. Salesforce Lightning

5. Microsoft Power Apps

6. Nintex

7. MarsX

Fleexy Approach

Fleexy is a product company specializing in the development of easy-to-use micro apps for businesses.

Our micro apps are backed by thousands of hours of development and continuous improvement, ensuring that your project stays up-to-date with the latest features and advancements.

The global vision of the company is to build the hugest foundation of micro apps.

Fleexy team builds micro-apps on top of relatively the new platform MarsX, which combines code and no-code together to unlock maximum flexibility.

What is a micro app?

A micro app is a little block that helps you to create your solution quickly. You can use it to build a platform for selling art, an e-commerce shop, or anything else you want.

So it's a reusable piece of logic. No need to develop it again and again, just use it out of the box.

How expensive it is?

It all depends on the app you want to build, the lower input is 299$ per month. It's enough to build a curly hair service marketplace like https://www.thedaba.com/. Just imagine, you get a product ready to go, which is possible to customise and update from a UI perspective with such a small amount of money.

How simple to customise?

If changes are complex and micro-app doesn't support the changes you want to apply then it will be required some coding, then it will be equal to custom development. Otherwise, you would just configure settings and make it work.

For instance, Auth Microapp Fleexy team has allowed the integration of more than 10 kinds of authentification providers.

Budget in case of this approach

How does the budget of the average MVP look like with Fleexy Microapps

As we reuse a lot of things from the foundation we need to develop much less. What we need still to pay for is for project scoping and for custom development - to make the product unique. Usually, it's enough to have just one month to build a product on top of micro apps, including some customisation and configurations.

Where is the Design/Marketing/Sales

They are optional as there are many templates you can use out of the box, as for marketing and Sales it's something we offer our clients to start from: "do basic research and find the niche".

Conclusion

In conclusion, developing an MVP is essential to test the waters before launching a final product. However, the custom development approach may have some downsides, such as the cost of failure, the cost of repetition, and a longer timeline.

No-code and low-code platforms can be a solution for these downsides, allowing for cheaper and faster testing while still having customization options.

Fleexy approach, with its use of micro apps provided by MarsX, offers a unique solution combining code and no-code development.

It unlocks a cost-effective and quick way to build a unique product on top of pre-built and reusable logic. Ultimately, choosing the right approach depends on the product needs, timeline, and budget.

Follow me on Twitter
Connect on Linkedin

Converting from a Developer to a Product Founder in 10 years

Igor Boky — Wed, 21 Jun 2023 10:58:18 +0000

A little story of a workaholic entrepreneur switching from a developer to a product founder in 10 years.

How it all started

I was a happy student when I started to develop for money. It was a lot of fun and it was super interesting: a lot of new technologies, a lot of challenges.

I had no family at that time (too early) and I had all the time to develop things. I was kind of a geek and I was always super strict to myself on deliverability.

The company I was working for was small and full of entrepreneurship or better to say startups, which was awesome, otherwise, I would leave it earlier.

But even it was a lot of fun, I decided to go further into the world of freelance. I was aiming to become an owner and boss to myself.

Owning the life being a freelancer

It always a lot of "benefits" of being a freelancer, ones could imagine a person sitting on a beach of an island drinking a mango smoothie and working on a project... This is a nice picture to imagine, but for most freelancers, it's not like this at all.

It was a long standup for me. It was necessary to produce good reviews and build a few products for free. But later I managed to find one customer, and then another.

Then accidentally I was working with 10 different customers at the same time: you know that, handling support vs active development.

Did I become super-independent? No, not really, instead of one boss on my previous job I got 10.

Yes, I have increased my income, from 3-4k to 8-12k per month.

Yes, I have increased my expertise quite a lot because I have joined more than 30 different projects in different niches, especially in e-commerce, travel-related and fin-tech.

No, I don't get any way to become happier in that setup, it didn't work for me.

