06 Sep 2010

3 great lessons from ngbot

ngBot has been my pet project since my first year at the university. Between that 7 years and now, a lot has changed both to the site, we its runners and the Nigeria web startup scene. But not just that. Running ngBot this far and following the naija startup scene; keenly watching the rise, hype, falls and all, has really taught me great startup lessons. Many good programmers are springing up; many Nigerian startups are coming up. But problem is, they are making the same mistake we made as well as others of our time.

It's not just a hobby
Number one problems is: to many of us, it is just a hobby. We get ourselves trapped in this [what's the word here?] cycle: have an interest in programming; turn out to be good at it; create something out of that cheer fun. The core problem with this is you will hardly take the startup serious - I mean business serious because ofcourse its just that other side thing.

Startups are business. They should be taken with that seriousity and commitment. They shouldn't be what you have time for that once in a while. They require proper time dedication. Ever wondered why some really successful startups/company founders dropped out of school? They had to take their little project more serious and give it more time.

Again I will repeat, building a successful starup requires a great deal of time dedication. Maintenance, iterations, scaling, customer service, bug fixes etc. For a month and there about, ngBot has been having serious scalability issues. However, namzo and I are on our final exams and sitting in front of a black shell console profiling the server, setting up firewalls, fixing config files, sniffing logs are the last thing on my mind.

Money baby, money!
Just proposing your startup? First rule - your monetisation strategy should come in straight up. Already launched and no monetisation plan? Well, good luck to you.

Really, monetisation is a big problem even some of the best of startups are having. I have seen a lot of great startups die because they couldn't hold on again for lack of money. Some were smart enough to kill their free model for paid subscription. Just some weeks ago, reddit had to openly voice out theirs. And come to think of it, even there, running startup is easier beacuse in most cases you can easily get funded till you grow very successful. It's different here in Nigeria. To start with, how many Nigerian angel or VC have you heard about? A lot of us run our startups with our personal earnings, which can turn out to be very difficult especially as the startup starts growing. So having a monetisation strategy from the onset is just so expedient. That you run a startup on a N10 grand/year shared hosting plan you can afford doesn't mean you shouldn't monetise. In fact that you are on shared hosting means you are probably just starting or your startup is not yet getting that traction or you are doing nothing that serious. If you run a VPS/dedicated server of about N10 grand/month, an internet bill of the same amount, not to talk of office and maintenance, then you will understand.

So how do you monetise? It's up to you and depends on what you are building or what you have already. Monetisation to most naija startups means contextual ads. Bad news however is that contexual ads only work best in certain environments. But that's a discussion of another time. I'd rather recommend a freemium based model. Direct sales of services and products (online sale of recharge cards, bulk sms solutions, etc) and paid subscriptions are even better. Reason that for one, your money starts coming in from day one.

After ngBot, there have been other ideas. But then the first question has always been 'how viable is this as regards monetisation?'

Communities are killers
If you can, I will say avoid creating communities. That sounds crazy right? But seriously do. Yeah, yeah, we are all social objects, and it can be fun to run a community. But another social networking site, forum, microblog etc may not just be it. In most cases, managing it can be more difficult than any other regular startup and has a lot of problems attached.

  • The input to output ratio is low. The ratio of the result you get to the time and dedication you put in is in most cases very low. And this will happen for a very long time till the community reach a certain stable level.
  • And then there is user retention issues. If users are not getting fresh content or actively involved, they get easily bored and away they fly. Thing is, user acquisition in communities can be easy but retaining them is a big one.
  • Monetizing a community is twisted. In most cases, the only way to do is via contextual ads which won't make sense till you grow a large user base; more appropraitely, a large site visit/usage. The best form of communities to easily monetise are ones that wrap round a certain type of service such that beyond regular free users, there can be paid users with access to more site features. devianArt as an example.
  • Did I forget to mention spam? Nah! This should even be the first item on the agenda. All startups fight spam but it is more difficult with communities. If you miss a day of maintenance, then you may find your whole site overrun by spammers. naijapulse was among the first set of microblogs in Nigeria that started out with great insights but the last time I checked over, it was spam ridden. ngBot mobile is really having a swell time with spammers too and sorry to disappoint, I'm yet to find a 100% foolproof way round it.

I have learnt my lessons and I'm glad I did. Now, I really look forward to taking the world with better ideas. But then sometimes it's not just about the rules. Some of us just get lucky. And on a lighter note, startups are fun, great fun.

What lessons have you learnt from your startups? Do feel free to share.

Follow me on twitter: @kehers

My name is Opeyemi Obembe. I build things for web and mobile. You should follow me on Twitter (@kehers).