My Old Startups and Lessons Learned

or 'How I became me'

shanedevane.com: circa ~1996

I got online around 1996 - I was 15, messing around with BBS systems and dial-up in Dublin. When I got an IOL (Ireland Online) internet subscription, it came with a free member homepage on homepage.iol.ie, and that's where I built my first web pages.

I registered shanedevane.com in October 2000 and moved everything over to that. The site was to promote my freelance web design portfolio. Back then, I was living with my parents in Dublin, Ireland, and trying to make a name for myself as a web developer.

One of my earliest memories was printing out A4 'Web Designer' flyers and walking around a town called Bray, handing them to any business that would accept them.

I succeeded in getting 2 clients that way, but I would not recommend it.

Freelance Ireland: circa ~1998

I started my first company at 18 years old in Ireland. Back then, if you wanted a .ie domain name, you needed to register a business name, and the only way to do that was to start a company! So lucky me, at 18 I had to walk into a solicitor's office for the first time so they could act as notary for me to create a single director limited company (a new thing at the time), and thus my startup adventures began.

Freelance Ireland started to address my own problems: 1) I needed to have some sort of good portfolio showpiece website that I could use in my web design portfolio so I could win clients, and 2) I needed to find freelance work, so what better than operating the site that people used to find freelancers! Unfortunately, the freelance web design business wasn't for me as the income was sporadic and the work was always 'lightweight', and I wanted to learn (and get paid) for systems database development and have a more stable source of income. Freelance Ireland ran for close to 18 years and eventually turned into a LinkedIn Group.

Milestone Detail
Lifespan ~18 years (1998–2018)
Peak members 4,197 registered users
Projects posted 881 across 51 categories
Freelancer profiles 325+ across 90+ skill categories
Outcome Evolved into a LinkedIn Group

Irish Coffee Club: circa ~2001

This was a voyage into building a product. Fresh from reading 'The 4-Hour Workweek', I figured I could build a site, integrate WorldPay, roast the coffee, and use An Post to deliver with a 24-hour turnaround time! The good news is it worked! The bad news is that the business needed real setup capital for buying coffee roasters and was a very low-margin business. It needed bulk purchasing of green coffee, sourcing good suppliers, buying a second-hand coffee roaster, fumigation challenges, and not to mention the health and safety requirements for food processing... so I exited this startup quickly. Still, it showed me what could be done and gave me confidence that I could build, market, and operate a 'business' end-to-end if needed.

Milestone Detail
Product shipped Working e-commerce site with WorldPay payments and An Post delivery
Turnaround 24-hour delivery
Outcome Exited quickly - needed too much capital for a low-margin business
Key lesson Proved could build, market, and operate a business end-to-end

GoingOut.ie: circa ~2003

This was my second foray into a large content site, designed to be similar in size and effort as Freelance Ireland. The goal was to create a competitor to LovinDublin and to learn some of the lessons I had with Freelance Ireland, i.e., the content of Freelance Ireland was user-based, which meant at times there was little to no content - whereas a content website wouldn't have these issues, and after building up an audience, I could potentially create services/products based around that.

The strategy was to keep the site in slow growth mode for 1-2 years, build it up, let SEO take over, and grow the audience slowly (while I worked day jobs), and I could keep costs super low. The plan worked, but the content didn't. GoingOut focused on events, and unfortunately, events go stale very quickly! This meant constant updates and curation. This was also my first time experimenting with 'pipelines' and ETL systems to crawl online content and then format the content for the event search engine. All in all, it was a great technical project to work on and gave me another showpiece for my portfolio. As a business, it really required low-cost curation and employees, which I didn't have and couldn't afford. So GoingOut was closed.

It was around this time I started to realize that the Irish market was just not large enough for a 'slow growth' tactic...

Milestone Detail
Strategy Slow-growth content site, SEO-driven, low cost while working day jobs
Tech First experiments with ETL pipelines and web crawlers for event aggregation
Problem Events go stale fast - needed constant curation and staff
Market lesson Irish market too small for slow-growth bootstrapping
Outcome Closed

shanedevane.net: circa ~2007

This was my first real attempt at a professional brand site. I already had shanedevane.com, but that was focused on graphic design and frontend work from my earlier freelance days. The .NET domain aligned nicely with where my career was heading - Microsoft's .NET ecosystem was the thing to know, and I was moving into that world. So I built shanedevane.net as the technical version of my online presence.

