Apparently, this year it’s particularly in vogue to express a certain ennui about the batch of Y Combinator startups that presented at Demo Day.
As I checked in with investors and friends who attended the event, the overwhelming response was that there were tons of solid companies but nothing “amazing” or “10x”.
Even the tech press can’t seem to get behind one single company, instead making their obligatory “top 7” and “6 favorite” lists. This morning’s story in the Wall Street Journal confirms the overall sentiment:
founders are tackling “boring” problems that generate cash
Founders risk everything, move across the country or further, endure endless stress, damage relationships, sacrifice their health and pour their every waking moment into creating something people want — and they’re rewarded with a yawn.
If companies are too serious they’re boring, and if they’re too social then they’re fluffy and we’re headed for the next bubble. Fuck that noise.
This TechCrunch article contains a graphic comparing eCPMS of desktop vs. mobile advertising:
TLDR version of this post: buying stuff on mobile is a terrible experience and fewer people do it than on the full-size desktop browser. Things you can buy easily (games, apps, content) is all cheap and single serve (< $10 per purchase).
Mobile & Desktop Impressions Are Not Created Equal
Advertising to a desktop browser is a relatively standardized experience, and you can assume the user has her full range of actions available at any given time (to see your ad as it was intended, to click, to buy, etc).
On a mobile device these assumptions fly out the window. There are multiple form factors of mobile devices, so you have to design collateral for all of them, test all of them, and you still can't be entirely sure your ad will look good or as you intended. So the view is messed up.
Then it gets even worse, with tons of accidental taps you're getting charged for because the advertising network places the ad in a place that is most likely to get tapped, even when the user has no interest in your ad.
The next problem is the marketer's own fault - the result of clicking through on a mobile ad is usually a terrible experience. Websites that aren't properly optimized for mobile, confusing landing pages, abrupt launches of the App Store with no context, mobile shopping carts that make me want to scream bloody murder and not trust for one moment with my credit card info.
And few do this well, so there isn't a gold standard to measure your own company's crappy experience against. Let's face it, most CMOs are not going to be the visionary to solve this usability problem - they're busy managing several other channels and a big investment here is very risky.
Why eCPMs on Desktop Advertising Remain Higher
At its simplest, the outcome of these differences is that eCPMs on the desktop are worth more because they a) are a more palatable user experience b) convert better and c) have greater competition among marketers trying to buy ad placements.
Beyond this calculation is the longer-term thinking marketing leaders generally employ when they consider various campaigns: the risk/reward ratio of any given channel.
If I am looking to build a new campaign, I can assume it will take longer to create the collateral and setup the keywords and other tracking needed for a mobile campaign than a desktop one - PLUS I probably have a lot of people on my team who are experienced and successful with desktop campaigns but would have a bit of a learning curve with mobile.
And this all assumes I have a mobile version of my website that is even suitable as a landing page for these ads. If not, there will be the additional work of creating new pages or optimizing the website. And don't even get me started on the shopping cart…
The Compounding Problem: Turtles All the Way Down
Stepping back from all this operational stuff for a second, we have to remember why marketers even buy ads in the first place. There are 2 main ways websites monetize - they either sell you products or they sell your pageviews to their own advertisers.
In the first case, they need to have some kind of shopping cart to capture the expected value of the impression they've just purchased.
In the second, they need to make their own mobile advertising experience decent enough that their own advertisers will pay MORE per impression on the marketer's own site than the marketer had to spend to acquire the traffic. And then their advertisers have the exact same problem.
I think value is only about one layer deep with mobile - you can't reasonably expect to resell your own filtered traffic to your own network when eCPMs at the first layer are already so low.
But If It's Really Hard and Few Are Doing It, Isn't That An Opportunity
YES!
If you can figure out how to provide a decent buying experience on your mobile site and/or a decent mobile advertising experience for your visitors (which your own advertisers are wiling to buy) then you are unique right now.
This knowledge is at the root of the gold rush to figure out how to make mobile advertising, and even entirely mobile-first businesses, work.
The company that can make the mobile traffic arbitrage opportunity more than one layer deep may well be the next Google. Until then, most marketers will continue to view this channel with trepidation and run small experiments rather than switch over their Adwords budget.
As I prepare my pitch for YC Demo Day I've found myself experiencing something that I haven't felt in a couple years - a lack of clarity around telling the story. I spent the first couple of days frustrated with myself.
“Why is this so hard?”
“I thought I was good at presenting my company.”
