Posts tagged startup
Posts tagged startup
How do you build a new product with constant customer feedback while simultaneously staying under the radar? - Eric Ries
If you are part of a startup, you have likely read Startup Lessons Learned by Eric Ries. The quote above is based on this post, and details his experience utilizing Google Adwords to attract initial users to IMVU. I wanted to slowly increase our user base at lloquy.com, allowing time to pivot implementation and limit mistakes to a wider audience. Therefore, his method to attract users really got me excited, and I wanted to write about my experiences utilizing this type of marketing.
My previous experience with Google Adwords was positive. Around 2003/2004, I used Google Adwords to gain users on a videogame-related blogging website I created. My daily budget was a mere $2/day, and the service consistently brought 30-40 users each day. That website wasn’t a business, though (merely something I did to learn new technology). I felt that if Eric Ries was having better success today, then this method could easily provide a high return on investment and warranted some of my attention.
Unfortunately, my experience in 2011 was less successful then Eric’s and my past experience in 2003 (by a wide margin). Google Adwords, like Google’s analytics’ service, has become extremely complicated and hard to use without a significant time commitment. For Adwords, I believe this is because common and popular keywords are being taken by companies with a much higher budget, leaving bootstrapping startups behind.
I tried many, many keywords, but at most times, my keywords ended up being in one of two categories. Either the keyword did not have enough data to adequately gauge the number of daily clicks (meaning a low percentage of users searched on it), or the keyword was really expensive (meaning my $5/day would result in 1-2 clicks only).
Since you only pay for clicks with Google Adwords, I decided that casting a wide net was still a good method, and utilizing a wide net of lesser-used keywords would still produce a sufficient number of clicks at a low cost. This hypothesis proved to be false. I did receive some visitors, but the majority of clicks were from Google’s Display Network, and not from the keywords I entered. With the majority of clicks coming from the Display Network, I was unable to figure out any information that could help me improve my campaign.
I’ll be the first to admit that if you’re willing to really study Adwords and put in the time commitment, then you may have better luck. For a startup, there are 1000x things to do at any point in time, and I reached my limit with what I felt I could accomplish with Google Adwords. I felt bummed that this technique wasn’t working for me and decided to ask Eric Ries whether he still believed in his blog post. His response is below.
Me: @ericries Do you still find your post on SEM… relevant? I had tremendous success years ago, but not so much in 2011.
Eric: @presstartgames yes very much so- but AdWords itself has gotten too expensive for most keywords. try reddit or stumbleupon
Great advice, and my test validated his response. Eric offered two alternatives, so I researched both and decided to try StumbleUpon next. StumbleUpon utilizes a system where you pay for users to “stumble” on your website. Users have the capability to “like” or “dislike” your website, and you can earn free stumbles if enough users like your page.
Unfortunately (again), StumbleUpon did not produce the results I was hoping to achieve. While I only put in a small amount of funds for this test, StumbleUpon did not provide one sign-up to lloquy.com.
The results so far? Two different services with two similar results. At this point, I concluded my issue had to be one of three things
I decided to ask Hacker News users (through this post) on issues with my launch page, and while I received some excellent advice on how to improve the launch page, I also received the following advice.
Stumbleupon traffic doesn’t work well. People are simply browsing from one site to the next looking for articles or funny cats and your bounce rate using SU is probably 100%.
Google ads might work but you might be better off getting on http://betali.st and http://startupli.st to get those early adopters.
Another request for advice and another excellent response. BetaList and StartupList are both free websites that attract users looking for early access to websites. I thought this could be a good approach and decided to submit lloquy.com to BetaList. After my submission, a few weeks had passed with no news on if it would be featured. I’d given up on that avenue and concluded the website would not help. I thought I was back to the first step.
Then BAM! it gets featured. The result was a steady stream of new users signing up on the launch page. In total, BetaList resulted in approximately 100 new users, which is exactly the goal, since lloquy.com has still not “launched” (in a marketing sense).
To summarize, the message of Eric Ries initial post is still valuable (and true), but the technology is constantly changing. For startups looking for the initial set of early users, I highly recommend BetaList over more established methods like Adwords. I haven’t tried StartupList yet, but plan to in the future.
Like this article? Have a comment? Leave it here.
