It's about time! ...Tech sites become more critical of startups and investors
First off, my hats off to any entrepreneur who is willing to put him/herself out there and take the risk to start a company, sell to customers, and raise money. I wish I could to it myself, but that’s another story.
A while ago, when I was a more avid reader of tech sites like BusinessInsider, Techcrunch, PandyDaily, and Betabeat, I noticed majority of stories about startups were rah-rah - everyone will conquer the world and every industry was going to be disrupted.
But over the last few weeks, I’ve noticed that writing has become more critical of both startups, founders, and investors. It’s about time!
It’s a great start to talking about the difficulty of building a sustainable business, how raising money is not success (already been beat to death), and money is too abundant.
There are many startups attempting to build content recommendation and curation algorithms. Their approches are varied and range from the simple to the very complex.
I would generally categorize the simple approach as using explicit profiles. Explicit profiles are built by asking directed questions (What are you interests?) and tracking usage (What articles have you read?)
The complex approach is to build implicit profiles. Implicit profiles are based on what you do, but also what you don’t do. This requires a lot more understanding of the content characteristics and mapping them back to the user profile. For example, you have be shown 10 different articles about a specific topic, but you only click on 1 of them. What was the reason you clicked on that specific article? What it the image? author? title? time of day? day of week? device (mobile/tablet/desktop)? What if you already read it from a different source?
Recommendation algorithms are designed to be safe. If you’re on Amazon looking at a book, you won’t be recommend a scooter. It will most likely recommend another book within the same genre and topic. This bodes well for a the explicit profile approach to work without the need a very strong implicit profile.
There are many other issues to overcome with content recommendations and curation, I don’t think implicit profiles will be the main hurdle.
In the beginning, there were a handful of sites where you would go to to seek information (CNN, ESPN, WeatherChannel etc.) As the amount of information grew, companies were developed to help categorize and sort information (Yahoo), then came search engines.
Today, we’ve had a massive shift in the way we consume content. We no longer have to pull or seek out content. The majority of content we consume now is pushed to us via email, Facebook, or Twitter.
As the number of friends and follows grow on the social platforms, the amount of information gushing through multiplies and we’ll need another tool to help sift through all that content.
Like many other startups, we use Amazon Web Services (AWS) to run Fitocracy’s website and iPhone app. Our main stack consists of half a dozen Elastic Cloud Compute (EC2) web instances, an EC2 Redis instance, and an Relational Database Service (RDS) instance that are all located in us-east-1…
One of the most defensible positions for a startup is if you can achieve the network effect. The network effect is so strong that it has kept large companies in business for a long time, despite bad products and numerous competitors. Craigslist is a perfect example. It is only recently that a…
The billion dollar tech startup valuations and acquisitions are starting to attract a lot of opportunist to the industry. A lot of these people are not technical and they know they need a CTO to get to get to MVP.
I don’t have any problem with non-technical founders teaming up with…
I tried out Twitter’s newly rolled out Discovery tab to see if it could do as good a job as Knowaboutit or News.me (two social media automated curation / content discovery newsletters that I had been recently using). This is what “filtered” to the top:
Follow the instructions in Chapter 1 carefully and you should be all set on the Ruby on Rails part.
2. Host your Git repository
You can use GitHub to host your repository, but private projects are not free.
I’m using Dropbox to host my Git repository. I like the Dropbox solution because I can access the repository from any machine I have dropbox installed on without having to worry about any other outside providers.
Without fail, I get at least one email a week asking about how to break into VC. I will give my short answer, and then point you to some other posts with great answers and advice, much better than mine.
There are two common paths to becoming a VC:
You worked in the startup space
You have transactional experience as a banker
Some other common traits:
Have a network of entrepreneurs and VCs/lawyers/bankers
Strong interest in working with startups.
Most of all… you have to be Lucky. There aren’t many openings, so you have to be there at the right time and place.
I received a copy of Reid Hoffman’s The Start-up of You at SXSW and read it on the flight back home.
