Category Archives: Starting up

Starting up: The £4k Startup

benjbins

My “office” next to the bins

A version of this post first appeared on FastCoLabs entitled “How This Journalist-Turned-Coder Built His Startup For $6,000″

Over the last few months, I’ve been writing about my startup adventures. First, on why I decided to leave the Guardian, learn to code, and do this in the first place. Then on the crude intricacies of chasing rich people around London. I left you at the bit when I decided to stop chasing money and focus on finishing the thing. I was gonna get it to the cliff, and over it, myself.

It’s at the cliff. I have a product. For the last two weeks, a handful of friends have been trying to break it. In wanky startup parlance, this was my “alpha” stage. This post marks the beginning of the “ private beta”.  Then it’ll probably sell to NASA for eleventy billion and I won’t talk to you any more.

Here it is: picfair.com (not much to see, drop your email in for a beta pass)

In the meantime, I’ve been doing some maths. All told, I’ve spent under £4k on the whole thing. Here’s how:

The product: £890

I started writing the code in February, beginning with the core structure in Ruby on Rails, and the basic front end stuff (HTML/CSS/Javascript&jQuery). I learnt how to code on evening courses last year after leaving the Guardian, and gradually fell in love with it – I genuinely now find writing code more creative than writing words. Between the General Assembly course, Stack Overflow (it’s like Quora for code), and the never-ending generosity of my tutor Rik Lomas from Steer, I’ve been able to get over every technical hurdle I’ve come across, albeit very slowly, and painfully.

aaaarg

Sometimes code makes you angry

There have been plenty of times I’ve hit a wall and thought, fuck it, I’ll throw some money at this, but then decided against it. By writing the code myself, I’ve got a very good fallback if it all fails, which it won’t, obviously. But if … I won’t just have the “experience” of building a business, in a spiritual I’ve-really-found-my-calling kind of way, I’ve got tangible new skills. I’ve gone from being a shit developer to being an average-to-ok one, and I’ve already made the money back that I paid for the courses several times over through freelance code commissions like this one and this one.

But average-to-ok isn’t good enough, so I paid two experienced “code consultants” to pillage my site after I’d done all I thought I needed to. A Ruby dude went through it looking for security holes and superfluous code, and, inexplicably, found nothing too horrific. A front end developer did likewise, sprinkling his significantly-better-than-mine CSS nous across the site. Nothing too major, but a noticeable upgrade. It looks hot, and it works.

tim

Front end dev Tim Parker up in my code

Legals: £1300

The basic stuff you need for a website – global copyright notices, privacy policy, basic Ts&Cs and disclaimer – you don’t have to pay for, you can write them yourself. They’re all pretty much the same. The gov.uk website has free sample templates you can use – just Google “businesslink + [disclaimer/privacy policy/etc]” and you’ll find them. A few tweaks and you’ve got your own.

However, my website involves the transferral of digital property, requiring a seller to grant a license to a buyer. This is a minefield, involving all sorts of scope stipulations and indemnification clauses. Mistakenly, I thought I could do this myself – I spent a week rifling the web for similar licenses, chopping them up and re-wording them to suit my specific transaction. I finished them and triumphantly showed them to a lawyer friend – “Look – I’m a lawyer now!”

They stank. So I found a solicitor to do them properly. A minor victory: my stinky legals were still better than nothing, so the instruction was a “revision” rather than a full draft from scratch, saving me about £400. The £1,300 still kind of pains me, because I feel a bit like I’m paying shitloads for someone to change “from today” to “henceforth accordingly”, but hey ho. I’ve salvaged some of the stuff I wrote for human-readable deeds over the top of the licenses.

Branding: £1250

Again, I thought I could do this myself. I’d made a logo, had a pretty good idea of an overall design principle and colours, and used to equate “branding expertise” with snake oil. I still think that, a bit. But I also realised the damage that shitty branding can have on a product, and was worried that my logo and overall style was veering towards the shitty, careless end of the spectrum. So I found a very good guy who put together a logo and a top-level style guide for me – something for me to work from.

logoforblog

My logo. I like it so much I sixed it.

Very much worth it. Together with the front end “code consultant” this top-and-tailed my design build. The branding guy sent me on my way, and the code consultant cleaned up any mess after I’d finished.