So I had to quit and find a better focus for me. It is not something I liked to do forever. I didn't want to come to ordinal work at the office as well.

So it was the moment when I had to switch to the product.

How do you get to product ideas?

I believe that the best product ideas could be born at the centre of your expertise.

While working on different client projects I always repeat the same routine, again and again. Their routing is the thing that everydeloper does every day.

Every developer reinvents the wheel every day to build the bicycle.

Example: How much time it's necessary to spend to build a simple ToDo list with backend and frontend, DB, Email notifications and such?

Well it could be various, but it will take at least one day if you are familiar with the deployment, you have good knowledge and high expertise on many aspects like

Hosting
DNS
Frontend Frameworks, HTML + CSS + JS
Backend Frameworks
Email Providers,
Email templates
etc.

How much time do you need to spend to get such expertise and deliver faster enough?

Remember we are talking about a ToDo list, the simplest app we can imagine.

These days I met John, who is started his work on Mars. It was a nice match, he got into the same set of questions but did a step further - started to deliver a product that could solve this.

Let's build a ToDo List as quickly as possible

Let's talk about the ToDo list again.

One day I asked myself what could be the easiest option to develop a ToDo list app.

For me, the answer is: don't develop it -> Reuse something that is already developed/tested/production ready.

The question is how? From top to the bottom in terms of the amount of knowledge required

You could find a few ready-to-go npm or *any-ecosystem* packages to get it faster
You could build something with AI - just ask to build UI and API and then improve
You could find a GitHub seed project
You can use a low-code solution
You can use a no-code solution

So which option to pick depends on many factors like

Ability to customise
Amount of efforts
Running costs (cost of hosting and maintaining )
Long-term support and improvements

Let's focus on Customisability VS Speed of development

I have tried to compare options and a simple chart

From my experience, the more you want the solution to be different to others -the more time you need to spend.

I have tried many no-code platforms, and they gave me good speed in the beginning, but I always had a lack of customisation and simplicity of updating as soon as the project grows. At the same time, the fully custom solution I have developed for my clients is something I'm not able to support as well because the amount of custom functionality the product eventually introduce is huge, you need a team to manage this, which is ok, but it creates bigger running costs.

Then I take a look at low-code platforms, it was a bit "unclear" thing for me. In simple words, it's a kind of platform where you can code, but at the same time, you can use a no-code approach. It seems like a perfect combination for me and that is where Mars is sitting.

How it works there

Everything on Mars is a micro-app. You can combine them, fork, reconfigure etc. You can do a lot of changes without coding and you can write a thing from scratch if it's necessary.

This is how I came to the product idea

Mars seemed like an awesome platform for me, but it's just a tool, the same as many others. If I want to build a ToDo list on Mars I need somebody to develop such a micro-app for me. Then I did a simple approximation on bigger products: an e-commerce store like Amazon, service marketplaces like Booksy or Renting platform like Airbnb

And I got this

Many people develop similar big and complex apps today.

And all of them repeat the logic.

Every founder pays for this.

So, to make such founders' and developers**'** lives better, I could provide them with ready-to-go reusable micro apps built by niche experts.

Being a developer myself and having quite a huge experience in many niches after 10 years of product development for clients I was in the right shape to take it off.

That is how Fleexy has been founded. Or flew up!

The Team, or Crew

Teams build things together, and Crews starts the spaceships.

From the products I have worked with, I remember a few very talented guys.

Perfect tech experience, passion to create things, open-minded and honest vision of things.

Together with Alexey and Andrey, we founded Fleexy. We aim to build the biggest foundation of micro-apps for Your Future Product.

It was a nice moment, it was always so inspiring to build a new thing, you start light, you have no dependencies, and you feel like you can fly. So we did.

But we are a self-bootstrapped company, we have no investors, that means we don't have a lot of money to burn.

That means that we have to find a customer sooner than later.

Building a product is a lot of fun

We started with an obvious thing: testing different niches, where we could be helpful.