Interestingly, my blog is called Perpetually in Beta, which is the what the book stresses - always be changing. The book challenges the reader to don’t become complacent and always be improving.
Overall it’s a good book and a worthwhile read. The world is ever changing, so we need to be ready to adopt and change with it.
Too many references to tech startup stories. Most of the general population won’t care about how Dropbox made it. I think they would more interested to hear about how a 30-something was able to change careers.
LinkedIn was touted as the tool to solve your problems. While LinkedIn is valuable, real skills learning is how you change, not networking.
Here are my tips to achieve “permanent beta”
Network locally - While it’s great to network with people you aspire to become, it is also useful to network with those around you. I learned a lot about retail buying from my neighbor. It helps me think about ecommerce investments. I chat with my college buddies about their careers. One does internal corporate communications, and the topic came up with a prospective investment. Everyone has information that will be helpful to you in one way or another. You just have to be open an receptive.
Learn - Focusing on your interests is dumb. The interests of most people don’t translate to their careers - fashion, entertainment, auto, etc. So learn something that will further your career and life.
Read a book about financial planning. Boring!!!! But that’s what you need to plan for your future.
Learn how to develop an excel model. Boring!!! But that’s how you’ll impress the boss when you understand the tradeoffs of an operational or financial decision.
With HootSuite surpassing 3m users and generating significant revenue, I thought that this would be a good time to reflect on the investment we made back in December 2009 and share what I’ve learned since.
For quick background, I reached out to Ryan, the CEO and founder, in spring 2009. It was…
CodeYear.com is a great idea. They will send you a weekly homework assignment that will teach you how to code. I also like the marketing push to coincide with New Year Resolutions. Sign up if you haven’t already. I did.
Great idea, but how many people will finish and gain enough skills to actually land a job or launch a successful app?
We all know majority of new years resolutions are never met. I know of only a few reasons most people can successfuly achieve such goals - your life depends on it and/or your career depends on it.
Codeyear is a start in the right direction, but there is a lot more needed than just a tutorial. It needs to create a community support group through forums, meetups, experienced newbies and mentorship.
Augment the codeyear experience by adding taking some additional steps:
1. Set a goal - launch an app or get a job. Keep it simple, a simple CRUD or better yet, a read-only application 2. State it publicly - accountability always helps 3. Get Support - a community to answer questions, guide etc. 4. Get a coding buddy/mentor/partner 5. Fix someone else’s bugs - Coding from scratch is hard. Modifying someone else’s code is a much better starting point.
I doubt many people will stick with the program. But I want people to make me eat my words. Prove me wrong!!
I’ve signed up for CodeYear and I’m going to seek out a programming mentor, thanks alexkehayias. I “plan” on sticking with it to dust off my development skills and learn some new skills and languages. But life happens …
Google 1998; 2004 Amazon 1994; 1997 Yahoo 1994; 1996 Oracle 1977; 1986 Apple 1976; 1980 Microsoft 1974; 1986
What does this mean to startups and investors?
The investment banks want the IPOs and the mega-mergers. They don’t care about sub-$50 million fund raises or acqui-hires. So I think they’ll do their best to help out the both the investors and the entrepreneurs with introductions for hiring, investment, acquisition all in hopes of building the relationship to be at the front of the line for the big deal.
Birchbox has gotten a lot of great press and buzz in the startup community and it’s generating a lot of copy cats. Birchbox works because of a specific set of circumstances, to replicate for a different industry and audience will be challenging. Scott Brit wrote an excellent post about his attempt in the space.
Does the consumer have too many subscription services? Is it just wasteful spending? -especially in the current economic situation.
5 years ago, I only had a handful of subscriptions. Now I have more than I need. Below is a short list of subs that I’ve had in the past or ones that I’ve considered.
I can’t seem to make time to craft my blog posts. So I’m going to experiment with speed blogging by just drafting in Plaintext in iOS, copying to Tumblr to add links, and then publishing - spending no more than 10 minutes editing.
