Welcome to The Startup Foundry.
From media darling, to selling out: 10 mistakes I made with my startup.
Published: Jun 6, 2011 3:47 pm
This is a guest post written by Elias Chelidonis. Elias is a web entrepreneur and blogs at thewebcitizen.com, the guide for the web citizens.
Launching a web startup has been my passion since the early 2000’s while I was working on my MBA in the USA. I was fascinated by the speed of how quickly things moved online (and the whole business process in general). I knew I had to get involved. The internet bubble had just burst, but there was a strong momentum of young and passionate people eager to develop the next AOL, Excite or Webvan ( For those who are young enough these were the hot e-businesses of the time ).
After I had a few unsuccessful attempts to launch a serious scale business (and was actually about to quit trying), the idea of online file storage came to me. I decided to launch Orbitfiles.com, and it appeared I was at the right time and place. We launched in early March 2006 and on April 23rd 2006 our service was one of the most talked about web properties, and we were even on the first page of Digg.com. For those who remember, Digg.com at that time was the today’s Twitter, once you hit first page, you were rewarded with a huge wave of traffic! Servers crash and you get gazillions of visitors. We were not expecting things to move so fast. Our service recovered in 5 days and the hit started to cool down. By the end of May 2006, I got a call from Google for potential acquisition as at that time they were planning to launch Gdrive. Unfortunately, nothing came from it.
It was an unforgettable journey, and I thought I would share 10 critical mistakes we made ( taking full responsibility as I was the CEO ) that I will not repeat for my upcoming startup (Hint: I will be a storage guy forever ).
Here are some of the mistakes we made
- Paid too much attention on web design. NO-ONE CARES as long as it is visually acceptable. What it counts the most is the design of the interface, UI. Make it simple enough that even a 10 years old kid can navigate and find all the info it needs. Although, we had this correct initially, we decided to change the UI to a more web2.0 style and the whole thing messed up, was too complex and the rest is history. Average user’s brain remembers only easy things. Tip: If you say “Wow, our platform rocks, very smart “then it means only you can use it but if you say “Oh man, it looks so damn stupid “then bingo!!! Keep it too simple.
- We had a solution for everyone, consumers, businesses, free users. This is acceptable in the beginning till you find your way and create momentum (build back links, traffic) but later you must decide what you stand for in your users mind. The reason is that it will help you allocate all your resources to this target market. In our case free users were abusing the system (as often happens) creating server slow down to business users, the two different target groups’ needs were in contradiction, it just does not work. You must decide whom you want to serve at some point. No such thing as all you can serve.
- We had very low prices. This was one of the most crucial mistakes, initially we thought the cheaper we are the more users we will attract. Well, we did, all the spammers, abusers and people who thought for $5 a month can own half the universe. Do your math and charge wisely, your prices will also determine your target group. Higher prices attract higher quality customers that require less support and are less demanding because they know that nothing is free.
- Focused on growth and not profitability. If you decide that for the next year you will focus on building a big audience, you need to consider these 2 factors:
- Does your business require heavy bandwidth usage – Online file storage services require huge bandwidth by nature, which means huge bandwidth bills. We started with just $100/month of bandwidth and in just 2 months the bill was $15,000/month.
- Funding – If you have deep pockets then got for it, if not, focus on profitability. I know your chances of becoming the next big hit are small but that is life.
- We did major changes in the user interface without asking our users first. Huge mistake as we were smart enough (sarcasm) to change the entire UI so when users logged in, it was a totally new platform for them. Digg.com and Myspace.com did that and now are suffering. Humans hate change, so do your changes wisely and not at one time and most importantly survey your users before proceeding.
- Did not hire support team and did not have in place a good support platform. By not having sufficient revenues due to mistakes 2, 3, 4 we ended up doing the support. It is ok at the beginning but as the business grows you need a support team. As a result we got so tired doing it so we gave up and ultimately users start to vanish in a matter of days. We had a huge drop of 15% month to month in the conversion from free to paid users ratio after we became sluggish in customer support.
- Wasted time trying to be acquired. During the early web2.0 mini bubble everyone wanted to be acquired so we ended up wasting almost 2 months trying to find a buyer. What we managed to do is lose focus of our core business and ultimately give up. You build a business to solve a problem and grow this business to a profitable entity and not be acquired. If it comes is acceptable but do not look for it unless you are the next Twitter.
- Did not have a good backend system. This is very crucial since it is the head of the entire business. It needs to be consisted of functions that help you have full control of your business at any time. For instance, include key metrics such as total number of users, total number of free users, total number of paid users, total number of users per subscription plan, total number of uploaded files (in our case) total number of each type of files ( docs, images, videos etc. ), total server space used and percentage of remaining space left. In terms of functions keep it simple, you do not need too many details, being able to search a user based on username or email and view all its data it is fair enough.
