If anything, this should be a cautionary tale about outsourcing your development work to India. Especially for a small amount like that, and then expecting something solid.

Its awesome to see companies coming out and admitting their mistakes and reasons for failure.  It provides tried and tested advice for those venturing on a new business or thinking about it. Well done Elan and team!

Firstly thanks to Craig for quoting me accurately: 

Some other things I would like to share about the experience:  - 

1.) With Zappon we tried to take a different approach to what would normally happen in a corporate. There was no extra capacity to work on Zappon and I still thought we need to get there fast. When we decided to move on it Groupon had not yet acquired Twangoo – so it looked like a window. – and as such the RACE was on. 

What I will be proud of despite how it went down was that in a matter of TWO months we built a product, assembled a great team and sold a pipeline of 6 weeks worth on deals at a time when suppliers understood little of the concept. That was 8 weeks of no sleep while we were all doing our normal jobs, ignoring loved ones etc. It felt like an amazing feat. (little did we know how the first week would turn out) 

My goal was to do it on a shoe-string out of a garage and it had no internal involvement from Avusa HQ. (a brave move on their part) 

In fact I was paying people on my credit card at a stage to avoid delays on the usual corporate red tape.

In hindsight we should have taken an extra couple of weeks and mitigated risk on tech. 

Another issue was that these kind of things need focus – dedicated focus. In a corporate where you have many other things on the go that is hard to achieve despite your best efforts. 

2.) When we tested with our Indian partners everything worked 100% running on a single server. At launch we load-balanced across two servers and that is were the code all came unhinged. All aspects around logins, caching, payments went belly up – which invited (and rightfully so) a barrage of attack from the social media community. We took the risk and paid. 

Other sites had the luxury of launching in obscurity and slowly building up via user feedback making incremental improvements. The big-bang approach was necessary at that stage in the game but always a major risk.  

Honestly I thought in the first week we would have to issue a release in week 1 to say we failed. It was one of the hardest weeks of my life. We just kept saying “eye of the tiger” and fighting fire after fire ignoring the abuse from the sidelines and focussing on the goal. The team spirit got us through. 

A local tech company stepped into help us but in the end it was too late. Our tech partners had used PHP cake as a framework and implemented it badly with the separation of logic and essentially it was a minefield for new developers to deal with. 

3.) So you can imagine we had a young team who thought they would be rock-stars suddenly dealing with all of this external and internal pressure. I commend them for doing this well while us senior guys took it on the chin. I have to say despite the outcome I am proud of the team and would work with them again gladly on other future projects. Ironically we all had a lot of fun and learned so much about this space and start-up challenges. 

Also though we had some unhappy customers on the whole I think we did our very best to put the customers first when mistakes were made. This definitely was a priority. 

The sales team also exhibited great INTEGRITY out in the field. Our attitude was to always look for a win-win and not deceive suppliers for the sake of a deal. They did this with flying colours and I am proud of their conduct since this is a space where one can easily take advantage. 

4.) On the value of print – yes print is not a great driver of traffic but it is good for brand. BUT the key strategy there was when talking to restaurant owners etc who are skeptical about online, print and our brands such as Sunday Times and Times is something they understand. As a result it made getting deals easier and we never had an unhappy supplier. Even if their deals did not sell they got to be in the paper – which made them happy. 

5.) On marketing – We also as publishers were not that sharp on optimising performance marketing for retail which is a specific skill. I think we learnt much around that topic but it would have been better to get out of the blocks with that mindset. 

6.) The quality of our deals were not good enough. And for that matter I see that as an industry issue. Nobody has adequately nailed this. Us senior guys did not have the capacity to go out and do the BIG deals we needed to be doing.  There is not enough skin in the game for the supplier. 

7.) I will be 100% open and say that on net  this project costs us R1.5m  (from the first to last paperclip all costs included) – which in this game is negligible. Less than some of the more aggressive players are spending a month on marketing acquisition costs.   I am sure most of the other players have burnt past that a long time ago. We wanted to get a taste before we went overboard and made a really expensive mistake. I don’t want to the be group-buying site who burns R10m+ and shuts shop. 

You may say this is a case of being half pregnant but …..

The reason we took a non-aggressive approach was because it was touted to be a gold-mine and we were always not sure about that. We wanted to test the water though the mistakes we made did not allow us to do that adequately. 

It’s a hard and dirty business having to go and get those deals and manage customer service. Much more than putting up a site. It is a lot harder than it seemed. 

Overall, I have no regrets whatsoever. 

Elan

Taking this to a personal level is petty and not cool. There is much I could say about that comment but frankly it is not worth it.

I was one of the buyers on Zappon.  I got the fantastic deal of a 3 months of Sunday Times for R5.  At the end of the 3 months I signed up for a year!

This doesn’t seem like it was an expensive experiment, and it had plenty of upside, so if I was an Avusa shareholder I wouldn’t be dissapointed that it was tried.  In business, if you can hit 1 home run for every 5 tries you’re doing pretty well. 

At the moment the group-buying market is dominated by Groupon, who seem to be buying up everything that looks like an ad space.. with their massive marketing overspend they make it tough for pretty much any competitor to really make a dent in their presence.

I say well done to Elan and his team — they did great even though their site architecture let them down. There’s no shame in Zappon failing, it’s far better to have tried and
failed than not tried at all.

It’d be a terribly boring world if no-one
started up something because they were afraid of failing, and  in my humble opinion South Africa needs more folks like Elan willing to take the risk in starting up something new and exciting.

“When we tested with our Indian partners everything worked 100% running on a single server. At launch we load-balanced across two servers and that is were the code all came unhinged.”

I am aghast.

“But it worked on my machine” is not, and NEVER has been a good excuse for why things croak when you launch them.  Anybody worth their salt in this industry could tell you that your production cluster and your staging cluster have to be identical to avoid this sort of thing.To try and excuse yourself by claiming that you didn’t have the luxury of a slow launch and a quiet build-up is ridiculous. You knew what you were getting yourself into – most of your resources should have been put into ensuring you had a rock-solid scalable platform.

I can’t take any of the rest of your comments seriously if you can try and slip that past us.

biting my tongue right now……

As long as Avusa is still willing to take calculated risks and not shy away from new ventures then the lessons learned would be worth it.