frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

Ask HN: How do I market to consumers as a solo dev about to go to uni?

2•Alex-Programs•5h ago
Hi!

I've spent my gap year making a language learning tool[0]. It's been somewhat successful, but I really don't have a sustainable marketing strategy (e.g. SEO or ads) - it's all from me sharing it, making blog posts, etc. I'm going to uni in ~2.5 months, and I'd like to be able to keep it growing, even as most of my attention goes towards my studies.

My users are really happy with it, the conversion rate is good, and there's been a surprising amount of word of mouth from them already, but it's just not quite enough.

I've had some limited success from Reddit text ads, but their algorithm is determined to optimise for cost per click, rather than conversions, and it's only getting worse over time. Text in general is a lot easier for me to work with.

X ads don't work at all. Nor do Google or Reddit banner ads, or Adwords.

SEO is an expensive labyrinth. I've made some progress there, paying for some backlinks and having a high quality blog, but not enough to rank for the keywords that matter. I do have some capital (single-digit thousands of £), and I'd be willing to use it, but only if I know it'll actually give me a good return.

I made an affiliate program with generous payouts (my entire profit margin and then some - 50% of all revenue for 6 months), aaand nobody's used it. I suppose I need to treat that as an entire separate marketing task.

I made the world's best machine translator (https://nuenki.app/translator), and that's cool and has got me some users, but again, doesn't scale.

Maybe it's the landing page? People are much more likely to convert if I give them even a one or two sentence summary of it beforehand, so maybe it isn't explaining it well enough.

Anyway, yeah, I'm not really sure where to go from here! I'd appreciate any advice.

[0] https://nuenki.app

Comments

codingdave•5h ago
> My users are really happy with it, the conversion rate is good, and there's been a surprising amount of word of mouth from them already,

If all that is true, you already have hit the dreams of startup marketers. People post ads on various sites trying to get enough people that their users start word-of-mouth organic growth. What you need to continue growth is to focus even more on your current customers. If you burn your energy on sales, you risk ignoring them, and that organic growth might stop. So talk to your users more, focus on their experience, and make them so happy that they keep talking. Build the features they tell you are needed to make things even better. Ask them where they think other people would be responsive to marketing, and focus your efforts wherever they say.

Alex-Programs•5h ago
I spent quite a while doing that, and I still kinda do. I quite frequently add new features for a single person! The problem is that 25 people who are happy with the product and likely to share it is still only 25 people, and there aren't many features left to implement - the product is pretty mature.

I could improve the definitions, but that'd cost huge amounts of money in order to use proper dictionaries rather than Wiktionary, or I could add support for multiple languages at once, which two people have asked for but which would require a lot of dev time, and after that there isn't really much left to change.

And I've tried asking where I can find people, and they generally suggest "subreddits" (most ban self-promotion, and I've already posted on the ones that don't), and "discords" (all of them are very hostile to self-promotion).

I think the problem might be due to my landing page? People are far more likely to convert if I've already explained it to them. But I'm not sure how to convert a text explanation into the landing page form.

PaulHoule•5h ago
My first impression is that this is tough because Google Translate is free and many people think that is good enough.

Now I find myself using Google Translate because there is some text in French or Chinese that I want to read. There is another market for someone who has some text in English (say) and they want text in French or Chinese to put on their web page or fill out a government required form or something that has higher quality requirements in terms of being correct, good prose, etc. That person is either doing the work themselves, paying a translator, or not doing it because they can't afford it. I think that is the person you're trying to find.

I don't think you want to tell that person how it works (finding the best commercial LLM for the task) because if I think it is just sending the work out to some particular LLM couldn't they just use it themselves and cut you out?

Alex-Programs•5h ago
To be clear, the main product is translating from English to the language you're learning - not the other way around - so that you immerse yourself while you browse.

The hybrid translator is kinda a side thing based on my research into LLM translation quality.

It's a bit more complicated than using the best LLM, because it combines the results of the best ones, but yeah, that's broadly how it works. I made that part open source, anyway - I'm not trying to sell the hybrid translator, just use it as a marketing tool.

But I guess this comes back to the fact that I think my landing page might explain it poorly. I'm just not sure how to explain "It finds English sentences in webpages, filters out so that it's only the ones at your difficulty level, then translates them into your target language, so you're immersed while you browse" in the form of a ultra-low-attention-span landing page.