Do not fall into the trap of Facebook, Youtube, these are exceptional cases of exceptional entrepreneurs. Have them as your dream goal but be realistic on what you can achieve.
For at least a year, our great customer support was the main driver of the business growth. We were replying to most of inquiries within a day and this had a huge impact on the brand, trust, reliability and free viral marketing. One great lesson learned is that you need to reply to your customers requests instantly, I do not say answer and solve their issue because this may take few days or may not be achieved but you need to show that them you acknowledge the issue, this has an enormous effect to your brand.
Another key feature not to be forgotten by any means is a good mailing system so that you keep in touch with your users. This must be in place from day 1 because as your user base grows it will be harder to integrate. You can either enable API from major newsletter services like Mailchimp or build a system from scratch. It does not need to be too complex since it will be for internal usage only, send mail to all users; send mail to specific groups (paid, free) and some good tracking stats would be just great.
- Did not keep in touch with our users. By not having a good backend system we could not keep in touch with our users on a regular basis, so as a result our brand deteriorated and ultimately was hard to bring them back. We attempted to extract all users’ names and emails once a month from the system and use a mass mailing script to keep in touch with them. Having to do it for few thousands is ok but for close to 500,000 users is hard to keep up.
Initially we thought that is ok if we do not email users since we were hardly getting emails from other web services but the truth was the exact opposite. At the first newsletter campaign we increased monthly revenues by 15%, the reason? People who signed up a year ago had forgotten our brand since there was no communication. But to achieve that you need a great product above all.
- Had too many features. Initially things were going fine, we had only 4-5 core features and people loved it. We messed things when we wanted to add more features. When we re- launched the service in March 2008, the platform was so cumbersome that was hard for users to navigate with so many features and most importantly none really knew what we stand for. Stick to 4-5 features, make them simple and do not fall into the trap of too many features, versions.
- How fast your business can grow – I do not believe majority of startups know that (forget exceptions), you get that gut feeling with your next entrepreneurial attempts. One key issue that would make me believe now that my next venture would possibly skyrocket is the offer itself. If you say to yourself “this is an insane proposition “then the chances are high because to create huge impact you need to create a huge disruption in your market. I still could not believe that is possible from 5000 unique visitors a day till April 23rd, the site would get 50,000 in just few hours. How would I solve this now? I would move on the cloud, probably AWS or any other provider from first day so the business can scale with no server upgrade and downtimes. This was the most crucial mistake we did, all pieces were in place, great product and a great momentum from social media but our infrastructure was not ready. Five days off service are good enough to have this momentum disappear.
- Be prepared for the unexpected – In an April Sunday morning back in 2006 I was unable to login to our admin system, after trying for few times I decided to contact the dedicated server support (today is the second largest cloud hosting provider). The reason? The overnight shift employee had mistakenly erased the entire server data. I thought that is, our dream is over but later I realized that the day before we had backed up the whole server because we were planning to move to our own network. If this had happened the night before everything would be over. I would never believe this could have happened, so from now on do occasional backups depending on your data volume.
- How service would be a year from launch date – Things change so fast when you finally get this growth momentum that is hard to predict how your business would be in a year from now. The only compass you can rely on is your users, they are the ones that would tell you where the market moves and it is your decision whether you want to follow or now (if everyone says more free space more free features does not mean that you need to do that, use your brain first). You may have to change your prices, subscription plans or even revenue streams as you gain more experience and understand what works best for your business.
Although most of these mistakes can be eliminated ( not all of course ) by listening more carefully to your customers and seeing what other successful start ups do, there are some mistakes which were simply not possible to be avoided. Many times I tried to connect some pieces of the puzzle and see how I could had predicted that the service would hit Digg.com first page and buy more servers in advance but nothing made sense. I guess experience is hard to substitute.
Things eventually did not end up as we were planning and we decided to sell off in mid 2009. It was an amazing journey with a lot of mistakes that made us lose trust with our users but also with the satisfaction that built a great product people still love using. We finally lost a good chunk of money and this is due to our mistakes. Don’t repeat them.
For more startup news, follow us on Twitter @startupfoundry or on Facebook.
👉 Filed Under.
Categorised in: Strategy
If you like startups, join our weekly mailing list. Good startup content, no fluff.
Hi, I'm Paul Hontz.
I'm a YC alumn and I love startups. I created TSF to highlight companies I find interesting. You can learn more about me here.
The Story of Cruise (YC W14): How 4 people built a self driving car in 7 months.
Jul 8, 2014 2:35 pm
I Want To Write About Your Startup – Relaunching TSF
Jan 17, 2014 5:14 pm
How To Do B2B Email Sales
Nov 23, 2011 2:12 pm
Nasty Bug in iOS 5.0.1 OTA Update screws up Address Book on the iPhone 4S
Nov 10, 2011 11:56 pm
It’s Easier to Answer to Your Code Than Your Customers
Oct 28, 2011 2:25 pm