But actually, I'm not good at presenting my company yet. I'm good at presenting Twilio. Getting to the point where that was a natural “look Mom no hands” process took a long time, and I need to keep that in mind.
And stop piling on even more pressure to make it perfect.
What Am I Even Talking About?
I've been going in circles about my deck and the delivery. Focus on consumer or business? Tell the whole marketplace master plan, or just explain what we've built this summer? Interject a lot of my personal story as a marketer or keep it more dry? Talk about about how much our competitors suck, or focus on what we do that no one else does? Use one “up and to the right” graph, or two, or three… or?
Everything that was coming out of my mouth feel so much like a pitch, and so little like me. Buzzwords, bullshit, boring stuff. Bleagh.
I was starting to wonder if I wasn't going to be able to get my shit together into a script I could deliver without killing my soul, so I sat down with my cofounders to talk it out.
10,000 Hours to Mastery
When we need to have a serious talk one of us yells, “Shed time!” which is the bat signal that we need to grab some beers and head into our converted storage shed in the backyard.
We settled in and I gave them the pitch.
“I know it says what we do, but it has no soul,” I lamented.
Al has a minor in theatre, and he started telling me about creating great improv. Al and Kevin had been watching the movie “Heat” earlier in the day, and Kevin remarked that some of the best Al Pacino lines seemed so natural they must have been improv on the spot.
“No way, those were great because he practiced the fuck out of them,” replied Al. “That's why he makes millions of dollars and the rest of them are just stock actors.”
Whoa.
I want to be like Al Pacino (on so many levels!) so this advice definitely got my attention. I'm going to have to practice a LOT more. Like, every waking hour between RIGHT NOW and Demo Day. There's no way I'm going to get to my 10,000 hours but I can make some progress.
Making the Words Your Own
Al didn't stop there.
“It's like you've been popular on Chicago Second City Improv, and now you've been called up to do SNL. You need to take the writer's lines and make them your own, like you're saying them for the first time.”
Aha, this!
This is the problem, I am hating the script because it feels like a script. Its just not my style. When I pitch, my preferred style is to sit down and have a conversation with all this rapid back and forth and iteration of a natural human interaction. I'm not saying I've perfected the pitch, but I rarely see a blank stare after the first two sentences. And its not the same every time, I'm constantly listening and adjusting depending on many factors.
But you don't get to do that on stage. Its not a two-way interaction and there is no opportunity to iterate.
The Pitch is Advertising
Mad Men's Don Draper is a character who has spent 15-20 years of his career pitching to clients. From 1:25 to 2:55 - less than the length of time we will have for Demo Day - he completely re-imagines the Kodak slide projector. Its not an investor pitch by any means, but its still inspiring.
The Goal
Up on stage it is all about lead generation.
Despite what some may think, the true art of marketing is saying the least possible words to convey the most possible information. Its not just about the limited space in an ad, a page or a screen. Its about respecting the limitations of the space in someone's brain that they're willing to give up to you. Their attention.
So despite being “a talker” it is not a completely natural thing for me to communicate well.
If I look at how long it took for my presentation of Twilio to get down to something concise that anyone from layman to hardcore hacker could resonate with took YEARS. Years to discover the word “telephony” is the worst jargon imaginable. Years to realize that you can skip saying “API” and just say “developer tool”. Years to cut a word here, a word there.
And then one day, its just natural and you're not thinking about it anymore and it flows and its beautiful and people smile and you smile and the cash register goes DING!
Until then, it is work.
So I'm working on finding my new voice, and I'm grateful to my cofounders and blogging for helping me work out some of the stress. Back to practicing!
TL;DR1: I believe you can raise a $200M hedge fun, why not a seed round?
TL;DR2: You have the rest of your life to be a “fat and happy” finance guy on the fringes of the financial industry (or run any lifestyle business, like our family consultancy) – but GO FOR IT while you're young and healthy, with tons of energy and little fear! You can come swim in our pool later.
A couple years ago I called my Dad and asked him what it would take to start a hedge fund. I was 25 and working on my second startup job after departing corporate America, and a college dropout. I was totally serious.
I wanted to create a company that would time the market around commodities and futures by evaluating the import/export manifests of ships coming into and departing U.S. ports. The information is public record, and the only thing in your way is parsing the EDI transmissions.
I considered applying to YC with this idea, with a non hedge fund twist. I won the first pitch competition I ever participated in with a similar idea many years ago, so I figured I might have a shot.
My wonderful friend David Weekly probably remembers me angsting in his kitchen as I debated whether or pursue this idea or found the company that ultimately became Referly.
