Here we finally are with my first “real” publication! 🥳

If you’re curious how I got here, look at my very first post here.

I don’t know yet what’s the best format to talk about the learnings, and the thoughts every week, so I’ll just start with an overview of different stats and follow up with the rest.

Since this is the first post, this might be longer than how it used to be as it contains also ideas and initiatives that I actually started a few weeks ago rather than this week.

Overview of current status

Here are some snapshots of different stats.

NextCommit

Users

In the last week, I got:

• +10 signups

• 20 active users

Stripe

Well, still at 0, so I don’t even know what’s the best part of the dashboard to include here. 😂 

I’ll wait until I get my first paying customer! 🤞 

Plausible Analytics

Google Search Console

Personal branding

I recently realized how important is to also have an audience to get a wider reach when talking about ideas and talking about my products. So I started being more active also on socials like X and started this newsletter.

X

Beehiiv newsletter

Thanks a lot for the first 9 subscribers to this newsletter! ❤️

Let’s now dive into the main topics! ✨

Early validation

What attracted me to build NextCommit is the combination of different factors:

• I have experience in the space, both as a candidate and as a hiring manager (announcement link)

• Other job boards proved to be able to have a sustainable business (RemoteOK, Remotive, etc.)

• I enjoy the technology involved in building this (but yeah, this doesn’t drive business numbers… 🤣)

I think I’m doing a better job than other job boards in finding more positions, but unfortunately, that’s not enough.

But how did I come to this conclusion?

Experimenting with paid ads

In the indie hacking community, they often say:

The idea is that you start with a landing page with the bare minimum:

1. A clear hero section

2. A way to subscribe and leave an email

If you get enough interest, you start building the product. An alternative that I often read about is to provide a link to pay in advance to get an even stronger validation.

All good, but all would be useless if you don’t get traffic right? 😅 

I even discussed this in X very recently.

With NextCommit I already have a product and a landing page, but I never tried getting that “real” validation yet. When talking about validation it’s better to split between “problem validation” and “solution validation”.

The problem I know it’s real, I faced it myself. The solution seems also validated, otherwise, there wouldn’t be successful job boards as successful businesses. But the key point is: is MY job board valid? Will users use MY solution?

This is a good opportunity to experiment as I’m still a noob in marketing and distribution, and I decided to experiment with paid ads. I worked many years in the video advertising industry so I’m aware of what CPM, CPC, etc. mean but being an advertiser and marketer is a different mindset.

Up to a few weeks ago, the platform was completely free. Also, it’s not like much could be monetized yet. 😅 

But I recently bumped into another job board Remote Rocketship that was showing more job posts behind a subscription. TBH I never thought that people would pay for it, but by looking at the X profile of the founder, the guy reached 6k MRR! 🤯

I was mind-blown.

Instead of measuring only signups, I can put the “load more” button that paginates the job entries behind a paywall and measure the conversion for paid plans. I anticipated already that I’m still at 0 paying customers, so you know already how it ended, but let me tell you how it got there. 🤣

I started with Google ads. What a pain, I needed to reach out to support due to the website being marked as malware. By searching on the internet for a solution I was very pessimistic about the outcome, many people were complaining about this issue and were not able to solve it even after months! Luckily after reaching out to the support, they were able to unblock me after a few days. 🎉 

The results? 🥁 

I spent around 50EUR and got 0 paying customers in return. 😭 

It was time to reflect on why this happened. On top of that, why Remote Rocketship was able to get paid users with the same model?

Here are my conclusions:

1. My landing page sucks:

   1. the hero section is pretty much what you’d see on every job board

   2. not much emphasis on highlighting the problem of looking for a remote job,

   3. no “about” section that could give more trust to users,

   4. no customers’ reviews for social proof are shown,

2. Very basic functionality: I’m proud of the AI-based scraper that runs NextCommit that enables to find a lot of remote tech job posts, but from a user perspective it only offers a filter by country and by category, and after the experiment started the pagination requires to upgrade to a paid plan. In comparison, Remote Rocketship has way more filters available.

3. Different personas: NextCommit focuses only on tech people. Maybe tech people are less willing to pay compared to everyone else? 🤔 

4. The price was high: 4.90EUR/week or 16.90EUR/month. I added a coupon visible during the upgrade with a 50% discount forever, but that probably scared also the majority of the people. How do I come to this price point? I mostly copied Remote Rocketship as well for this.