It actually started life as a MediaWiki installation - basically a personal wiki where I dumped notes, CV stuff, and links. Later I rebuilt it as a proper ASP.NET site, and eventually used C# MVC 1.0 as a portfolio piece when that came out. I was writing articles on everything from Google Analytics custom reports and keyword strategies to Clean Code book notes. The idea was to position myself as a .NET developer and IT consultant in Dublin who also understood SEO and web analytics - and to practise what I preached by keyword-optimising the site for terms like "IT Consultant Dublin" and "Web Developer Dublin".

In the end, the effort versus reward wasn't worth maintaining two separate sites. I started folding the .NET content back into .com, and as I got deeper into larger side projects in the evenings, splitting time between both just didn't make sense. So SD.NET was quietly absorbed.

Milestone Detail
Purpose Professional brand site and technical portfolio
Started as MediaWiki personal wiki (~2007)
Rebuilt as ASP.NET site, later C# MVC 1.0
Content Articles on Google Analytics, SEO, Clean Code, .NET development
SEO targets "IT Consultant Dublin", "Web Developer Dublin", "SEO Dublin"
Outcome Content merged back into shanedevane.com

indexmark.com: circa ~2008

Enter LinkChecks! This started life as indexmark.com around 2008. In 2011, I moved the free version to link-check.org as an experiment in open source, with the idea that indexmark.com would become the premium version. I decided I wanted to build a tool that I could use and then perhaps monetize - learning from my past experience, I focused on a simple tool and a simple concept - check if the links on your website are working! Basic, simple, easy. A user enters a URL, it would kick off a background crawler, crawl a website and all the links, store metadata such as meta tags, image amounts, broken images, keyword density analysis, duplicate content detection, canonical tags, H-tag analysis, and link juice tracking, and then email a report to the user complete with old-school .xls file! Super, everything someone could need to check if their website had good or bad SEO. I offered it for free initially and started to build up a steady user base, nothing too special, maybe 20-30 users each month (given that competition in this space via ahrefs and Moz and of course, Google Search Console!).

This project was great, again, as far as technical projects go. It unfortunately broke often (as web crawling often does), took too long (10+ hours if there was a backlog), and required constant code updates. At the time, my goals for this were to explore and build in .NET and I wanted to learn more web crawling. In hindsight, this 'version' was a pure MVP and after initial validation should have been rebuilt in Apache Nutch (very fashionable at the time). LinkChecks also suffered from the explosion of React.js Apps and jQuery frameworks which didn't take kindly to crawling!

Milestone Detail
Product shipped Free broken link checker, website analyzer and SEO tool
Output .xls report emailed to user
Monthly users ~120
Competitors Ahrefs, Moz, Google Search Console

go2conferences.com: circa ~2012

This was a global conference search engine - users could search for conferences by country, city, industry, and date, and conference organisers could submit their own events. It was basically the same ETL pattern I'd developed with GoingOut.ie, but applied to conferences on a global scale.

I built a set of batch crawlers that scraped conference listing sites and ran them through a parsing pipeline to extract, deduplicate, and categorise the data into a SQL Server database. At its peak the database held 113,381 events across industries like pharma, tech, telecoms, and education. The front end was ASP.NET MVC with search, per-country pages, and forms for users to submit conferences or sign up for alerts.

Same story as GoingOut though - keeping aggregated content fresh and deduplicated was a constant battle, and the crawlers needed ongoing maintenance. I never got to monetisation. It was a solid technical project and a deeper version of the pipeline work I'd done before, but the operational overhead of running a content aggregation product as a side project was, once again, the thing that killed it.