In my career, I’ve build hundreds of financial models and reviewed thousands more for everything from VC funding to M&A to LBOs.
Lately, I’ve been receiving half backed financial models.
I’m not talking about seed companies where the focus is the product and getting the first customer. I’m talking about companies with a product and paying customers.
I’m not asking for a complete 3 statement financial model that bankers put together. I’m asking for a model with the key performance metrics highlighted and understanding of the levers that drive your business.
The most important part of the financial model is the revenue build.
I’d like to see your assumptions and understand how you got to the revenue numbers you presented. Don’t just plug 20% growth for the following period.
I want to see that you’ve thought through the number of customers, their spend, their time as a customer, and cost to acquire. This will lead to your KPIs metrics like LTV, churn, and break-even. Then compare those to industry numbers and be able to explain why yours are different.
It’s not about being accurate, it’s about thinking it through. Remember back in high school, you had to show your work on those algebra exams to get full credit? Same here.
Every industry and situation is unique, but a well thought out model is important to move an investment along quickly.
It was 1998 and the dot-com boom was in full effect. I was making websites as a 22 year old freelance programmer in NYC. I charged my first client $1,400. My second client paid $5,400. The next paid $24,000. I remember the exact amounts — they were the largest checks I’d seen up til that…
Beautiful Savings was the first business I’ve ever started. I wanted to take a minute and write about the experience of bootstrapping your own startup. Some things I did well, and many things I did poorly.
I’ve shared this tip several times and heard great responses. So now I’ll let everyone in on my secret interview question.
During the course of my last job search, I came up with a great question to ask the interviewer. I used it 18 times and was able to predict the outcome of 11 interviews. I got 3 wrong and I couldn’t get a read on the rest. That’s a success rate of over 50%!
The question is “What concerns do you have about me?” or some variation. It’s a simple, but effective question. I really like the question both as an interviewer and interviewee because it cuts out all the bullshit and formalities. It forces the interviewer to give real feedback and brings any shortcomings front and center. As the candidate, you get truthful feedback and most importantly, an opportunity to address it.
During your interview, you are trying to prove yourself. But you might not fully address all the requirements, both explicit and implicit. Asking the question turns an unstated requirement into an explicit one. Each time, I was able to address their concerns and I walked out with a good sense of the outcome.
Here are some of the answers I received.
Note: The interviewer will probably not offering up their criticisms directly (at least I have never heard them), but they may be the reasons you don’t get the job.
- not enough experience/not right experience - not the right fit/startup experience - over qualified/will leave for a better position/not a management position
If you can directly address their concerns, you’ll increase you chances of getting a call back. Otherwise, you will be letting the interviewer to deliberate critical points without your input.
If you use this tip, I would be very interested to hear how it went.
A few months ago I started a mentoring relationship with a person who wanted to break into investment banking. He made up his mind and jumped in head first; he quit his job and enrolled in an investment banking course. I worked with him on real world modeling exercises and general interview prep. We talked about how difficult it would be land interviews let alone a job. He graduated a number of years ago and didn’t have the relevant work experience.
He got several interviews with boutiques, but nothing panned out. But he he kept at it, and finally landed a M&A investment banking position with a middle market firm. While it was not perfect position, it is a great starting/jumping off point. (He really wanted to be in NYC)
He got the job by being out there and being lucky. While visiting some friends, he reached out to some local firms and continued networked. His resume was passed among 7 firms before landing his coveted position. It turned out the firm was starting a search to fill a slot in the incoming class because a new hire backed out. It just happened that his resume was near the top of the pile, so he was selected to interview.
While I think of it as “luck” that he got the call, he worked very hard to prepare and position himself to be successful in the interviews and ultimately as a banker.
Approximate 600 links flow through my lmai twitter feed daily (will try to chart later). I may see 20 of those because I happen to be in twitter at those times. I ignore most, I favorite some (which gets added to my ReadItLater list thru iftt), and a few I’ll read on the spot. On top of that, add Facebook, emails, and RSS. There is just too much stuff.