5. The job board itself doesn’t require authentication, but the link in the landing points to the /sign-in page and this seems to be a problem.

The number of new users increased, but not that much. As you can see there are 1.8k clicks, but I have only 34 users in total. I’m doing a more in-depth analysis later in the “Marketing, marketing, marketing“ section about this.

Differentiation and a better value proposition

The value proposition is bad, how can NextCommit shine among the other job boards?

If the value proposition was exciting enough, I shouldn’t have experienced that drop between landing impressions and users.

Rethinking the journey of landing a job

If you think about it, finding a remote job position nowadays it’s not that hard. Even if I provide 10x more open positions to show, that still doesn’t solve the real user problem of actually landing a job.

Landing the job it’s the real pain and the final goal of the user, not finding a position to apply to.

I started thinking about how I could help in that journey.

For sure, giving visibility about the opportunities (the job board) is the first step, but then the journey moves towards applying to those that match the candidates’ interests and skills.

The entry point for this step is almost always the CV.

In the last few years, I reviewed hundreds of CVs, and some of them were, really, bad. The interesting part is that the quality of the CV was often not correlated to the quality of the candidate. But I think that most engineers in general are not very good at selling and marketing themselves, and the CVs reflect this.

Someone might even say that there could be an inverse correlation between the quality of the CV, and the quality of an engineer. 😂 

For this reason, I’m currently working on providing a way to automatically analyze a CV. The analysis is split into two parts:

1. an analysis of the quality of the CV: is it clear? is it comprehensive? Is it well-organized? etc.

2. an analysis of the candidate’s profile: does the candidate have experience as a backend or frontend engineer? does the candidate have any leadership experience? etc.

The idea is to change the value proposition towards:

I’ll write a separate post once I launch it, but here’s a peek at how it will look like.

Let me know if you have some thoughts on this screenshot!

Everyone user will be able to see the summary of the CV analysis for free!

I tried to validate this feature with a few people I know by asking them to give me their CVs and so far the response has been positive!

LinkedIn AI

Last week a friend of mine sent me a video of this new LinkedIn premium feature. AFAIS it enables you to chat with a job posting to understand whether you might be a good fit or not and how you could improve.

TBH I think the current landscape is relying on a chat-based UX just to show off “Hey we’re using LLMs ourselves as well!“, but sometimes they’re just not the best UX, at least in my humble opinion.

Why not show directly what the user might be interested in rather than let them ask? I think many products just followed the ChatGPT UX, but it works well for broader use cases, while sometimes just a page with some text and clear insights and CTAs are more effective.

Anyway, I’m kinda digressing a bit here… 😛 

Overall, I don’t think this feature would be bad for NextCommit. The fact that I’m also focussing on a specific target persona (tech persons) enables me to build features specific to them (e.g. see the profile analysis screenshot above) compared to targeting a broader audience (and that’s probably indeed why LinkedIn opted for a more generic UX like the chat).

Multiple possible business models

I’m more than happy to help people land their dream job, but It’s important to remember that NextCommit still aims to become a sustainable business, and business means generating revenue. 😅 

I thought and found both B2B and B2C ways to monetize the platform.

B2B

Sponsored posts

This is probably the most common model for a job board. It’s super simple:

1. you bring a lot of traffic to the platform,

2. you ask companies to pay for having a sponsored job listing.

This could be done in any way that could lead traffic to that listing: pinning the entry to the top, changing the background color to a sparkling fluo, or whatever you think will drive traffic.

As an example, have a look at the RemoteOK form for sponsoring a job post:

Of course, RemoteOK drives a LOT of traffic, but if you’re able to get there, with some automation everything should just go on autopilot.

Beware, I’m not claiming that it’s an easy thing to do! 🤣 

Pieter Levels claims to have a 50k MRR from RemoteOK. Not bad, huh? 💰️ 

This would probably be the safest approach and once I get the traffic I might reach out to some companies to test a partnership.

ATS API

I don’t know if someone does this, but to build the CV analysis feature, I needed to expose an API for my front end. Would some company be interested in paying to use that API on a pay-per-use?

