More Accurate Acquisition Funnels with Mixpanel

I’m a big fan of Mixpanel for analytics. I’ve written about how I do A/B tests in Mixpanel, but over the past week I learned some really important things that I’d like to pass on.

Imagine a typical registration process from a homepage. The person visits your homepage, decides that they are interested, and creates an account. Clearly this signup flow can easily be tracked in Mixpanel. It might look something like this.

mixpanel-funnels

What’s the conversion rate on this funnel above? Here’s where things get interesting….

Returning Users

Typically if a user is authenticated, they are sent straight to the application dashboard. That’s great, but what happens if a user isn’t authenticated and decides to visit the homepage?

They would be tracked in this funnel.

If a visitor hits the homepage for the first time, Mixpanel sets a cookie on them, which tracks future actions and knows who exactly triggered a specific event. The “Viewed Homepage” event isn’t smart enough to track returning users from new users, which is why we need to instrument this.

How to Create Better Acquisition Funnels

The easiest way I’ve found to fix this takes a little work, but it’s not too bad. It doesn’t require a bit of thought around what events to track though. In short, you need something set on the user when they create an account. I’ve quickly included some code below to show the “signup funnel.”

//Homepage
mixpanel.track("Viewed Homepage");

//Registration Page
mixpanel.track("Viewed Registration");

//Visitor Fills out information and hits "create account"
mixpanel.track_forms("#id-of-button", "Account Created");

Now how do can we filter out new visitors from past visitors on that first “Viewed” Homepage event? For example, if someone makes it through the signup process, and then hits the app dashboard, we could set information on the user….something like this. This could be account_type, user_created, really anything that shows that someone made it through the signup flow before.

mixpanel.register({ account_type: "$whatever" });

Then, in our funnel, we would want to filter by people where account_type is not set. Mixpanel allows you to do this easily :
mixpanel-account-type

 

You should now be able to more accurately track how new users interact with your signup flows, especially how they respond to your homepage. I hope this helps, and if you have a better way, feel free to let me know in the comments.

Growth/Marketing Interview Tips: My Learnings

Over the last few weeks I finished up my time at YesGraph, and have been exploring other opportunities and doing some contract work in my free time.

I’ve had the opportunity to connect with several companies here in Boston and SF, and the demand for marketers who work on driving growth is very high. This should serve as a no-brainer (as every startup is looking for exponential growth), but I was surprised and happy that companies are placing a high priority on this skill-set/mindset.

I’m compiling this post because there’s many articles about specific growth “hacks”, but very little information about what it’s like to get hired as one of these growth/marketing people. I’ll try to explain my learnings along the way and provide practical insight on how to navigate this process.

interview-tips-growth-marketers

Actively apply or wait?

Similar to marketing, hiring is very much a numbers game, yet certain channels are higher quality than others.The companies that I connected with was segmented into three major channels:

  • Inbound – people that find me via Twitter, blog, LinkedIn, meetups
  • Outbound – actively applying to companies who are seeking to fill a position
  • Referrals – personal friends/networks that know of companies looking to hire (typically no job posting)

At YesGraph I learned a ton about how hiring works, so I decided to run an experiment. I actively applied to 3 companies, and then reached out to friends/mentors to see if they knew of any opportunities that I should look into. I also posted my employment status to followers on Twitter/LinkedIn.

At this point in time, my primary goal was learning more about the company, and seeing if there was a mutual fit.

Here’s how the top of the funnel looked:

  • Inbound – 4 opportunities
  • Outbound – 3 opportunities
  • Referrals – 3 opportunities

 Learnings:

The companies that I applied to are well known in the tech space, yet I was disappointed by their lack of process.  I blame this largely on the channel being low quality, and I am “just another applicant” in these scenarios.

There also was a major difference in the attitude of the company; for inbound and referrals the companies were very interested in meeting, but for outbound channels, they seemed very passive. Put simply, actively applying to jobs is a low quality channel.