That is how we get into 2 of them

By doing these demos we also created a solid number of micro apps we can use now for our next project/product. We were happy to share that with the world.

So we build a website and describe what we are and why. And guess what?

We got 0 visitors.

How Coders switch to Marketing

That's where we are now. We are in marketing mode. We do posts, we do tweets.

We literally spend days writing blog posts (and we still do), we do keyword analysys, and we do back-link building to get higher domain authority.

That already gave us some leads!

This helped us to get our first customers and even released one of them The Daba. It has been built on top of the Services-Marketplace solution we have. It has a lot of customisations, especially in terms of UI, and it looks and works awesome. Try it out.

Fleexy is just at the beginning of the fly

We have quite big ambitions - we want to disrupt the current way of app development. It should be faster, quicker and cheaper. The more we do this, the more I believe we will make it work.

And as for me

For sure I lost the income. It was a lot of pressure and it still is.

But I got much more:

Having a mission I believe could impact many things tomorrow
The crew I can trust
Freedom to do things and be able to define my next step
The long journey I'm happy to be part of

Follow me on Twitter
Connect on Linkedin

How to Build AI powered App in 5 minutes in Node.js

Igor Boky — Wed, 14 Jun 2023 07:39:32 +0000

Hello

Did you try AI for real in your projects?
Specifically OpenAI API or just announeced Vertex from Google?

Here we are.
We will create a product generator for your next marketplace solution!

Create a new project
Init npm init es6
Install open AI package npm install openai
Create a file index.js
Add this code

import { Configuration, OpenAIApi } from "openai";

const configuration = new Configuration({
  apiKey: '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' 
// Input Your Open AI key here from
// https://platform.openai.com/account/api-keys
});
const openai = new OpenAIApi(configuration);

async function getImage(info) {
  const response = await openai.createImage({
    prompt: info,
    n: 1,
    size: "256x256"
  });

  return response.data.data[0].url;
}

async function getProductInfo(info) {
  const completion = await openai.createChatCompletion({
    model: 'gpt-3.5-turbo',
    messages: [{ "role": "user", "content": `Generate a creative info in JSON for product type ${info}, use following structure  "{name: string, price: string, currency: string }", respond only with JSON` }]
  });

  const output = completion.data.choices[0].message.content;
  return JSON.parse(output);
}

function getProduct(info) {
  const productInfo$ = getProductInfo(info);
  const image$ = getImage(info);
  return Promise.all([productInfo$, image$]).then(([product, image]) => ({ ...product, imageUrl: image }));
}

async function execute() {
   const topic = 'Kid Toy';
   const productNumbers = 4;

   const promises = [];
   for (let i = 0; i < productNumbers; ++i) {
     promises.push(getProduct(topic))
   }

   const products = await Promise.all(promises);

   console.log(products);
}

execute();

6) Execute the file node index.js
7) Check terminal or console, you would see the following

Can you imagine? you just generated quite real products, it could be anything else. Be creative - test it out.

Ready

Now you have a list of products including price and image. Simple as that. You can render it into HTML and get results like on the image below?

What is the next step?

You can play around with prompt and expirement to make it
You can convert this into the server and build API or do a SPA

Let me know if you want to see the evolution of this project!
Will post more.

Follow me on Twitter
Connect on [Linkedin] (https://www.linkedin.com/in/igorboky/

Essential Product Metrics Every Startup Founder Should Know

Igor Boky — Mon, 12 Jun 2023 20:56:49 +0000

Introduction

Product metrics vary across products and industries, but there are common ones that benefit any product. Choosing the right metrics as key performance indicators is crucial in product development. Experts recommend having metric tracking in place when launching a product or feature.

In general most of them works in way - The Bigger the Better. So try to increase them, but they defines and explain different aspects of your product and your users behaviour.

In general product metrics can be grouped to evaluate performance. A simple categorization divides them into economic and engagement metrics.