Milestone Detail
Lifespan ~3 years (2010–2013)
Events indexed 113,381 across multiple industries and countries
Data sources 100+ conference listing sites crawled
Tech ASP.NET MVC, SQL Server, batch crawlers
Features Search, per-country pages, user-submitted events, alerts
Outcome Never monetised - closed

Newslinn.com: circa ~2016

Newslinn was a citizen-to-journalist photo sharing network. The idea was simple: when something newsworthy happens, the people on the ground have the photos first — but they have no way to get those photos to the journalists who need them. Newslinn gave them that connection. Citizens could share photos of events and incidents directly with verified journalists, and the platform included photo verification technology so journalists could trust what they were receiving.

In 2016, we won a grant from the Google Digital News Initiative (DNI) Innovation Fund — one of around 128 projects funded across Europe, and one of only two funded in Ireland (the other was The Irish Times). That grant was the catalyst — I co-founded Newslinn in Dublin with two others and we used the funding to build the platform and develop the real-time photo verification technology. We were also accepted into Google's Adopt a Startup mentoring programme, and made the Sunday Business Post's Top 100 Start-ups list in September 2016.

The longer-term vision was to bring the service to developing countries using SMS-based news sharing for people without access to smartphones. Total funding raised was under €100K. At its peak, we had over 400 journalists on the platform worldwide.

Newslinn was the most externally validated project I'd worked on up to that point — Google funding, press coverage, a real user base of journalists, and recognition in the Irish startup ecosystem. The challenge was always the same one: building a two-sided marketplace (citizens and journalists) is hard, and scaling it required more capital than we had. Newslinn closed around 2019.

Milestone Detail
Lifespan ~3 years (2016–2019)
Founded Dublin, 2016 — three co-founders
Journalists on platform 400+ worldwide
Google DNI grant 2016 — one of ~128 funded projects, one of only 2 in Ireland
Total funding Under €100K
Recognition Sunday Business Post Top 100 Start-ups (2016), Google Adopt a Startup
Outcome Closed — two-sided marketplace needed more capital to scale

startupteamlabs.com: circa ~2017

This was a content blog about building and scaling startup teams. It ran on Ghost and covered topics like team lead evaluation, boosting morale through showcasing achievements, and increasing empathy to scale teams. Each post was published as an 'MVP version' - a deliberately lean format that I could iterate on. I also created all the illustrations for the site by hand - marker-drawn visuals of meetings, sprint boards, retros, and team dynamics - which gave it a distinctive look.

The illustrations also became the basis for a Planning Poker app I built in React Native for both Android and iOS. Each Fibonacci number in the sequence had its own unique hand-painted character - a yellow smiley with glasses for 0, a red bouncy guy for 2, a basketball jersey for 5, a skull for 13, and so on. There was also a side experiment with developer-themed t-shirt designs.

It was a short-lived project but a fun one, and it captured a lot of the thinking I'd built up from years of working in and around startup teams.

Milestone Detail
Platform Ghost blog
Content Articles on team culture, leadership, lean processes
Visual style Hand-painted marker illustrations
Planning Poker app React Native (Android & iOS) — hand-painted Fibonacci characters
Active Late 2017 – 2018
Outcome Closed

ScrumLeads.com: circa ~2019

This came out of my time in New Zealand. The idea was to build a scrum master agency - recruitment, outsourcing, and upskilling, all focused specifically on scrum masters and agile leads. The tagline was "Helping Teams Deliver". I also set up a Scrum Community on Slack where teams could get dedicated content and private company channels.

I went fairly deep on the groundwork - contracts, onboarding forms, marketing materials, a recruiter rate card, and job description research (I was scraping Glassdoor NZ and Indeed for scrum master roles to understand the market). I also explored funding through a FasterCapital regional partner agreement.

I liked the concept, but it needed a sales engine and a network of paying clients that I just didn't have the time or capital to build while in NZ. It never launched publicly.

Milestone Detail
Concept Scrum master recruitment agency, outsourcing, and community
Tagline "Helping Teams Deliver"
Community Slack-based Scrum Community with company channels
Location New Zealand
Outcome Never launched publicly - lacked sales engine and capital

Produxly.com: circa ~2019

Produxly was a visual content brand -"Product + User Experience + Agility == Produxly". The core asset was a library of over 100 original diagram illustrations I designed in Photoshop, covering things like go-to-market strategy, UX mental models, sales funnels, and digital transformation frameworks.