I worked in supply chain for about 3 years (age 19 - 22) in a Fortune 500 company, and had a sense of the business opportunity in predicting things like how many iPhones would get delivered at Christmas, or how many Harry Potter books would ship via Amazon.
My Dad is a registered financial advisor, so I grew up working on the family business - which advises high net worth individuals (people with $10m+ of wealth) on how to invest and protect their legacy (hint: not startups) and helps Fortune 1000 companies set up 401k programs that benefit their employees. So of course I called him, because legally we could start a hedge fund or fund-of-funds, if we wanted to.
Dad's Advice
I remember sitting on the couch of my very beige and boring SOMA loft, on a Saturday afternoon, glass of wine in hand. “Hi Dad, I am considering starting a hedge fund.”
“Hi Elle, what's the angle?”
I love that he doesn't question it, coming from his 25 year old daughter.
“Arbitrage on parsing shipping data,” I reply.
“Well, everyone has an angle.”
Hmmm.
Everyone Has An Angle
My Dad explained that hedge funds are truly a dime-a-dozen in the finance world, not dissimilar from angel investors. They generate more fees for investment bankers because they have larger capital requirements to reach scale, but ultimately its the same game. He pointed out angels have an investment thesis, and hedge funds have an angle.
Then he said the thing that was THE THING:
You can do a hedge fund anytime, why do it now?
Well thanks Dad, for the vote of confidence (more like “holy fuck you think I could raise hundreds of millions of dollars in this economy!?”), but what do I do with that?
He continued to explain that creating a hedge fund, or any kind of fund, was something i should be thinking about in my 40s or 50s when I wanted a more traditional role in finance with a lower risk/reward profile.
That's funny, because the fees from that kind of work would make me very wealth much faster than this startup stuff. I pondered his comments though - I'd watched my parents go from bankrupt to rich in the past 10 years because of their business, so it wouldn't shock me to discover they know many things I don't about creating wealth.
What Money is For
My family is upper middle class, but as service providers to the very wealthy I've learned some valuable things about money, at a much younger age than I otherwise would. It has changed how I approach people with money, for better or for worse. I expect this will have an interesting impact on my ability to raise funding for my company (and I'm not saying all positive here):
1. Trust is Everything or “Stewardship”
I'm 14 years old, sitting in my Dad's office organizing his spreadsheets. He says, “Elle, do you know what business we're in?”
“We help people keep their money right Dad?”
“Yes but what do we sell?”
I look at him with a blank stare.
“We sell trust, honey,” he smiles. “We're honest people just like the steel workers, union laborers, and employees these companies serve. We help them keep their promises.”
2. Investment Has to Mean Something
When you have an annuity that yields millions each year its not enough to get a return, because you already have the money. In fact, people like that expend and impressive amount of energy trying to figure out how to give it away.
When you get right down to it, very rich people tend to be a bit uncomfortable having so much money at their disposal and they're looking to advisors to help them deploy in a meaningful way. Passionate about orphan children, female entrepreneurs, bringing music, art, and culture to suburban America? There is a cause for that.
3. The Body Wears Out
A series of deaths in my family in the past couple years have made me more aware of mortality than I ever was before. My Dad pointed out, as we discussed the hedge fund business, that it was something I could do without actually leaving a chair. Move to New York, get a Blackberry, iPhone and 3-line desk phone and I could operate from anywhere.
I'd be nothing like the tired steel hammering laborer who I'd met this morning to discuss retirement benefits, who would retire early due to work-related injuries. I'd never push my body to endure the repeated strain of that work - and at 50 or 60 I'd likely be comfortably happy to run a hedge fund.
But I would never be able to recapture, at 50, the energy of being 25. My Dad had me when he was 28. To him, I am so early in my life (I'm 27 now… this post reflects on a convo 2 years ago).
Okay Dad, these are great reasons. But here is the reeal mindfuck.
Dad thinks I can raise $200 million dollars to start my fund, no problem.
Whoa.
This was the first moment I realized the difference between starting a startup and being a CEO, instead of being an employee, was primarily a question of having the balls to go for it. Chutzpah, gumption, whatever. They're all euphemisms for having balls.
Its about showing up, describing the value you can create, and asking for money to make it happen. My Dad, my hero in business, thought I could totally start that business but was questioning why I should do something so low risk and “old” or “slow” instead.
For me, doing something in finance would have been a logical next step after working in the financial industry, getting deep domain knowledge in shipping and global supply chain logistics, and coming up with an angle. It wouldn't have been easy, and whether I could actually raise that fund is to be seen (yes I still want to do it someday!) but the point is that it would have been a predictable course for me.