I’ve been playing with several apps and products that are attempting to solve this.
These are the apps that I have come accross, let me know if I’m missing any.
Zite - iPad only, well designed and works. They were recently served a cease-and-desist notice from publishers. It seems it was only relevant to the “reader” mode where all the ads are stripped out.
Flipboard - Beautiful visualization of your feeds, I suspect they’ll move into personalization with their acquisition of Ellerdale
News.me - Originally developed by NYTimes, now part of Betaworks. It was recently launched and currently only uses your twitter feed as a source. The biggest drawback to News.me is the subscription cost of $.99 a week. While it’s not expensive, why pay when there are so many free options.
Trove - Developed by Washington Post. It’s feels cumbersome.
Knowabout.it - Developed by @falicon and@willcole, some neat features like peeking into someone else’s collection of stories and items from quiet sources.
Summify - Developed by two hackers, @mirceapasoi and @cgst. They focus on showing what’s “important”. From my experience with the Summify emails, “important” seems to be defined by popularity.
My6sense - Mobile only. I haven’t tested the app yet, but looks very promising.
Read everything - VC blogs - Larry Cheng has complied a list which he updates annually. I added the top 20ish to my RSS reader to help make reading more efficient. This will help you get some insight to how VCs think and what they are interested in. - Twitter - follow VCs and entrepreneurs. Read what they link to. - Read BusinessInsider SAI and BetaBeat. SAI will usually link to the most popular postings of the day, so add that to your RSS list too.
Join the community - Network - attend events and shake hands with VCs and entrepreneurs (Meetups, pitch events) and then grow the relationship. - Twitter - start conversations by asking and answer questions - Comment on blog posts - Add value to the discussion. - Join every alpha/beta/new product and play with it. Take some pictures with Color, Check-in with Foursquare, get a Kohort username
Process the information - Take what you read/what you know/who you met and form an opinion. Will Company X be successful? How can Company Y monetize? What do you like/not like about Company Z? Know all the hot startups and some relatively unknown ones. Have an opinion.
Help a startup From the networking events you attended, hopefully you made some connections. Now reach out to the entrepreneurs and ask to help out. Propose a plan that will leverage your skill set. Got social media skills? … help with marketing. Got finance skills? … help with an Excel model. Got software skills? … well, everyone can use help here. How you will help will vary on your availability, your skillset, and their needs. It could be volunteer, hourly, project based. Either way, you’ll learn something.
I consulted with numerous startup building Excel models, writing software, doing QA, and just thinking through marketing plans and providing feedback. They all gave me some insight into their startup which was helpful in interviews.
Be lucky Luck plays a role in any job search, so be prepared when your time comes. During Q12011, there were at least 4 openings - OATV, Spark Captial, First Round and Time Warner. I haven’t never seen so many VC openings in such a short period, and those were the advertised ones! Each have their own specific requirements, but they all require the applicant to know the venture space and love it.
I recently concluded a search for an investment banking analyst/associate as my replacement (another post). I laid out the specifics, mainly work experience and education, and got flooded with qualified, under-qualified, and over-qualified applicants. There was no way to determine who to take to the next step.
I needed a screen for minimal competency, but I couldn’t find anything for excel and more specifically financial modeling.
In the tech world, there are many programming/technical assessments. See Facebook’s Puzzles, Palantir and TopCoder (recommended by Google). They all have different methods of testing competence. One can also review Github and StackExchange (fka StackOverflow) profiles. Besides testing skill, this also shows the applicant interest in the company and industry. No one wants hire an unenthusiastic robot.
Because I couldn’t find a suitable solution, I created my own. Download. I created financial statements for fictitious bicycle shop and asked the potential hires to forecast the business based on certain assumptions I provided. I chose a simple retail store because it is very tangible without a lot of moving parts.
Depending on the level of the hire, you can add complexity to the model and tasks.