The site was built on Bootstrap with a landing page, a slides gallery, and a pricing page with Free/$9/$49 monthly tiers - the plan was to sell access to the diagram library as a subscription. There was also a Twitter and Instagram presence, and I experimented with branded merchandise (t-shirts, leggings).

Produxly was related to startupteamlabs but took a different approach - where startupteamlabs was hand-painted marker illustrations and blog articles, Produxly was polished, digital, and designed to be a sellable product. It never got traction as a subscription business, but the diagrams themselves ended up being useful assets I kept coming back to.

Milestone Detail
Concept Visual content library for product, UX, and agile professionals
Content 100+ original diagram illustrations (PSD)
Site Bootstrap landing page with pricing tiers (Free/$9/$49)
Tagline "Product + User Experience + Agility == Produxly"
Outcome Never gained traction as a subscription - closed

Vunch.co: circa ~2019

Vunch — short for "Virtual Lunch" — was a networking prototype from mid-2019 in Auckland. The concept was structured video calls over lunch breaks as a way to grow your professional network. I went through YC Startup School (not Y Combinator) that summer, built a landing page and branding, and started a Medium blog — but it never got past pre-launch. Gohono had real traction with real venues, so that's where the focus went.

Milestone Detail
Concept Virtual lunch networking prototype
Built Landing page, branding, Medium blog
Outcome Shelved — Gohono took priority

Gohono.com: circa ~2019

This one actually started offline. In 2018, myself and my co-founder were organising board game nights and coffee meetups in cafés in Auckland, NZ. We were doing it manually - finding venues, coordinating small groups of people with shared interests, getting them to show up. It worked surprisingly well, and we kept thinking "this should be a product".

So in 2019 we founded Gohono and built a platform that did what we'd been doing by hand - it helped people with shared interests find each other, chat, and then meet up at local cafés and restaurants. Venues could control when and how groups used their space, and customers could join just by scanning a QR code.

We tested with a couple of venues in NZ, logged 399 customer engagements, got press coverage in three publications, and later made the semifinals of the HORECA Challenge - a startup competition in Barcelona with 100+ applicants. We didn't make the finals, but it was good validation.

The product was real, the engagement was real, but ultimately we didn't have the capital to scale it. The core lesson from Gohono was that the concept of helping small groups meet offline at local venues has legs - the manual version worked in 2018 and the platform version worked in 2020 - but selling to hospitality is a brutal sales cycle when you're bootstrapped. The other lesson was about the value of starting with the offline version first: we knew the product worked because we'd already done it by hand before writing a line of code.

Milestone Detail
Customer engagements 399 recorded during venue testing
Press coverage Featured in 3 publications
HORECA Challenge Semifinalist from 100+ applicants
Product shipped Working platform with venue dashboard, customer apps, and analytics
Venue partners 2 venues tested in NZ

treb.ly: circa ~2022

This was an interesting one. This was my first attempt to build a 'social network' based on a b2b model - it was a prototype Meetup clone designed specifically for large sporting brand in Spain. The goal was to showcase that the brand could engage it's customer base on it's own Meetup version specifically for it's customers, as part of a 'customer rewards' concept, whereby it's customers could engage and oragnize their own Meetups but without their customers having to pay Meetup / nor having the overhead of organizing on-going Meetups (and all the hassle that comes with maintaining a group) - the name comes from the concept of 3 ie. the minimum amount of people needed to form a group. This got past the idea phase and a prototype version was built using the stack I'm familiar with, Django, Channels (web sockets), AWS. Sadly this never got past the prototype stage and I didn't have the capital to continue development/selling. Still, the overall 'implementation strategy' of building a prototype and then sourcing a venture-customer is something I will repeat....and perhaps re-surface the idea again especially now that AI in 2026 makes building easier.

Profesica.com ~ Antler: circa ~2023

This was part of the startup Antler programme in Madrid. Didn't get investment nor '2nd stage' but it was a good experience. I'm a fan of the Antler startup programmes, they can be a mixed bag in terms of value but the fact it exists and is actively getting better is a huge win for the global startup community. Will definitely write more about this at a later stage.