The domain: £330

Choosing a name for your business is incredibly annoying. I spent days talking gibberish to myself. I’ve got pages and pages of nonsense in my journal – portmanteaus, words that end in “r”, quirky territorial domain endings (.ie, .io, .ck etc). I even registered three of them in piques of certainty – mirapic.com, picpacket.com, and picrise.com (want them? £20 each. Ok, for you, £18)– before deciding they were rubbish the next day. I abused the Domain Scout app on an almost minute-by-minute basis. At one stage I think it told me I’d “already searched for that domain 13 times”.

scribbles

Stupid bollocks in my journal

I was determined to get a .com. I know this is old-fashioned, but it’s also what everyone assumes. Especially when they are looking for something. I settled on Picfair, which was already taken, but available on SEDO for $1,000. I got it down to $500. This still slayed me. $500 for a word. But I was set on it, and it’s great. Easily the stupidest money I’ve spent on this whole thing, but hey.

Testing: £200

Two weeks ago, I finished the site, and unleashed 10 friends on it, challenging them to try to break it in return for dinner. They all broke it, pretty much. They all had plenty of suggestions too. Invaluable, and maddening – I had a total of 73 bugs to fix and tweaks to implement. But I think – THINK – It’s just about ok now. I’m gonna regret saying that.

So… totting up:

The product                              £890

The domain                               £330

Legals                                      £1300

Branding                                  £1250

Testing                                       £200

Total                                                 £3970

**Disclaimer**
I realise, of course, that there has been a significant “opportunity cost” in making this. Dedicating the majority of my time for the last four months has meant, yes, I’ve lost out on freelance commissions and some paid work. But despite operating at a minor loss, I’ve gotten by, and haven’t missed rent or starved.

And also, more importantly, this is the final blog I’m gonna write before releasing the thing into the wild, so, with my recently-learned legalese: I hereby reserve the right to be a total perma-shill, shameless promotion-obsessed pain in the arse once it’s launched. You’ve been warned.

Starting up: Chasing money

Screen Shot 2013-05-19 at 20.02.20

Three weeks ago I wrote my first post about starting up: a summary of my last year – leaving the Guardian, learning to code, and deciding to do a bizniz. I left you at the bit when I was about to go chasing rich people around London. Here’s what’s happened since:

“Smart money” vs “Dumb money”

When hunting cash, you need to know what kind you’re looking for.

In essence, “smart money” are investors with form: angels or funds with a history of investing in startups, or with experience in the industry you’re aiming to join. The “smart” bit is the added value you get with the cash – advice, contacts, experience. And mostly … a fair old dollop of PR. If a “smart” investor backs you, it’s a story – your startup is given an immediate credibility injection, and its first press release.

“Dumb money” is everybody else. This could be a friend’s rich uncle, a floozy fund looking for anything worth a pop, or a City prospector wanting to get in on some of this Silicon Roundabout thing. A big part of their draw: the SEIS scheme, giving investors a 50% tax break on investments up to £100,000 in early-stage businesses, and exclusion from capital gains on any profits made. I don’t like this smart/dumb distinction. As far as I’m concerned, if someone has a few hundred thousand to risk on a chunk of a startup, it’s very unlikely they are dumb. But anyway …

Hunting

I started with a list of seven “smart money” investors. Here’s how I found them, and made contact:

1. Read lots of TechCrunch Europe and Tech City News pieces about recent angel/fund movements and investors, and have a proper rifle through AngelList. There’s plenty of investors out there, but they aren’t particularly visible. For obvious reasons: They expect to be sought out, and startups need them more than they need startups.

2. Get their email addresses. In order:
• Work out of you’ve got a mutual contact, and ask for an intro, preferably bigging you up. If you don’t have any, look through the people they follow on Twitter for anyone you know, and go via them.
• Look for their personal blogs, there’s often an email address on them.
• Look through their Twitter history for anyone else who’s asked them for their email address and got a public reply
• Ask them outright on Twitter… “mind if I drop you an email?”

3. The email. Only one of the emails I sent hasn’t been responded to –  the only one in which I didn’t mention The Guardian in the email title. This is one of the few advantages I have as a first-time founder – I’m not just a random dickhead starting a business, I’m a random dickhead who used to (still sometimes does) work for the Guardian starting a business. I kept the emails short – a brief explanation of who I am, a brief summary of the business … wanna go for coffee?

Meeting

Of these first seven: one never replied; one passed immediately because she isn’t currently investing; one I’m meeting later in the month; and four I’ve met.

Two of them I probably shouldn’t have. I’m not sure if this is a bad thing or not. If I’d done my homework properly, I’d have realised that these two funds have never invested in anything without proper traction (i.e. a company that exists, and has plenty to show for itself), and usually invest well above the kind of cash bracket I’m looking for.