5 years of growth at 10% per annum
Gross profit margin improves due to higher volume of sales
Pay back bank debt over next 4 years
Addition of second store financed by a combination of debt and equity
LBO for competitor
I then spot/sanity check the numbers and formulas. Do revenue/gross profit/EBITDA numbers/calculations make sense? How do the secondary models function? (e.g. PPE and debt) I also look for attention to details - formatting, colors, linkage etc.
Results: For a 3 day post on Doostang and some personal networks (Fri - Mon), I received over 200 resumes. About 25% were given the financial model; about 50% of those submitted their work. This led to a much more manageable set of candidates to review. The remaining pool was very strong - they had the necessary skills, showed interested in the job because of the extra time commitment vs. just sending in a resume, and cut thru the resume BS.
Lesson Learned: There are a lot of people interested in breaking into investment banking and private equity. This had led to the development of “vocational” financial training programs such as WallStreetPrep and TTS. As more applicants enroll in these programs, the baseline for financial modeling will be raised, which is great.
However, this will make it more difficult to screen via technical skills. A while back, I could assume that if you knew how to model, you understood accounting and finance. Now if you took a program, you would know how to model, but do you really understand accounting and finance? An understanding that is usually gained thru deal experience and the related building of the models and decks.
A lot of people start of by saying, “This is a dumb question, but …” and my reply is “there is no such thing as a dumb question”. Not true with Quora. There are dumb questions. There are also spam, marketing, redundant questions and answers.
Don’t get me wrong, there are some great conversations going on, but in time the “good” content will be overwhelmed by the “bad” content and the quality and utility of Quora will diminish.
Stackoverflow.com is a niche site for programming questions. For the most part, the only developers ask and answer software development questions. Granted, most software questions have a “correct” answer, so its easier to select the right solution, and close the question. There are also other niche sites based on the stackoverflow platform like http://photo.stackexchange.com/
None of these niche sites will never scale to Quora’s potential, but they have value to its loyal users and that value will determine its survival.
*Quora could become more valuable if there was a better method of ordering/filtering questions and answers. I haven’t thought it out yet.
It seems like every company wants to have an iPhone/Android app, from the big brands to the small retailer. Why?
From my experience, most people don’t use all that many apps on a regular basis.
This is what I use:
Almost daily: Mail, Safari, Twitter, Evernote, iTunes, Google Voice, NYTimes, Instapaper, WSJ,
At last weekly: Mint, Maps, Foursquare, Skype, RunKeeper, Yelp, rotation of games
My others apps are only for specific uses.
This all relates to my overall theme of habits. How much value add is your app that I will change my habits to include yours?
Personally, most apps don’t bring that much value. They won’t be get enough of my mindshare to load them up.
In the long run the aggregator and filter apps will will out. I use Maps to find directions. I use Yelp to find lunch. I use Twitter to find interesting stuff (in fact, Twitter and Facebook have gotten too messy to wade thru that I need an aggregator for my social network - http://knowabout.it)
Yes there are apps for niche use, and they are necessary. But not every company needs an app that replicates their website.
Who is ultimately responsible for privacy if that privacy is violated through the use of an API? The user of the API? The owner of the API?
I was recently using a photo website that connected to various social networks to display my photos. I found a security issue where if you change the id variable in the url, you can view another person’s pictures. I’ll address this in a separate post. This isn’t even a hack - it’s just changing one number in the url.
So who is responsible for protecting my data? Is it Facebook? I uploaded my photos to Facebook and Facebook is the source of the third-party photo app. Should Facebook be responsible for checking the apps that connect through it’s API? Is it the third-party photo app? They’re the one with the security flaw that is exposing my photos. Is it me, the user? Should I not be posting my photos knowing someone, somewhere, somehow can view them?
I’m leaning towards the third-party application, but it’s not going to look good for Facebook either.
Fash911 is finally up and running. The site has changed directions a few times and this is the latest incarnation - crowdsourced fashion and beauty news.
It’s a Digg/Reddit for fashion. Digg and Reddit are too general and random. Yes, there are categories and channels, but they’re not well populated. We feel a niche site can attract and serve a better audience.