Product Economics

Economic metrics measure things that impact the profit equation directly, such as revenue or costs. Product leaders rely on these metrics to assess the quality of marketing efforts or the viability of the business model. Below is a list of five of the most important economic metrics.

Monthly Recurring Revenue (MRR)
Monthly Recurring Revenue (MRR) is a vital SaaS metric that provides a reliable and predictable view of your monthly customer-generated revenue. It not only reflects your current earnings but also indicates your growth over time. Consistent MRR enables accurate financial forecasting and empowers informed business decisions. Learn more about the magic formula involving Average Revenue Per Account (ARPA) for a detailed explanation.
New customer growth rate
New customer growth tracks the number of brand new customers a company obtains. It distinguishes between new and repeat business and indicates future growth. The growth rate is calculated by taking the difference in customers between two periods, dividing it by the number of customers in the previous period. For example, going from 100 customers to 500 customers yields a growth rate of (500-100)/100 = 4, or 400%. If you have such Growth rate for months - you are the next unicorn!
Churn - The Lower the Better
Churn is a key metric for subscription businesses, representing the percentage of lost customers within a specific period. It's calculated by dividing the number of cancelled subscriptions by the total customer base. Lower churn indicates a strong alignment between your product and customer needs.
Cost per acquisition
Cost per acquisition (CPA) or customer acquisition cost (CAC) measures the expense of acquiring a new customer. It's a significant product metric that affects pricing decisions. The price must cover both the cost of delivering the product/service and the customer acquisition cost. CPA is calculated by dividing the total direct marketing costs for a specific product by the number of new customers acquired.
Customer lifetime value
Customer lifetime value (CLTV) gauges the value of an average customer to the company. It's calculated by dividing the average lifetime spending per customer by the average cost to serve them. For instance, in a subscription business, you multiply the subscription margin (price - variable cost) by the average subscription length to determine the average gross margin per customer.

Product Engagement

Engagement metrics track customer/user interaction with the product, revealing what's effective and what's not. Product leaders rely on these metrics to assess customer value and the performance of various product aspects/features. Here are five vital engagement metrics.

Net promoter score
Net Promoter Score (NPS) is a widely used metric that gauges customers' likelihood to recommend your product. It's calculated by surveying customers and subtracting the percentage of detractors (ratings of 6 or lower) from the percentage of promoters (ratings of 9 or 10).
Active users
Active users count the number of users who actively engage with the product within a specific timeframe. It provides a more accurate measure of the user base compared to total account numbers. The optimal measurement period for active users, like daily or weekly, depends on the expected frequency of product usage by users.
Session length
Session length measures user engagement duration and reflects user satisfaction with the product. To calculate session length, subtract the start time from the end time for each session and then determine the average session length. Typically, very short sessions (a few seconds) are excluded to filter out accidental or bot-related activity.
Number of sessions
The number of sessions tracks user frequency within a specific period, typically aligning with the active users metric. It provides insights into product stickiness and user importance. Low numbers of active users or sessions may indicate a product's failure to meet user needs effectively.
Feature usage
Feature usage metric assesses the percentage of users utilizing different product features. It aids product leaders in prioritizing important features and potentially removing or deprioritizing unused ones. To calculate feature usage, divide the number of users who used a feature by the total user count within a specific period. For more details on measuring product feature usage, refer to this post.

How to track them all?

Well, today we have huge number of options to do that, usually it requires a bit of mix of technologies/approaches to make it work. But it always better to start simple

Analytics Tools
Start from using a solid tools like Google Analytics
that you can install with a simple snippet, or go with UserMaven, also there is quite nice heatmaps and recording you can get via Hotjar
Do User Feedback and Surveys
Gather user feedback and conduct surveys to gain qualitative insights and supplement quantitative metrics. This can help uncover valuable information about user preferences and satisfaction. To make it work you can use Google Forms, TypeForm
Custom Data Collection
As you grow you can consider a bit more precise configurations and data analytics approaches. Implement custom data collection through or event tracking for example via Mixpanel. This allows you to capture specific metrics tailored to your product's unique needs

Conclusion

When it comes to success, there's no one-size-fits-all formula. The metrics that matter most depend on your goals and situation. If you're running a social platform with affordable products, focusing on maximizing the Net Promoter Score can help you bring in more users and boost sales. But if you're in the B2B SaaS game with higher subscription prices, it's all about prioritizing Customer Lifetime Value.
By keeping your existing customers happy and growing your business steadily, you'll be in a solid position to win over new clients.