There are many ATS, and maybe some of them could be interested in relying on this API to provide some sort of pre-filtering automation to help hiring managers use their boards more effectively.

Pieter Levels (this guy is everywhere…) also recently tried to move toward the hiring manager side and shipped Applicant AI. It’s a whole new ATS powered by AI to help screen candidates more effectively. That’s an interesting idea, but that would imply a whole pivot of NextCommit.

But I fully understand the pain, I remember how much time I was spending on going through all the CV submissions and evaluating them.

B2C

In this case, it’s about providing some functionalities behind a paywall, either a subscription, a pay-per-use, a one-time payment, etc.

As mentioned before, the experiment I ran tried to monetize the pagination, but nobody upgraded.

The CV analysis is another feature that I’ll partially add behind a paywall. Why not fully? I want the users to grasp the value of it by at least being able to read the summary of the analysis. Seeing its value might increase the conversion rather than paying blindly beforehand.

The Biggest problem of B2C is the “organic churn”.

I don’t know if there’s a specific term for defining this type of business, but basically, my users do not face the problems I’m trying to solve every single day of their lives (luckily!). They’d use the platform only when they need to look for another job, so users will “organically” churn.

It’s a leaky bucket that needs to be continuously filled, but the water that goes out at some point can be reused for re-filling the bucket (happy customers who end up relying on NextCommit more than once for a job change).

This leads to a few questions:

• does it make sense to provide a subscription? Usually, subscriptions start from a monthly plan, but given that the usage might be shorter, I added a weekly one as well.

• is there a way to retain users AFTER they land a job? This is related to the leaky bucket idea. Maybe I could provide something more towards career development, but that would cause a lot of focus shifts.

If we exclude ads, B2C might be the only way to monetize the platform before getting enough traffic and trying sponsored job posts.

Marketing, marketing, and marketing

So far I haven’t allocated much time to marketing and distribution other than just a few posts on my personal LinkedIn and X and building a pretty basic landing page.

It’s not easy to both work on building the product and think about its distribution. I’ll likely start working in cycles: a week or so building, and a week or so distributing. That should help me better focus on each outcome.

Landing page and acquisition funnel

At this point, I criticized the landing page a lot. 😂 

Dan Kulkov summarized in this tweet what an effective landing page should look like:

Again, the hero section focuses on a too broad and common value proposition, the available features could be better explained, I could include better visuals videos of the platform, etc.

For now, let’s ignore the organic search as it’s almost none, so let’s see the funnel for paid ads:

• you see the ad,

• you click the ad,

• you reach the landing,

• you click on “go to app” or “start now“,

• you reach the /sign-in route of the application,

• you register,

• you start enjoying the board. 🥳 

Let’s now have a look at the stats.

Between the 11th of April and the 26th of April, from Google Ads I have:

• 66k ads impression,

• 1.8k ads clicks,

• CTR: 2.7%.

The CTR is not bad, 2% should be the delimiter between good and bad, so the ad itself doesn’t seem to be an issue.

This is also confirmed by the stats I have from Plausible for the same date range which says I got 1.8k unique visitors.

But at the same time, look at how many /sign-in:

It’s an approximation because an active user would need to sign in again once the token expires, but more or less this means that:

• 1.8k unique visitors on the landing,

• 179 clicking the CTA (remember that from the landing page, all the links point to /sign-in)

• CTR: 9.9%

So after the first interest, almost 10% of the people were convinced that this was worth exploring and checking the app. I don’t have a benchmark for this, but I feel that this should be higher. I’d need to track if this number increases after refactoring the landing page.

Finally, we have:

• 179 sign-in in plausible

• 21 new users in Clerck in that time range

• CTR: 6.7%

Ok, I don’t know if there’s a standard here, but… WTF! 😂 🤯 

How come that 93.3% of the persons didn’t proceed?! You clicked on the ad because you’re interested, you even clicked on the CTA because the landing, despite being bad, triggered something in you to make you willing to try, and then you just leave. 😭 

The only explanation that I have is that the login is a strong blocker.

This is the current login:

Maybe people are scared that this could be a scam? Or maybe these login options are not enough? 🤔 

The funny thing is that the root route of the app is not even blocked behind a login! 🤣 

I’ll try to change the link on the landing from /sign-in to the root and verify if the conversion from the landing CTA to becoming a new user increases. But I need to find a way to track that to distinguish those who reach the app from the landing from those who are going directly to the app.