My Dad knows I'll be damned if I do things the predictable way.
Staring down Demo Day, I hold this conversation in my mind.
Over the past week or so YC companies in my batch have begun launching in anticipation of Demo Day. Congratulations, your startup is in the news! Now what?
I've done 3 company launches and more than a dozen product launches, and contributed to many more. Through those experiences one of the most important things I've discovered is that what you do in the hours after launch is crucially important to maximizing the impact and total reach of your news coverage.
Not only did you spend an incredible amount of effort building the product you just launched, but preparing the news coverage from pitch to publish probably cost the equivalent of $10,000 to $15,000 in time (or money if you used a PR agency). Don't go to bed. Don't leave your desk. You're on a mission - for this one day everyone on your team is either part of marketing or part of customer support. As far as “talk to customers” goes this is the ULTIMATE DAY for that.
Your Mission on Launch Day
Keep the website up, which really shouldn't be hard anymore for 99% of companies, and get every possible eyeball you can back on your website.
What You Should Be Doing (TLDR version)
Monitor traffic with Google Analytics realtime
Get it on Hacker News and get the comments flowing
Make your own company blog post
Send out an email to any existing users
Send out an email to friends, family, and existing/potential investors
Customer support emails
engage with people on Twitter
share on your personal Facebook
Update/make a Facebook page
Comment Section on the news coverage
Not having a professional marketer is no excuse - this is a simple checklist of items and anyone can do them by simply taking it step by step. No professional marketing title or skills are required, just some writing chops and a passionate love for your customer.
Not sure how to start, what to do, or how much time they should take? This detailed guide will give you some answers to get you started.
Launch Timeline from T-0 minutes til Coverage
Monitor traffic (Immediately)
I highly recommend Google Analytics realtime because you should already have Google Analytics running on your website and its easy for everyone on your team to use. Project it on the wall or have it take over a monitor if you can. Keep an eye on traffic sources to see new places where people might be discussing the news and driving traffic back to you.
Get it on Hacker News (Launch +10 Minutes)
There are many ways to do this and I could dedicate an entire post to how to leverage Hacker News - but the most important thing is that you should not have your entire team vote from the same IP address or you will trigger the voting ring detector. The lowest effort highest reward tactic for this first news story is to ask for votes via Facebook chat and Instant Messenger, and get the comments flowing.
Share on your personal Facebook (Launch +30 Minutes)
Get the entire team to share on their personal FB pages, this improves the edgerank of the link when it gets shared on your Facebook page.
Make your own company blog post (Launch +45 Minutes)
Don't have a company blog? Now you do. Get one on Wordpress, Tumblr, Posterous, etc. and get a post out. You'll need to switch over to something more permanent on your own domain but this is fine for now. Don't forget to hook up Google Analytics to your blog. Your post should be unique from the news, not a sales pitch, and offer unique insight into your company and vision. Give it an eye-catching title and post it to Hacker News as soon as traffic starts to die off for the original news story.
Bonus points: share pictures of the team, screenshots of early iterations that have never been seen before, and mention if you are hiring.
Engage with people on Twitter (Launch +1 Hour)
Make sure to retweet the official tweet from the publications who cover you, in the order in which their stories go out if there are multiple. After that check out who retweets it and try to engage them. A simple “Thanks for helping spread the word about our launch!” tweet is fine, although you should try to make each tweet unique. Also engage in any more substantive conversations about your launch/product/company taking place.
If you see people with very large follower counts tweeting about you, make sure to pay special attention to them - they could be your next whale customer and can drive hundreds of additional visits per tweet.
Send out an email to any existing users (Launch +2 Hours)
Whether you have dozens, hundreds, or thousands of existing users make sure they get to share in the fun and thank them for helping you get to where you are. This is also a great time to focus the email on a single call to action: “come back and kick the tires, we want your feedback”.
Haven't sent any email marketing out? You can use MailChimp or Campaign Monitor to make beautiful email templates in minutes, without any HTML/CSS.
Update/make a Facebook Page (Launch +2.5 Hours)
Its time. List your company as a business, upload your logo and invite all your friends to like the page. Add a Facebook like button to your website if you can. Post links to all the news cover, screenshots, and anything else you think they'll find interesting.
Send out an email to friends, family, and existing/potential investors (Launch +3 Hours)
Keep a list of people who are supporting you and your cofounders and helping your startup. This list should include your parents, family members who still don't believe this is a real job, other press who didn't cover the story but who are friendly to you in general, existing and potential investors, advisors, high value customers and anyone whose influence could help you build this into a much bigger business. These are your VIP insiders - make them feel special by providing your own color commentary on the launch and letting them know how their support has played a role in getting the company this far.