This isn’t to say they were disasters. The first was in an office in Mayfair. I did my pitch – a five-minute “deck” talking through exactly what I’m building, with lots of figures and market research, then we talked. I didn’t fuck it up, and my business lasted half an hour with a serial investor without falling apart.  But the punch was was in the first few minutes – he explained how his fund worked, and I realised there was no way they would invest in me. I was too early, and too low.

The second was in a coffee shop in Shoreditch (Ozone – it’s the de riguer Silicon Roundabout meeting spot, I’ve learned). She was furiously smart. She didn’t order coffee, she just wanted me to get on with it. She stopped me before the end of my pitch and asked me straight up what I wanted. Crap cut. I explained: cash to pay a lead developer / technical co-founder as well as offering them equity, so I can focus on marketing post-launch. She tore into my logic – why would you pay a co-founder who is getting equity? Then she tore into my idea – how was I going to get traction? Why would I succeed where others had failed? How big is the market? I think I did OK. Not brilliant, but OK. Then I showed her what I’d built, and her good cop emerged. She was impressed, and started emailing a few people about me. But, again – ultimately, I’m too early, and too low.

In a film, these scenes would end with the investor saying “well, we’ve never invested in anything at this stage, but I’m in a good mood and you seem nice, hell, have half a mil, it’s in my bathtub, go get it.” This is not how investors get rich.

The third was with an angel who has invested in a dozen companies in the last 18 months, a couple of them pre-launch. In my head, he was the one. I’d read everything I could find about him, and had found a video in the nooks of YouTube where he explained that he always knew if he was going to invest in someone within five minutes of meeting them.

We went for a pint in Clerkenwell, and before I’d even started he explained that he was “going fishing” for the next half a year – he wasn’t looking to invest in anything. Bollocks.  I’d somehow failed his five-minute test without saying anything. This briefly derailed me. He liked me, and  my idea, and was impressed with what I’d built. Then he asked “what’s stopping anyone else just copying it?” I gave a terrible answer, essentially saying “nothing”. I now have 23 better answers I could have given. Next time someone asks me that, I’m gonna answer their face off. Anyway, watch set for six months.

The fourth was with another fund, at the right money level, again with a few examples of pre-launch investment. We met on the top-floor of a shared workplace in Clerkenwell, and, as far as I could tell, it went very well. But this, of course, means nothing. I’ve followed up with a few emails, and I’m waiting.

But then …

Back to the code

Then I realised that I don’t need the money as much as I thought I did. Not yet, anyway. I’d worked myself into a money-hunting frenzy, convincing myself that I couldn’t possibly launch without smart money and a Ruby ninja behind me. I’m confident I can get both, but this process could take months. I’m impatient. I’ll keep up the hunt, but in the meantime ….

I’ve spent the last week diving back in to the code. I’m gonna get this thing to the cliff myself, because it’s just so… damn…. close. I’ve set myself a deadline of June 3rd to have written all the code I want to write. Then I’m gonna spend a bit of money on two code reviews. I’ve lined up two very good devs, both of whom I’m gonna kidnap and lock in a cave until they sack off their jobs and join me. A Ruby dude is going to pile through my back end (no sniggering), while a front-end dev is going to pillage my CSS, design and UX. Any small tweaks, they’ll do …. anything major will go into a List Of Possibility, which I’ll whittle down to absolute vitals. I’ll either do these myself, or see if I can nag them into the consultancy deal.

After that, I’m gonna get 10 friends to use it and tell me all the reasons they don’t like it (hopefully not too many), followed by some final tweaks. Then I’m gonna launch the fucker.

I don’t know if any of this is the right way to do it,  but it feels more productive. And, more importantly, it’s quicker.

In the event that my startup bombs, at least it will have been swift, I’ll have spent under £5k on it, and I’ll have learned a shit load of code – arming me for the next one. If it doesn’t, my business is a hell of a lot more attractive to investors.

Onwards.

Starting up: An intro

logoforblog

Almost exactly a year ago, I decided to leave the Guardian. I didn’t really know what I was doing. I’d been travel writing for them (and others) since my final year of university and had moved in house as a writer and digital producer in 2009. A good stint, but time to go.

I spent a few weeks shouting at the moon, did a few freelance gigs, and then signed up to a General Assembly front end development course. I knew I wanted to work with developers, but didn’t know that I would actually become one. I loved it. I found myself waking up at three in the morning to move a <div> four pixels to the right. The thing that surprised me the most: I found writing code just as – if not more – creative than writing words.