Remember, customizing your set of metrics to fit your unique business needs is the secret sauce to making smart decisions and achieving your desired outcomes.

What do your track on your product and how? Share in comments!

Follow me on Twitter
Connect on Linkedin

How to add Facebook login to your app

Igor Boky — Wed, 24 May 2023 14:30:17 +0000

You have any kind of marketplace business, or you provide services on your own, or you have a business sales for consumers online.

Fewer steps the user need to do to get the purchase - a bigger chance to get him to pay.

Quite often users leave at the very end of the flow: payment. Payment often requires a login. Asking the user to provide an email name and password is a terrible option there as then we ask the user to do extra actions he most likely doesn’t want to.

Do we need the user to be logged in to get a transaction gone through?

Nowadays we have the following less-disturbing options

Ask the user to pay without registration. Just ask for an email to be able to notify a user about the order and track the order properly in your system. Then user doesn’t do any long action during the flow, but we don’t get a fully registered user.
Use social-media or OAuth-based logins approach: Facebook Login, Google Login, LinkedIn Login. They are much smoother and faster compared to the Email + Password option, also they provide other benefits we will describe later. They are free.
Use Phone number login with a short-code verification. It’s also a good option today, but usually, there are some operational costs on top of it, and on top of this you still usually need to ask for email and other information anyway.

Let’s focus on Facebook Login

What are the benefits we get right away:

a Facebook login eases the process for your users to create an account and login to your website. By integrating Facebook login, users can directly use their existing Facebook account credentials to access your website, eliminating the need to create and remember new account details. This can save your users time and hence enhances their user experience on your site.
Facebook is one of the largest social media platforms in the world. By integrating a Facebook login on your website, you allow your users to connect with your business through their Facebook profile. This can help you widen your reach and enhance your business's visibility.
Facebook login assists in the acquisition of user data. Facebook's API enables you to retrieve necessary information about users, such as their name, email address, and profile picture. This user data can help you personalize your users' experience on your site and enable you to segment your users better.
a Facebook login on your site can enhance your website's security. By using Facebook's infrastructure to authenticate user credentials, you eliminate the need to manage sensitive information, such as passwords, on your website. This helps ensure that your user's information is secure.

Quite a lot of good stuff, agree?

But how?

Even though Meta did quite a good work of explaining how to setup it up for you, it still quite a lot of things to understand and to be done before you can make it work. Also, it usually comes to development costs

How does it work in Fleexy?

To remove all the pain of integration on our end we made a bunch of microapp for social logins

By using micro apps you would be capable to integrate the following Social Media Logins in a few minutes:

Google Login
Facebook Login
LinkedIn Login
GitHub Login

All you need to do is update a few fields on our end to enable Social Media Logins Specifically for the Facebook app

Just go into the Facebook app you have created before according to the guide

And put them into the Auth Setting on Fleexy Auth Microapp

That's pretty all. Now you will get the Facebook Login up and running

Conclusion

Integrating a Facebook login on your website has significant benefits. It streamlines the login and account creation process, broadens your reach, enhances personalization and segmentation, and increases security. With various cost-effective tools available to simplify this integration, implementing a Facebook login should be an essential consideration for businesses looking to establish an online presence.

Thanks to Fleexy Microapps you can make this in with no headache and get happy customers with no extra costs for your business.

Follow me on Twitter
Connect on Linkedin