Customer support emails (Launch +4 Hours)
Hopefully you've already got support@ or help@ your domain set up. If not, get it and hook it up to Zendesk or some other dead simple email support tool you can use to manage the influx of support requests. You'll probably need to divvy up the requests across the entire team and respond to them fast and furious if you want to keep up.
This is some of the best customer development there is. After launch is over be sure to go back and read through all the tickets.
Monitor Comment Section on the news coverage (Whenever, Maybe Never)
No one reads comments except for people who love you and want to support you, and haters. Like and thank people for positive comments, and ignore the trolls.
After Its All Over
Usually your news will break between 6am and 9am Pacific Time, and resonate with readers throughout the day. You will see a lot of traffic during lunch hour on the East Coast and again on the West Coast. It will begin to taper off around 6pm Pacific Time.
Doing Support in Shifts
The best case scenario is that your launch is so huge that support tickets are coming in so fast and furious that you aren't sure when it will end. You may not have slept for days, and you can feel your tact and grammar slipping and its been 10-12 hours since launch. Assess your team and the volume, and depending on size and exhaustion level you have to FORCE someone to go to sleep (in the same building, do NOT let them go home you will not get them back).
You will have to force them because everyone will be running on adrenaline. You will really really regret it if you don't do this. Sleep in 3-4 hour shifts throughout the day if you have to, but get sleep or you will hit a wall as a team and if something breaks you might be mentally incapable of fixing it.
Recovery Day
Clean up bugs, the office, and your face. Encourage everyone to sleep and not get burned out. Gather the team together and thank them for a job well done - mean it and make sure to sell the vision to your teammates at this crucial moment.
Team dinner and post mortem
Spend time together, raise some glasses and eat some seriously nourishing food. Talk about the fun, the highs, the lows, and get ready to do it all again.
Did I miss anything you think absolutely must be on a startup's post launch checklist? Tweet it to me @DanielleMorrill
It might not have been obvious before, but this morning's announcement of the Referly API and corresponding launch of our “customer acquisition as a service” shopping marketplace should leave no doubt. Referly was built for distribution hackers. To be more precise, it was built to help people become distribution hackers.
The mission: you put in money, we output customers. 'Nuff said.
The reality is that its a little more complicated than that, but the promise of CAaaS (ughhh!) is huge, and definitely worth building a startup on. Google Adwords and AdSense revolutionized how SMBs bought ads 10 years ago, changing the paradigm from CPM to CPC. Can we turn it to CPA?
Getting the Most Out of This Post:print out every single page of your website and tape it all up on a huge wall in your office or lay it out on the floor. Bring together key stakeholders from every department in your company and give them post-it notes. Ask them to place a post it on every page that drives a business outcome they feel/are responsible for.
In most startups, he who writes code makes the rules.
Historically, engineers have owned the company's website and everything on it. Other teams like marketing, sales and customer support could produce content, but software developers were the gateway through which all content eventually had to pass.
More often than not, animosity would grow between engineers tweaking marketing content that had hundreds of hours of thought put into it, and marketing constantly creating more work for engineering with a rapid succession of iterative changes.
Both acts, while taken poorly by the opposing team, were well meant. Engineers who were hands-on building the product, had contributions to make and saw this as the only place in the workflow where they could interject their suggestions. Marketers (and other content producers) who were closely in touch with customers needs and wants tried to be responsive to the feedback they received.
Design Leads the Way
In the past few years I've seen a wonderful trend of collapsing together design and front-end web development into a single role. This allows designers to take back the reins, no longer accepting “I just can't do that in HTML” or an endless litany of unqualified feedback as excuse for poor implementation of beautiful Photoshop mocks.
By chopping their own assets and checking in their own HTML/CSS design has reclaimed the implementation of their work, allowing it to make it to production intact.
Designers at many companies have become a go-between: technical enough to be respected by engineering and crucial enough to the marketing team to be included in every meeting and understand the why behind decisions on both sides.
New Paradigms Breed New Processes
While the original goal was to get design work out the door without interference, another fascinating change has taken place in the design and engineering relationship.
Engineers have developed a vocabulary that enables them to be part of the design conversation in a way that is welcome, and many engineers are actually crossing over into design as designers cross over to coding.
With processes like Scrum and Kanban requiring more cross-team coordination of projects, engineering is getting a chance to weigh in on design much earlier when their feedback can be considered and not tossed aside as an 11th hour insight.