This should be a low-hanging fruit. 🤞 

Another thing missing in the landing is some form of social proof. I’ll be exploring Senja and maybe I can ask already some of the people I involved in experimenting with the CV analysis to leave a review there.

Content marketing and SEO

As shared in the beginning, the Google Search Console dashboard looks pretty bad. 😅 

I haven’t invested in SEO and content marketing yet. The only thing I have done so far is a blog with two blog posts and some announcements. I relied on Medium here simply because I don’t wanna manage a blog and Medium seemed the easiest approach, and should already be SEO-ready.

Interesting fact: those blog posts have been written using the POC of another project that I’m not actively working on, but I’d like to better revive at some point towards content marketing rather than generic blog posts.

I started reading “The art of SEO”.

For now, the main takeaway related to content marketing learned is to write blog posts targeting the different types of users, but part of the target persona, by considering that they are at different moments of their journey. ✨ 

What does this even mean?

Think about it this way. An engineer looking for a job might be just exploring, or actively looking, or already applying, etc. Different types of content will attract different users depending on where they stand in their journey to landing a new job. Users in their early phases might not be converting (to a paid plan) straight away, but you start by planting the seed by giving brand awareness.

I still need to properly go through technical SEO, keywords, and backlinking stuff, SEO tools are damn expensive! 🤯 

Socials

I created an account for NextCommit on both X and LinkedIn, but I didn’t use them. For now, I’m only using my personal account. I was considering making the NextCommit X account post daily the newly discovered job posts, but I didn’t have the time to work on it.

I’m currently also trying to improve my audience on X, especially within the solopreneur community. The awesome thing is the support and all the learnings you get from hundreds of people at different stages of the solopreneurship journey! ❤️

This is an example of a tweet of mine sharing a problem where people shared their experiences, pains, and learnings.

This newsletter is hopefully also going to increase my reach, hoping to share something useful with the community.

As I said on both X and LinkedIn, if you told me a few months ago that I’d have started a newsletter I wouldn’t have believed you. The same applies right now when it comes to TikTok videos, I’m wondering whether I’d end up putting my face on a TikTok video to market NextCommit (or other future projects). 🤣 

Public Launch

As soon as I wrap up the CV Analysis I’ll try to launch on ProductHunt and see what happens. 🥁 

Product spin-offs

When I was early working on NextCommit and shared it only with a small circle, a person just co-founded a startup, reached out to me asking:

"Hey Marvin, if you’re scraping job posts, it means you have information about the stack of the different companies. Can you tell me which companies are mentioning Github Actions?“

That was super easy to do as I store the extracted raw text of the job positions in the database, so I was able to get it.

He was looking for leads.

The scraper itself could probably be re-purposed to be a standalone service and scrape whatever the user needs with a pay-per-use pricing model.

This is something that I’ll definitely keep on my radar.

Weekly grinding

I admit that the solopreneur journey can be tough. I mentioned already some of the challenges I faced in this issue, and on top of that, you might sometimes start wondering “am I really doing the right thing?“

As I wrote in the tweet, what works best for me is to plan what I have to work on in cycles, and trust the process. These sparks of self-doubt usually happen during the development of some features, but it’s important to split the two moments:

• planning: think about what has to be done, criticize what has been done, and tune the plan accordingly.

• execution: don’t think, just execute and get the shit done!

In the meantime, I’m waiting for the injection of dopamine the first Stripe notification will give. 😂 

Recap on what’s for the next week?

What follows is not an exhaustive list of things that I’ll work on, but action items as takeaways from this newsletter issue.

Product

• Deploy the CV Analysis feature

Marketing and distribution

• Change the landing CTA to point to the root of the app instead of /sign-in and measure the conversion rate difference

• Announce the CV Analysis feature on my personal socials and write a blog post on Medium

• Start refactoring the landing page and get some social proof

• Start looking at how, where, and when to launch on different platforms: Reddit, ProductHunt, etc.

Conclusion

I realized that I condensed a lot of stuff in this issue, I hope you enjoyed it! 😄 

If you’re not subscribed yet and you’re curious about this journey do it now! 🚀 

See you next week for the next issue or on my socials!