I get it. Hipster got 10,000 people to sign up by offering “early access” to their product if they invited their friends, and LaunchRock started an entire project based on the concept. With these successes (as well as fork.ly), it seems many new projects are utilizing this concept, even when it doesn’t seem to make much sense.
I really respect the companies who have had success utilizing this marketing method. I also think LaunchRock is a great idea. So understand it’s not the concept that I’m against. There are many situations where using this type of sign-up method (for early access) is valid and reasonable. For example, if you are performing problem validation, then this type of splash page is particularly useful. Or, if a user’s experience is enhanced by having their friends on the site (such as a social network), then it makes sense to push this requirement.
Using “early access” as an incentive to obtaining a viral launch page is a nice idea. My frustration arises when this concept is used for ideas that do not benefit the people I must invite. For example, being informed that I have to invite friends to access a new blog, newsletter, or email is frustrating. If I want access to a new web application, such as an analytics service, I don’t want to pressure my friends to sign up so I can get access. In today’s world of information consumption, I quickly forget about the newsletter or service and move on.
While using early access on launch pages have provided success, I believe it’s time to offer other incentives to encourage launch pages to be shared with your friends and contacts. LaunchRock is seeing this as well, as their latest blog post shares how one company is utilizing the service for more then just early access. Startups should offer enhanced services to people who share, or something that is harder to achieve without sharing with your friends. lloquy.com enhances your influence score, which is one of the main indicators of reputation on the website. The ability to increase your reputation even before you have access provides a reason to share the site with their friends.
In summary, while early access may work for some websites, it does not work for others. Services like LaunchRock have made it incredibly easy to create a launch page, but it doesn’t replace the strategy of attracting users to your website.
When analyzing problems or questions, it’s important to get past the first answer and find the root cause. In my experience, the first answer usually only looks at the immediate problem or obvious solution, ignoring everything around it.
The URL provided takes a similar approach to risk. In many instances, we may not do something because we view it as risky. We may avoid showing people our website, or posting a blog on Hacker News, because we risk that people will tell us it’s not good. By looking at the overall picture though, we may learn that many actions pose very little objective risks, and our ability to get over subjective risks is key to being successful.
What’s incredible is that lots of things pose largely a subjective downside but have an objective upside. That means that if you can get past caring about those downsides, you can leverage yourself heavily and can repeatedly expose yourself to risks with big objective upsides with little effective downside.
Check out the full article here.
The first habit, or the foundation, of the 7 Habits of Highly Effective People, is being proactive. In short, you are responsible for your own life, and therefore possess the initiative to make things happen.
I’m not through the entire book yet (I’m actually only on Habit 2), but this first habit applies to startups as much as life. If you own a startup, you are responsible for that startup’s success, and you must make things happen.
The following is a scenario I’ve seen before. You create a business you know customers will love, because you love it. Whether you spent money to create it or did it yourself is irrelevant. You go-live, but the people don’t come. You decide to post a link to your Twitter and Facebook accounts. Nothing. You add features because you thought of cool new stuff. Still nada.
Here’s the point where many entrepreneurs make an incorrect critical decision. They either continue to add features, or they quit, assuming they’re product won’t be successful.
You need three things to create a successful startup: to start with good people, to make something customers actually want, and to spend as little money as possible. Most startups that fail do it because they fail at one of these. A startup that does all three will probably succeed. - Paul Graham
Paul Graham is really good at taking complex topics and breaking them down, such as the paragraph above. You only need three things to create a startup, but you can’t achieve any of them without being proactive. Startup life is difficult, and successful entrepreneurs can not let blame various circumstances for their results. They need to proactively find ways around those circumstances to achieve success.
Lean Startup principles are proactive by definition. You talk to customers to learn and validate assumptions until you achieve market fit. You’re continuously deploying changes, and testing small updates to understand their value. You remove waste that is not providing value to customers.
Being proactive sounds obvious, and I’m sure most feel they are already being proactive. However, looking at the definition of proactive from the book, I’m looking for ways to do better.
I can’t wait to read about the other 6 habits.
I launched TweetAfter.me with a minimal feature set. I had (and still have) many ideas to enhance the service, but made a conscious decision that none of them were necessary on the very first day. This allowed the service to get in front of users faster, allowing feedback to be received quicker, which ended up changing my view of which features were more important in my backlog.