Own Outcomes, Not Ideas
With the rise of distribution hacking I expect their will be similar trend with content producers in the marketing, sales, and customer service organizations campaigning to reclaim ownership of the content (via Wordpress and other CMS tools), conversion funnel (via Salesforce, Adwords, etc.), and many other aspects of the company's website. There are already massive tools like Eloqua, Marketo and (my favorite) Pardot which strive to produce a suite of tools.
Distribution hackers, who often are a powerful mix of coder and marketer, will hold the keys to the castle because they can both check in code and sit in the weekly sales meeting. You'll hear me say this over and over again: with great freedom comes great responsibility to deliver results.
It's easy to forget that success isn't about control, it's about results.
Control is about making sure the design you submitted is picked above all the others, instead of picking the design that has the best impact on the conversion funnel.
Control is about making sure the product management tracking tool you prefer is used, instead of picking the tool that the most people on the team will engage with every day.
Control is about getting caught up on the process at the expense of the goal and there are thousands of ways to do it.
Its so easy to do by accident, so you have to be vigilant or you might find yourself wasting time spending time fighting over what tools to use instead of what results to measure and what goals are worth achieving.
My greatest fear in writing about tactics of distribution hacking is that this new movement will breed a culture of tools elitists who say things like, “I refuse to join any company that isn't on Wordpress”… or Salesforce, or Trello, or Ruby on Rails for that matter.
Some of the best distribution hacking is the result of constraints, and some of the companies with distribution most worth hacking are old, staid, struggling to be modern. The distribution hacker can embrace this and make it work anyway. They are the kings of workarounds, and then when the results roll in they speak for themselves and can provide the leverage needed to make a big change (like having a content management system for example).
So Who Owns the Website?
No one.
Each functional group owns outcomes, and the website is a tool for achieving those goals. It is an important tool, but not the only one.
Any part of the company that thinks it can use the website as a tool to deliver results against the goals of the business should have access to it.
The role of the distribution hacker in this context is a conduit for communication across teams, coordinating and prioritizing goals to avoid a tragedy of the commons on the website. Not all business goals are created equal, and the distribution hacker is often able to see a broader view of the company and its needs than a lot of other roles. They are likely to have relationship and respect for engineering and sales, for marketing and HR, for business development and support.
Often these cross functional coordinator types have gone by the title of Product Manager in the past, and if you are a bigger company trying to attract the right kind of hire you might find that Product Manager or Product Marketing Manager are HR-approved titles that will get you the right kind of people.
Get Out of the Way
There are probably several people in your company right now who feel they deliver results despite your website. Find out who they are, what they want, and try to make it right. Distribution hacking isn't always looking out at the world for answers from customers, data, and analyst reports or stats. Sometimes its about looking inward, assessing the speed bumps facing other people on the team, and getting out of the way. What's the worst that could happen?
More than any particular skill, like how to run an Adwords campaign or write a blog post, software developers struggle to accept distribution hacking as part of their professional repertoire because they find a simple fact, and its corollary, intellectually abhorrent:
Markets are not efficient or “fair” because people participating in the market are not rational actors.
Accepting and internalizing these two facts is particularly challenging for developers because they are used to performing their work with a machine that is supremely efficient and fair, while surrounded by people who are some of the most rational in the world.
Distribution hackers, on the other hand, disregard the reality of this statement at their own peril.
Distribution hacking is about understanding and accepting the conditions of reality, and then creating systematic unfair advantages that leverage asymmetry in markets for the company's benefit.
Creating Unfair Advantage
Startups need to use unconventional and “hacky” tactics for achieving distribution of their products because the conventional ones are crowded channels being executed on by companies with 100s of times more money, time, and manpower. SEO, SEM, display advertising, PR, even social media are channels any marketing person worth their salt can execute on.
If you hire a marketer and tell them to execute “by the book” you will get average results. For startups, average is usually failure.
Forget about the list of tactics you put in your job description – the distribution hacker is not a marketing monkey and they crave just as much freedom to act an experiment as any software engineer. You're going to have to give them some room to operate if you want to get the best results.
With great freedom comes great responsibility to deliver results.
The Birth of a Distribution Hacker
As the titles “growth hacker” and “distribution hacker” grow in popularity there will be people who recast their “social media” or “advertising” careers in this light. While I think these folks are well positioned to become distribution hackers they are generally focused on executing a particular channel, and often I find they are not nearly rigorous enough with their experimentation and metrics.