On the other hand, inbound interest was the best channel. These people were excited to talk, and every company I met I could see myself working for.

What does growth mean to you?

There’s a bit of buzz around driving growth, so the first thing I would do is seek to understand what “growth” actually means to the company. I think it’s easy to get caught up in hype, so my focus was to fully understand what needle they were trying to move. I might be able to help, I might not.

I measure opportunities based on potential impact. My goal in every job is to deliver more value than I take.

Writing Code

I personally believe a marketer is most effective when they have a solid understanding of code. This doesn’t mean you need to be a programmer, but instead, you should be able to diagnose issues with metrics, and not look stupid when talking with the engineering team. *important*

I tried to make it very clear that while I understand more programming than most marketers, that should NOT be the reason why they hire me. I view writing code as the vehicle to delivering business value, and if you can write it yourself as a marketer, you make this happen faster without external dependencies.

I know many marketers who build websites with Weebly and are much more effective than a marketer who can hack. It’s because they understand what really matters….and it’s not writing functions. It’s making the company $$. As a marketer, make that your goal too.

Specialization

As much as I love working on a variety of marketing tasks, it’s also important to understand what you’re good at (and where you need to improve.) It’s okay to be upfront and tell a company that you specialize on one aspect of the funnel. Just be honest, if the company likes you, great! If they aren’t interested, you likely saved yourself a headache down the road. Don’t over-promise and under-deliver.

Timing

Timing is key. I met with a local startup here in Boston, and they are doing some great things, but it was apparent that the timing wasn’t right. They still had some product work they needed to do, and driving growth wasn’t the most important thing they needed to work on. As a young startup, focus is critical. If a company isn’t ready to use your skills, don’t try to force it. Once again, you will save yourself a headache in the future.

Less Magic, more Process

Over the course of the interview process, many people asked about specific tactics that I’ve done to help a company grow. I’m happy to share examples, but I focusing on tactics can be misleading. For example, let’s say I increased conversions by 300%, and the sample size was 10 visitors. That’s useless. Instead focus on process, and how you come to specific tactics.

Last example: I’d rather hear an Airbnb employee talk about how they figured out that Craigslist was a great channel for cross-posting instead of exact implementation details.

Understand your Worth

As a marketer, you are constantly dealing with metrics. As a result, you should be able to track how much value you are delivering, making it very easy to prove your worth. For example, If I implement an A/B test that improves the bottom line by $20k, that should be a big deal to my boss. Growing a company is a team sport, but if you move the needle in a big way, don’t be scared to ask for what you’re worth…just be reasonable.

This might be controversial, but what’s the difference between you and someone in sales? You just made the company $20k. A top sales rep is compensated according to the deals that they close. You are a sales rep on steroids. Take a look at Patrick Mackenzie’s talk on this….he gets it!

Compensation

What about compensation? I’m not going to disclose exact numbers, but I found companies are looking to pay an engineering-level salary for someone with these skills. I think that’s quite reasonable, but always measure in terms of value you bring to the company. That’s the determining factor.

I found startups to pay lower than market rate as cash is tight, but they offered equity. I personally view equity like lottery tickets, so it doesn’t sway me very much.

Tip: Don’t ever mention a specific salary compensation level first. If a company is looking to hire someone, they have already established this beforehand.

What matters most to you?

Alone same vein as compensation, you must make sure to figure out what matters most to you. For me, this was my criteria:

  1. Interesting challenges
  2. The team
  3. Flexibility/opportunity to move fast
  4. Work/life balance
  5. Post product-market fit
  6. Compensation

Biggest Learning: Don’t discount BigCo’s

I enjoy working in a startup environment because everyone’s moving fast & there’s flexibility, but that doesn’t mean big companies can’t do the same. At first I was not interested working at a larger organization, but many of these companies have spin-off projects that they are trying to grow in-house. Better yet, they have more stability than most startups.

For me, a big deciding factor is whether these side projects have buy-in from senior management. If the CEO is behind the project you are working on, that’s a positive sign. If he/she doesn’t care, you may want to look elsewhere.