I completed that course and released my final project – The Reddit Edit – into the wild. It did nicely, and was featured by The Atlantic, Mashable, TheNextWeb, and others. I used it to flog my next project, Interactive Stories – essentially pitching myself as a freelance interactive news designer. This has done well too, and has been commissioned by The Guardian and Rough Guides. Together with other writing gigs and a bit of consultancy, it paid the bills. I even got a few job offers from some notable meedja houses, but I knew it wasn’t right for me to go back.

Then I did a Ruby on Rails course – the back end to my front end (non-devs: this is the database building stuff, it’s fucking hard). I’d been sitting on a startup idea for a few years, and this meant I could actually start building it. After the Ruby course, I was incredibly proud to say I was a bad-to-OK developer, but that wasn’t enough. I did a deal with my horrifically-talented friend Rik Lomas who had taught the GA course: he would tutor me in Ruby and fix my bugs once a week in return for copy-writing help and marketing advice for his code-learning startup Steer (they’re great, give them all your money).

I am now dangerously close to having a startup. So I thought it would be cathartic and mildly interesting for others to start writing about the process. The rules: I will spend no longer than 30 mins writing these posts, I will not guarantee their frequency, and I will not apologise for humblebrags or flagrant self-promotion.

What is the startup?

It’s a search engine for brass instruments called Boogle. Not really. I’m not going to tell you yet. But, a few clues: it’s based around images, and money. It’s borrowing a model previously used to wake up a very old reeling industry, which I aim to apply to another very old reeling industry. It isn’t porn.

Does it exist?

Almost. I’ve built most of it – the back end in Ruby, the front end in HTML, CSS & Javascript. I’ve drafted the licenses and built marketing databases. The only cash I’ve spent is on a logo (that’s it, above, tiled, for no particular reason) and a little branding. I’d say I’m 90% there.

What am I doing now?

I’m steadily approaching my technical ceiling. Well, that’s maybe a bit unfair. I’ve found that pretty much every technical hurdle I’ve been faced with is surmountable, with a little help from Rik and StackOverflow. However, I’ve realised that I’m just not technically good enough to maintain the code, while managing the business, flogging the crap out of it, chasing potential clients and the like, and doing all the other stuff you need to as a startup founder. I know that an hour of my time is better spent as a marketeer than fixing a filthy CDN API-parsing RegEx bug that would take a better developer five minutes to solve.

So, I’m looking for a lead Ruby developer / CTO / Technical Co-Founder type person. I know that I can’t do this with equity, as 20% of nothing is nothing. So I’m simultaneously looking for  investment.

These are two things I’ve never done before – hiring someone, and raising money. I’ve just started, here’s my current progress:

Finding a CTO-shaped person

Due to the age-old shitness of computer education in the UK, there is a dramatic imbalance in the supply and demand of good computer programmers. Really good devs cost a lot of money, are rarely available, and – this is a good problem – usually have ideas of their own. The kind of Ruby dev I’m looking for can comfortably charge a bank £500 a day for their services. While I am offering them the chance to change the world and make loads of money, these kind of offers are magic beans to the savvy dev, offered to them on a daily basis. So I need to offer them cash too. If my startup fails, at least they’ve been able to feed their kids during the process.

I’m stalking Ruby forums and email lists, chasing contacts and reading dev blogs, and will begin my assault soon. I did get one fantastic Ruby dev very drunk last Wednesday, and pitched him after the fourth beer. I failed, as I knew I would, because he’s setting up his own business, but he agreed to be my adviser and help me with my dev hunting, which is very handy indeed. Onwards. If you are a very talented Ruby dev with itchy feet, I will know who you are within the month. You can email me too, if you like.

Finding investment

Now this really was something I knew fuck all about. I’ve spent the last little while reading as much as I can about how it works, and talking to lots of people who know more than I do. I honestly didn’t know what “seed funding” was until about a month ago. A handy guide: Friends and family funding (up to £20k) >> Seed Funding (£50k-£300k ish, from “angels” or seed funds) >> Series A (up to around £2m, usually from Venture Capitalist (VC) funds >> Series B/C (shit loads, from rich aliens). This is probably wrong.

Anyway, turns out I’m looking for “seed funding”. I’ve scoured old TechCrunch posts and Angellist and have made a list potential seed funds and angels in London. I have met a few other very nice startuppers like George from Rentify who is pointing me in a few directions too. I’ve made my business plan, my deck (this is a word I don’t like but it’s basically a presentation), and have got lots of friends to ask me lots ­of questions that might come up.

I made my first pitch to a seed fund person last week and I was pleased with it – I didn’t puke or accidentally headbutt them, and they didn’t laugh me out of the building. I have some more meetings lined up.

Soooo….

That’s my blog. I already broke one of my rules, as this has taken about 45 minutes. I’ll update you all soon, probably.