There is a very real risk that startups hungry for customer acquisition will bring a self-styled “growth hacker” on board and then find themselves disappointed after a few months without results.
It is so important to talk to your potential hire about how they imagine creating an unfair advantage for your company. They might not have a detailed plan, but you should look for strong signs that they can think outside the box. If someone claims to be a growth or distribution hacker as them to give you specific examples of how they've successfully hacked the system in the past, and demand numbers.
For a distribution hacker, speaking in terms of hypotheses and metrics is second nature because they live in their spreadsheets and mySQL queries day in and day out.
If you're hiring someone who is just getting started with distribution hacking you can usually figure out how creative and analytical they are by leading a brainstorm around tactics and metrics. I think there is a much broader market for bringing someone on board to learn distribution hacking than there is to hire someone who is experienced. Please send them to this blog.
Distribution Hacking as a Mindset
One of my former employees at Twilio, John Sheehan, once quipped “they can copy our hands but not our hearts” upon observing the rate at which Twilio's developer marketing tactics were being copied by other startups and telecom companies with developer programs.
Don't get me wrong, I'm all for copying tactics that you see working for someone else, especially when they address the same audience you are trying to reach. When I went to meet with these companies about marketing strategy I found that the imitation of our tactics at Twilio were surface level.
Companies who are more interested in the appearance of distribution hacking than the results are playing a dangerous game of “success theater” by going through the motions, without understanding why these tactics were chosen or how to measure if they were working or not.
Quite often we heard the decision to execute on the tactic was driven by board members of a company pointing out “it worked for Twilio”. They were completely focused on the what and how, with little insight into the why of the strategy, since they didn't have access to our internal tools or processes. We met with many of these teams to help them out, and I encourage you to always remember that no matter how well something appears to be working for someone else you should always keep your own metrics and trust your own results above all else.
Distribution Hackers: Your Job Description
The distribution hackers job is to conduct hundreds of experiments and objectively quantify the results to discover unfair advantages, and exploit them.
The ultimate goal is to build systems and processes that create a long-term competitive advantage for your company that will last long after you are gone.
It starts with you, but over time you will build a team to scale your processes as you make more discoveries and layer in more tactics. By the time I left Twilio, I had hired dozens of people who worked for me or on other teams to execute on the distribution hacks we had turned into business processes.
Distribution Hacking is a Methodology, Not a Tactic
This blog might be more popular if I listed a roadmap of unconventional tactics to try. While I will explore specific tactics I've used with analysis around their performance, the reality is that there is no quick fix for figuring out how you can get attract, convert, retain, and monetize customers. At the request of many readers I will be digging deep into various vertical markets, business models (B2B, B2C, B2B2C, C2C) and demographics.
Distribution hacking is about more than executing a collection of tactics. It is a methodology of experimentation designed to discover and exploit opportunities gain results and own a new channel.
There are plenty of shortcuts you can try to get short term results, and I think its important to have those in your toolbox, too. For an example of that, check out my post Go Ahead, Feed the Trolls about gaming Hacker News votes by engaging with negative commenters.
While these tactics are great for a quick hit of traffic, ultimately distribution hacking is about thinking and acting for the long term by re-imagining your company's entire approach to product distribution.
“Dude, are you feeding the trolls with your Feed the Trolls post? This is so meta.”
Common Internet knowledge tells us “don't feed the trolls”. This post is about when it is okay to give them some tasty snacks, while luring them into your trap to do your bidding. See this Hacker News comments thread as instructional material for this post.
Often when I put something up on Hacker News, or when I see a news story that I am invested in getting to the front page or top ranking you'll hear me say, “Okay… now we need some trolls” in a tone of mischievous delight. But I'm totally serious.
Warning: This tactic is not for the faint of heart, or those with medical conditions triggered by adrenaline.
Bring in the Trolls
On Hacker News, any comments are likely to help your post be more interesting for clicks, votes, and additional engagement. A nasty comment (hence my desire for trolls), or one that asks questions and invites conversation is the best. Each comment can individually be voted up as well, adding to the velocity and overall score of your post.
While a throwaway account is okay, what you really want is someone with a few Hacker News karma points in there.
Protip: Instead of campaigning for votes you might consider campaigning for comments and see if it helps your posts make it to the front page. Don't ask people to be negative, ask them to share what they really think and to encourage more conversation. There's no need to make the thread quality crappy. This post is about how to make the best of it when things are already going downhill.