With that being said, I’m joining Safari Books Online, where I’ll be working on Safari Flow. It’s a cool product with tremendous upside, and I love reading, so it’s going to be a fun adventure!

The Startup Marketer’s Experimentation Process

Over the past month, I’ve really tried to take a hard look at how I personally do marketing online. Specifically, I’ve been trying to create processes that I can use at any company as I progress through my marketing career. I wrote about how I onboard myself at a new company, but my goal with this post is to dive a bit deeper.

Before I jump into specifics, I believe that establishing a process around marketing experiments is one of the biggest competitive advantages you can have. So many marketers try tactics, and never document results (I’m guilty of this.) Even worse, there’s no “formula” to how you learn. This is bad, and I’ve been trying to change this, so I figured I’d share my process around experimentation. It’s still a work in progress, so any feedback is appreciated.

Startup Marketer's experimentation process [Read more...]

Gallery Leather Journals vs. Moleskine

For some reason, my friends in the tech space are obsessed with Moleskine notebooks. While I still don’t understand the proper pronunciation of the name, I decided to break down and purchase one yesterday. Here’s what it looks like below:

moleskine-notebook

The total price tag was $20. Simultaneously, I decided to match up a Gallery Leather notebook for the same price. Gallery Leather is based out of Maine, and clearly being from Maine, it’s important to be true to the “Motherland.”

I decided to pick up the Oporto journal – it has the closest resemblance to the Moleskine.

Gallery Leather Oporto

What I like about Moleskine

  • Minimalist design
  • Elastic enclosure
  • Rear pocket
  • Dense paper quality

Overall, I don’t think Moleskine is a bad notebook. The quality is decent, plus there’s a “cool factor” that they’ve built as a brand.

What I like about Gallery Leather

In short, the Gallery Leather notebook is extremely high quality. Backed by a no-questions asked return policy, this notebook is amazing – the binding is much better than the Moleskine and it’s obvious that they care immensely about the craftsmanship.

If I had the option, I would prefer not having “journal” on the front, but blends in with the cover and is barely noticeable. Put simply, this notebook makes me proud to be from Maine – it resembles everything a hard-working Maine-based company cares about.

 

Republic Wireless Moto X Review (and why I’m ditching my flip-phone)

Last year I wrote a post explaining why I do not have a smartphone. In a nutshell, I work in front of a computer all day (am constantly near wifi) and think spending $100/month is insane for something I truly only need for a few specific occasions.

Well, fast forward a year, and I finally made the decision to get a smartphone. It came with quite a bit of consideration around when I would use the device, and whether or not I found value in being constantly connected.

Fortunately, I don’t have to sacrifice a great device for an expensive contract. I just purchased the Moto X with Republic Wireless, and I’ve been amazed with the device as well as the quality of service I’ve received.

Moto X Review

First, the device is awesome. There’s a pretty extensive review below. There’s also Google Now, which is pretty cool. It’s essentially a Siri competitor, just not as smart. The phone is $300 dollars through Republic Wireless, which is quite a bit cheaper than trying to buy it elsewhere. I’m very price conscious, but this phone will pay itself back very quickly.

The setup process with Republic Wireless is very easy, I was making phone calls and checking my email within minutes. The camera isn’t that amazing, but I’m happy (especially since I didn’t have a camera on my old phone.)

Why Republic Wireless?

First, for those unfamiliar with Republic Wireless, it’s a low-cost service that uses a combination of wifi and cell signal (from Sprint) to make and  receive calls. For example, if I’m home, the device will be connect to the wifi, which makes the cost of calls/texts next to nothing for Republic Wireless.