Besides deploying with a minimum feature set, I also felt it was important to reduce recurring costs until higher revenue is generated. Let’s face the facts. A new product or service has zero users when it launches (TweetAfter.me went live on 02 January 2011), and no product or service is guaranteed to increase users. Therefore, it makes sense to lower your recurring costs to reduce your overall burn rate.
Sometimes that includes making tough choices regarding features. While sometimes, you completely remove a feature, tailoring the feature may be the only solution. A real example from TweetAfter.me is that tweets are currently scheduled in one hour increments. The goal has always been to provide a lot more flexibility in duration between tweets, default schedule durations, etc. However, right before deployment, I made the tough decision to reduce the initial scope of these features. Why?
To achieve my initial goal would have had a recurring cost 12x more expensive then current recurring costs. With zero users on day one, it made little sense to pay 12x more money for features of a just-launched service. The truth is that startups should test the market to verify those features are needed. If they’re needed, then users may pay for them, increasing your return on investment for these features.
Therefore, just don’t assume you need expensive features on day 1. Higher costs give you less time to validate decisions with customers. On day 1, one of your highest priority goals should be to give your startup enough time and money to reach customers. If you spend the money now, you lower the chances of succeeding regarding that goal.
If I had to describe the Lean Startup Methodology in less then 5 words, I would say “Ideas, Product, Learn, Repeat.” First, people have ideas and deploy a minimal product. You then learn from customers who use your product and repeat the same process, modifying your ideas and product(s) based on customer feedback.
Learning is the toughest step in this cycle, as it is the only state that requires external people to complete. Very early customer development and validation is also a state that doesn’t get much attention. Although that is starting to change (due to the Lean Startup movement), most material I have read focus on struggles with idea and product, then shift to funding and scaling. It’s a shame because the learning phase can often be the most frustrating, and without persistence, may create a lot of waste, or even critical mistakes. Take these two examples.
Just Keep Adding Features: Many entrepreneurs skip the learning phase completely if they become frustrated with the lack of initial customers, electing to implement new features from their backlog instead. This can be counterproductive, as one of the biggest forms of waste is creating features users do not want. A decade ago, I made this mistake with a blogging platform I created (before blogging was actually a term). I became impatient, and by the time users were on the platform, the usability of the platform had degraded due to the feature additions I made combined with the lack of validating the usability of those features when integrated.
Pivoting too Soon: The 2nd potential mistake is entrepreneurs giving up and moving on to the next idea. They repeat the “idea, product, learn” loop, as they believe they “learned” their product or idea was flawed and the best course is to start over with something else (Note that sometimes, this is true). Entrepreneurs are being told to “Fail Fast”, which I think is great advice, but if entrepreneurs can’t learn why the product failed, then they may need to keep going.
So, how do you enhance the learning stage when you’re trying to entice “earlyvangelists” in the crucial beginning weeks or months? It’s not easy. Here are a couple ideas to start the conversation.
Search Engine Marketing: SEM services, such as Google Adwords, could be a cost-effective way to get initial users to your website. Eric Ries has a great article on this approach, and I encourage you to read it, as there’s not much I can add. Note that Adwords has become significantly more complex then even a few years ago. Therefore, I advise that you spend a great deal of time optimizing your campaign. If you don’t, you may draw the conclusion that something is wrong with your service, when in reality, your ad campaign sucks.
Use Social Networks: It is so much easier to increase your reputation with the plethora of social websites. 10 years ago, there was very little. Blogs allowed non-technical entrepreneurs to increase their reputation and become their own brand. Now, websites like Twitter, Facebook, Hacker News, Quora, and Stack Overflow allow you to share your knowledge and insights with a larger audience. If people trust your insight and thoughts, users are more likely to try your services and products.
These ideas are more indirect methods to better achieve your goal of reaching initial customers. I purposefully stayed away from other, more talked-about ideas, such as A/B testing or customer interviews. No matter what you do, understand most methods take a significant amount of time to “get right”, and many times, it’s easier to go back and add a new feature. Don’t fall into that trap!
I’m interested to hear your methods regarding this topic, as everyone benefits from new ideas, and no idea will work for every situation.
Like this post? I encourage you to visit TweetAfter.me and provide feedback.