An Example: My First Post on this Blog
I submitted the announcement of the launch of this blog on Hacker News, and after 10 minutes it only had 2 points. “Well, that's a dud” I thought as I headed out to Taco Bell with my cofounders. Oh well, lots of stuff I post to Hacker News doesn't get voted up - no big deal and on to the next one. I figured it didn't matter if we tripped the voting ring algorithm since it wasn't going to make it to the front page anyway, so I asked my cofounders to vote it up. Hmmm… that actually worked, 4 points at #17 on the front page. Not what I expected.
And then came the throwaway (brand new account) comment:
“If every IT person I've ever hated assembled together like Voltron and wrote a blog, I think it would be a lot like that one.”
Yes, we have a live one! Its not the best kind of comment since it is from an account with very little weight, but its a start. Before I could even begin to execute on my engagement strategy (usually a retort that could drive additional conversation) two other commenters chimed in.
Boom, from #17 to #3
Keeping the Magic Going
Hacker News is a huge traffic driver (if you don't know, now you know!) and once you're on the front page and particularly in the top 10 spots its all about figuring out how to stay there as long as possible to get the most possible click-throughs to your content.
Stop thinking about my silly post, imagine if this was your company's launch. The same mechanics apply.
So now what do we need? More trolls!
“If you're going to promote yourself on HN, is it too much to ask that the link contain some useful content? Otherwise it looks like your main skill is self promotion.”
Psssh
This particular flavor of comment I like to call “sour grapes” and its usually good for a tired little debate about how HN is going to shit (yay, more comments!) and I'm always up for a little instigating. I'm dmor in the thread by the way.
Btw, I was fortunate enough to have someone come to my defense, but really you don't have to defend yourself against the claims of trolls and I don't suggest asking others to come to your defense… its not about winning, it just about making a bit more heat to stoke the click-through fire.
Here's the thing - as you're going about doing all of this fancy footwork there are people voting up the post. If not, it will not remain on the front page, comments are not enough.
If you don't post something of AT LEAST minimal value then no amount of hacking the system is going to work. This is an optimization play, to maximize the amount of time your article spends on the front page once it gets there.
Even better than feeding trolls is engaging in meaningful conversation in the comments with people who respond positively (or negatively) and constructively to your post. Intelligent dialogue is what Hacker News is intended for. These tactics are for gaming the system when your post doesn't generate the beginnings of those conversations and you need to jumpstart things with only a couple crappy comments to fuel the fire.
Don't Let Them Get to You
You've got to have a thick skin, because the Internet is mean and nasty place. Feed the trolls candy and get them to help your posts resonate as long as possible to drive traffic, but don't ever think you can win them over with intelligent debate.
Here trolly trolly trolls!
Followup Question: Does this tactic work other places besides Hacker News?
I think this could work anywhere there is a voting system, including Digg and Reddit, but I haven't tested it. In general, trolls in comments on your blog, forums, etc. can be ignored or moderated out if they are really abhorrent. Its still better not to engage them there, as it only serves to bring the whole tone of the post down to their level. This works on Hacker News because the majority of visitors click straight through links on the front page without ever viewing the comments, so tone is less important and doesn't usually reflect on the author of the post that was submitted.
I've been asked a lot lately about being a growth hacker. People want to know how to become one, what one does, what one is. It seems to be the new startup job title du jour - possibly replacing the trendy “community manager” of 2007.
At first I wasn't super excited about being labelled, but I've had “distribution hacker” in my Twitter bio for quite awhile, and I like it. It feels like decent shorthand for describing what I do. I just never thought it would be so trendy. So I'm over my backlash and ready to embrace spreading the word about doing work that is analytical, results oriented, and – gasp – winning the respect of developers who have traditionally been skeptical of marketing.
Part developer, part marketer, part analyst, part product manager, part social coordinator. What the hell am I? I stopped worrying about that awhile ago, instead I focus on the incredible things that can happen when I put this diverse skill set to work. I'm a distribution hacker and proud.
This blog explores distribution hacks.
We'll dig into hacks I've employed and ones I see working for others. I'm not going to be overly concerned with definitions, or whether something is technically distribution hacking or just really well executed marketing. The fact is there probably is a very blurry line between the two. It really doesn't matter.
I'm going to focus on things that deliver results. I will also provide methodologies, tools (like some crazy spreadsheets), and candid analysis based on advising over a dozen companies, growing Twilio's developer community to 100k strong before my departure in April of this year, and building my company in YCombinator this summer.
Advice at Scale
If you have some specific questions or areas you'd like me to address just drop me a note at danielle@refer.ly or tweet to me at @DanielleMorrill and I'll try to keep a steady stream of answers flowing. Enjoy!