If you’re away from the home/office, it then automatically switches over to the cell signal – it’s a hybrid phone of sorts. Here’s a quick video to show it off:

The best part about this approach is that Republic Wireless can offer “smartphone” plans for a very low cost to you. They currently have 4 packages to choose from:

  1. $5/mo (unlimited talk/text/data from a wifi signal, but no cell access from Sprint)
  2. $10/mo (unlimited talk/text/data from wifi and unlimited talking/texting from Sprint, but no mobile data plan)
  3. $25/mo (unlimited talk/text/data from wifi and unlimited talk/text/3G data from Sprint)
  4. $40/mo (unlimited everything, including 4G data from Sprint)

I chose the $25/mo plan. I need to periodically use Google maps, Uber, and check my email while traveling for work, so it’s a perfect fit. Considering I also work from home, nearly 95% of my time using this device is connected to wifi.

Something else that’s really cool about Republic Wireless is that you can switch plans (up to 2x/mo.) For me, I may consider downgrading to $10/mo unless I make trips for work and really need the data, which I will then simply upgrade.

Put simply, I have an amazing phone and spend $25/mo, while the rest of the population pays over $80/mo per phone. I truly feel like I’m hacking the system!

Downsides of Republic Wireless?

When I first received my phone, there was an error when booting up the phone for the first time. I didn’t see it again, and proceeded to use the phone.

I did notice after talking on the phone with someone, about 5 minutes into the conversation I could not hear them. I also noticed that there was a small “blinking” of the background, so I shot Republic Wireless support an email. Within hours, I had a replacement on the way – I received a shipping confirmation in the evening, and it was on my front steps at 8am in the morning.

I setup the new phone, and haven’t had any problems since. If I would change one thing about Republic Wireless, I would consider adding a few more “onboarding” videos, since the process may be confusing for less tech savvy individuals. Overall, their documentation is very good, but I think a few tweaks would be nice.

In Conclusion

I’m going to purchase the same phone for my wife shortly – which will bring our total for 2 lines to about $50/mo. This is amazing! I was spending $25/mo through AT&T for a flip phone with 100 minutes of call time. I will NEVER go back.

Lastly, there’s very little risk to try it because there’s a 30 day money-back guarantee. That was the nail in the coffin for me.

If you’re looking to purchase a smartphone and don’t want to break the bank, you need this phone!

Tips on finding a Marketer for your Startup

Two months ago I started working at YesGraph. It’s a team of 7, and we’re launching our paid tier in the near future.  My role is what I’d classify as a startup marketing role, yet it’s dramatically different than my past role at Boundless (also a marketing role.)

It’s not a difference in company culture, or the fact that YesGraph is a distributed team, but instead it’s based around one important distinction. Product-market fit. 

Has your company found product-market fit? The best explanation I’ve seen is by Sean Ellis:

I’ve tried to make the concept less abstract by offering a specific metric for determining product/market fit. I ask existing users of a product how they would feel if they could no longer use the product. In my experience, achieving product/market fit requires at least 40% of users saying they would be “very disappointed” without your product.

If there’s one takeaway from this article, it’s that if you are looking to hire a marketer for your startup (or join a startup as a employee) you need to find out if the company is in “search mode” or “growth mode.” I’ve created an ugly chart below to summarize what I’m going to discuss.

Growth Mode vs. Search Mode

In the rest of the article, I’m going to break down the different distinctions I’ve experienced as a marketer. [Read more...]

The Remote Worker’s Onsite

I’m currently on week #3 of working remotely for YesGraph, and last week I had the opportunity to travel to Palo Alto, and work with fellow remote employees on-site. At first I thought this would throw a wrench in my plan to measure my productivity, but I realized that this is a perfect opportunity to do some benchmarking.

For those of you who haven’t seen my earlier posts:

  1. My plan to track my remote work experience
  2. My first week as a remote worker

An argument from critics of remote work says that remote employees aren’t as productive as people who are on-site. There’s so many variables in making such a statement, but at the very least, I should learn what works for me on a personal basis. This post is a culmination of my learnings.

Productivity

From a productivity standpoint, my assumption was that I would be much more productive on-site, due to the fact that I could focus solely on work. This turned out to be accurate, although my productivity in terms of hours worked was very close to the week before. I started work on Monday, and flew home on Friday (which I didn’t track.)