Ads are "rolling out gradually" to WhatsApp

https://arstechnica.com/gadgets/2025/06/ads-are-rolling-out-gradually-to-whatsapp/
1•Willingham•29s ago•0 comments

Russian Spies Are Suspicious of China

https://www.nytimes.com/2025/06/07/world/europe/china-russia-spies-documents-putin-war.html
1•gmays•47s ago•0 comments

Data-Parallel Types (SIMD)

https://www.modernescpp.com/index.php/data-parallel-types-simd/
1•ibobev•1m ago•0 comments

Don't Argue: Build

https://lemire.me/blog/2025/06/16/dont-argue-build/
1•ibobev•2m ago•0 comments

Auto-Updating AI Expert Using Telegram Channels and NotebookLM

https://www.junctionbot.io/articles/creating-a-self-updating-ai-expert-based-on-a-telegram-channel
1•astuteficus•3m ago•1 comments

Facial recognition error sees woman accused of theft

https://www.bbc.com/news/articles/cdr510p7kymo
1•croes•5m ago•0 comments

GhostNet: The Cyber Espionage Operation That Shocked the World

https://adaptive.live/blog/inside-ghostnet-the-cyber-espionage-operation-that-shocked-the-world
1•debarshri•6m ago•0 comments

Why today's graduates are screwed

https://www.economist.com/finance-and-economics/2025/06/16/why-todays-graduates-are-screwed
1•petethomas•7m ago•0 comments

Why content is becoming less and less interesting to consume

https://citizendot.github.io/articles/the-great-content-blah/
1•CITIZENDOT•8m ago•0 comments

Show HN: Canine – A Heroku alternative built on Kubernetes thats 10x cheaper

https://github.com/czhu12/canine
2•czhu12•9m ago•0 comments

Is Tahoe macOS 26 or 16?

https://eclecticlight.co/2025/06/13/is-tahoe-really-macos-26-or-16/
1•alwillis•9m ago•0 comments

Startup Slop

https://jihad.house/notes/startup-slop/
1•mxstbr•12m ago•0 comments

Andy Goldsworthy on 50 years of making art

https://www.nytimes.com/2025/06/14/arts/design/a-decade-of-bruising-labor-a-6-mile-work-of-land-art.html
1•bookofjoe•13m ago•0 comments

Novo Nordisk will lose billions because it didn't pay a couple hundred dollars

https://www.marketwatch.com/story/the-story-of-how-novo-nordisk-will-lose-billions-because-it-didnt-pay-a-couple-of-hundred-dollars-bcef7c01
5•shadowtree•13m ago•2 comments

I love this site – attempt at chatbot for Medical Clinic

https://web.archive.org/web/20250616181520/https://valby-rtg.dk/Valby/index_UK.html
1•arbirk•14m ago•1 comments

Agent Management Interface Patterns

https://lukew.com/ff/entry.asp?2106
1•ahmed_sulajman•15m ago•0 comments

What I mean when I say "my union"

https://posts.demarko.org/2025/06/07/what-i-mean-when-i-say-my-union/
2•meithecatte•15m ago•0 comments

Paper-based devices diagnose malaria in asymptomatic people

https://news.osu.edu/paper-based-devices-diagnose-malaria-in-asymptomatic-people/
1•geox•16m ago•0 comments

Build Stuff

https://andrewpwheeler.com/2025/06/12/build-stuff/
1•apwheele•16m ago•0 comments

Tourists damage crystal-covered chair in Italian museum by sitting on it

https://www.theguardian.com/world/2025/jun/16/palazzo-maffei-verona-italian-museum-police-tourists-van-gogh-chair
4•mykowebhn•18m ago•0 comments

Show HN: I built an open-source Google translate widget

https://github.com/JigsawStack/translation-widget
4•ykjs•19m ago•1 comments

HTAP: The Rise and Fall of Unified Database Systems? – InfoQ

https://www.infoq.com/news/2025/06/htap-databases/
1•rbanffy•22m ago•0 comments

Cherbourg Project

https://en.wikipedia.org/wiki/Cherbourg_Project
1•mhb•22m ago•0 comments

Tropical forests are heating up. Can they cope? – Science – AAAS

https://www.science.org/content/article/tropical-forests-are-heating-can-they-cope
2•rbanffy•23m ago•0 comments

Scientists genetically engineer a lethal mosquito STD to combat malaria

https://newatlas.com/biology/genetically-engineered-lethal-mosquito-std-combat-malaria/
2•burnt-resistor•23m ago•0 comments

Improving Brain-to-Image Reconstruction via Fine-Grained Text Bridging

https://arxiv.org/abs/2505.22150
1•PaulHoule•23m ago•0 comments

There's a Tron Treasury Company

https://www.bloomberg.com/opinion/newsletters/2025-06-16/there-s-a-tron-treasury-company
1•ioblomov•24m ago•1 comments

Bender browser automation – low token solution

https://github.com/sjoerdvanderhoorn/bender
1•sjoerdvdhoorn•24m ago•1 comments

Sovereign Public Cloud for All Microsoft Cloud Customers in Europe

https://blogs.microsoft.com/blog/2025/06/16/announcing-comprehensive-sovereign-solutions-empowering-european-organizations/
1•smartbit•26m ago•0 comments

AI Interview Copilot

https://www.interviewcopilot.info/
1•dimakrivolap•28m ago•0 comments