Monday

Monday Productivity

Tuesday

Tuesday Productivity

Wednesday

Wednesday Productivity

Thursday

Thursday Productivity

Overall, a pretty productive week! If you were wondering what the “news & opinion” was above, this is what happens (reading blogs) when you are doing research on competition and learning more about the market.

Team-Building

A clear goal of being on-site was to get to know each other – we definitely could have worked the entire time, but building a culture was a huge focus.

I’ve noticed many startups point to inanimate objects as symbols of a healthy company culture. The beer fridge, ping-pong table, and free snacks are easy scapegoats, yet as a distributed team, we don’t have this luxury. We have to look at things like communication, honesty, and trust. These traits are much more important (and yes, much tougher to communicate/track.)

For example, during the course of the week we had lunch as a team, and used that opportunity to get to know each other. The perk (free lunch) is what I’d call a culture-creator, opening up the opportunity to build a bond with fellow employees.

yesgraph-palo-alto

Timing

I’ll be the first to say that working on-site has certain benefits. For example, as a young company working on major product improvements, it’s easy to get feedback and move forward when working in-person. In this situation, I look at working on-site as a tool in the remote worker’s toolbox. It’s something that can be very beneficial in certain phases of a company’s growth, yet is something that should probably be used sparingly (a few times a year.)

Onboarding

Being a new employee, this was a great week to build relationships with fellow coworkers, and better understand the company and market. Most of my competitive research can be done remotely, yet working on-site is particularly helpful when mentally “crafting” a pitch on how to tell the YesGraph story.

This week was extremely helpful in this regard, as I learned more about the vision and focus of the company in the next 6 months.

 

In Conclusion

In a nutshell, working together is awesome, and is perfectly suited for specific occasions. If you’re a distributed team, I highly suggest doing this a few times a year. It’s a great opportunity to bond as a team.

End note: it’s tough to write weekly, so I’m going to measure new experiences that I have as a remote worker, instead of constantly writing about productivity. Here’s what you can expect:

  1. I’m going to visit coworking space and track results (productivity, etc)
  2. I’m doing a home exercise program (T25) and seeing what happens (it’s really tough to exercise as a remote employee, plus I’m really lazy too)

 

Have feedback on this post? Let me know on Hacker News.



Working Remotely: Stats on Week #1

If you didn’t have a chance to read my first post, I’m working remotely for YesGraph, and attempting to document the entire process with data from Rescuetime and Fitbit (as well as keeping detailed logs on what I do during the day.)

I’ve never worked a full-time remote job, so these posts are an attempt to share my learnings. I’m also hoping that by publishing the data, it forces me to develop good habits. Here’s how my first week went, although I started my documentation process on Tuesday.

I’ll first begin by summarizing high-level learnings, and then dive into the specifics later.

High-Level Learnings

This will be a constant work in progress
If I learned anything over the course of the week, it’s that working remotely will be an iterative learning process. Like any skill, this is something that will take time to get used to.

Human Interaction is Important
I’d consider myself a social individual, so going from constant interaction with coworkers to working on my own is a significant shift. I’ll discuss this more in detail later.

Productivity (Am I working too much? Or not enough?)
I expected this to be the case before I started working remotely, but I think it’s not easy to balance work hours. Working remotely values what you produce, over sinking man-hours into a problem, which I find fascinating. It’s also pretty challenging to measure (which is why I’m using Rescue Time as a benchmark.) Fortunately YesGraph is focused on results, not man-hours.

Employer Trust
One of the most eye-opening aspects of working remotely is the sense of trust that you feel. Working remotely is a privilege, and it’s surprisingly rare that a company trusts you enough to know that you will produce high-quality work, no matter what location you work from.

Methods of Communication
When working on-site it’s easy to resort to in-person communication as a first action. This is a blessing and a curse. It’s a blessing because human interaction is a great way to build a bond with co-workers and quickly find the answers to certain problems you may face, but it’s a curse because it’s often destroys coworker productivity.

Exercise isn’t Easy
It’s easy to stay inside all day, and this first week I didn’t exercise nearly as much as I should have. This will be a major effort of mine in the future to change.

Ownership of Time
An interesting aspect about working remotely (and measuring results over “butt-in-seat” time) is you feel an intense amount of responsibility to maximize every minute of your time. At a “normal job” when you get bored, you might browse Reddit. While working remotely, you still might get bored, but instead of doing a mindless activity, you can do household chores. You feel more ownership over your time.

 

Daily Routine

As I mentioned before, I tried to track as much as possible, and below are the data points from each day of the week. My goal is to show that this is a process of experimentation.

Week #1 Aggregate Stats

Overall, it seems like I was pretty productive, below is my RescueTime pulse. It’s important to note that these stats are NOT over a complete 40 hour week (I started this experiment on Tuesday and did onboarding tasks on Sunday.) While I’m happy with a “92”, I think I can do better (and should do better.) It’s important to note that RescueTime is not and end-all solution on productivity, but I think it’s the best solution out there.

 

Overall Productivity

Overall Productivity

Next, you can see which days I was most and least productive.

Productivity by Day

Productivity by Day

Next, you can see my productivity throughout the day. I don’t see anything obvious, but I felt more productive in the morning. Clearly the data shows otherwise.

Time of Day Productivity

Time of Day Productivity

 

Exercise

Unfortunately my Fitbit had a problem syncing on Tuesday, so I only have stats for Wednesday, Thursday, and Friday. I’ll break these down into the daily reports below.

 

Daily Reports

In the daily reports, the purpose is to show my attempt to do something new everyday. My overall learning is that mixing up activities is one of the most exciting parts about working remotely, yet certain activities can help/hurt your productivity.

Tuesday
On Tuesday, I went to a coffee shop to work in the morning (8:40am – 11:20am), and worked from home in the afternoon. I got a decent amount of exercise this day, but unfortunately my Fitbit crashed and I lost the data.

Tuesday Productivity by Hour

Tuesday Productivity by Hour

 

Wednesday
On Wednesday, I didn’t get much exercise at all (the goal is 10,000 steps/day), yet I was pretty productive throughout the day. I worked the entire day from home.

Wednesday Productivity by Hour

Wednesday Productivity by Hour

And for exercise…

Wednesday Exercise

Wednesday Exercise

 

Thursday
This was my most productive day of the week (in terms of hours worked), and ironically, this was also the one day out of the week that my wife worked from home as well. Take a look at the stats below. Compared to other days, there was more fluctuation with my productivity, but overall, it was pretty awesome to work from home together.

Thursday Productivity by Hour

Thursday Productivity by Hour

 

Thursday Exercise

Thursday Exercise

 

Friday
I would love to see overall stats of the productivity of the American workforce on a Friday. I ended up walking nearly 4 miles before beginning work at 9am, yet for some reason my productivity suffered. Perhaps this was because it’s the end of the week, I’m not sure? You can also see my productivity started dropping rapidly in the afternoon. While I think this is quite normal, my goal is to stretch more productivity out of my Friday afternoons.

 

Friday Productivity by Hour

Friday Productivity by Hour

 

Friday Exercise

Friday Exercise

 

Final Note

This week I’m working on-site at YesGraph HQ, so it’s nearly pointless to report on my remote productivity or exercise. I am tracking my productivity on-site, as to benchmark the difference between working remotely and in-person. I’m really excited to see what I learn!

 


Tracking A/B tests in Mixpanel using url parameters (Rails Application)

So you have a Rails application, and you’re trying to setup A/B tests. You probably also aren’t in the mood to pay for Optimizely  because it makes more sense to roll your own. I understand, yet you also want to be able to track the results of these A/B tests in Mixpanel (specifically in a funnel.)

You’re reading the right article – I’m going to dive in and show you a super-simple way to set this up. [Read more...]