Originally seen on 500.co
Today I’m going to share with you 4 “crystal ball” analytics tricks for accelerating growth.
I learned these tricks the hard way — I was doing them wrong for 2 and half years.
If you read closely, hopefully I can save you from the headaches I went through, not to mention then tens of thousands of dollars I wasted from not having my analytics dialed in.
Alright, I know what you’re thinking “This lady is no oracle, how is she going to tell me how to get more customers for my business when I’ve tested a ton of channels that didn’t work.”
Well, I’m going to stop you right there. Hey, believe it or not, it’s not your fault. It’s hard to translate all the data your business produces let alone know what to measure.
But before I tell you how to build a crystal ball for you business, let me quickly tell you a bit about myself.
I’m Tammy Camp, currently a Partner at 500 Startups, the most active venture capital firm in the world with over 1,500 companies in our portfolio.
I specialize in growth marketing, so that means I help 500 portfolio companies with their growth strategy and implementation. Aka, getting more customers.
Over the course of 14 years, I’ve generated hundreds of millions of dollars in affiliated sales for companies such as Nordstrom, Expedia, Eddie Bauer, eBay, Amazon and Priceline — all while working for myself.
I also was on the demand generation team at Walmart Labs building out their product API that generated tens of million of dollars in revenue in the first few months.
In a nutshell my career has been 100% focused on getting more customers and generating more revenue.
I was Doing it Wrong
Before I pivoted my current company, Action Factory, I was spending about $100K per month on advertising AND MAKING A PROFIT.
Now, it didn’t start that way because I thought throwing up the Google analytics tracking code on the header of every page was enough.
Seriously, I thought if that tracking code was on the header of every page, analytics would work like magic and alert me if something was wrong in my business.
There came a point in time early on when I was at my wit’s end with stress because I was barely breaking even with the cost of advertising and my business revenue, let alone paying payroll and keeping the lights on.
One weekend I decided to take a break from the stress of running a business and visited a northern Californian farm.
While on this farm, I realized I was isolated with no Internet and no mobile phone connection.
I was trying to relax, but I was worrying if something went wrong with the advertising campaigns, I wouldn’t be able to manage it because I was 100% unplugged.
Being unplugged that weekend was probably the best thing that ever happened to me.
It wasn’t because I came back relaxed, because I didn’t come back relaxed……It was because I made a mistake on a campaign end date and it ended 3 days prior when I arrived on the farm.
When I realized this, I immediately ran all my reports and discovered the most shocking thing — I had made the exact revenue with spending only 50% of my advertising budget for that weekend.
Let me say this again — I made the same amount of revenue with spending only 50% of my advertising budget.
“How could this be?” I ask myself.
Immediately I started breaking down what had happened.
From various reports, I found I was only making profit from a couple traffic sources and the rest was not profitable at all.
It just happened that the campaign with the non-converting traffic ended while I was up at that farm with no Internet connection.
I was basically lighting dollar bills in flames of fire with the other traffic sources for over the course of 6 months wasting tens of thousands of dollars.
Do you know how irritating it is to waste money as an Asian person? I mean, it’s like an abomination in my culture.
In the end, I felt pretty blessed to have found this sooner than later.
But as this unfolded I realized this one thing.
This could have all been avoided if I knew what to measure with analytics and how to measure it.
The “RIGHT” Things to Measure
Here at 500 Startups we use Dave McClure’s infamous “Startup Metrics for Pirates”. These metrics makeup the customer lifecycle, which is the DNA for a startups success. We call them AARRR:
Acquisition – users come to the site from various channels
Activation – users enjoy 1st visit: “happy” user experience
Retention – users come back, visit the site multiple times
Referral – users like the product enough to refer other
Revenue – users conduct some monetization behavior
We use these metrics to determine the outcome of our growth experiments.
For instance, with the overwhelmingly large amount of acquisition channels out there, you’d be crazy not to have your analytics dialed in before you start your campaign.
Again, take it from someone who’s lost tens of thousands of dollars with this mistake.
And if you are still and the fence about it, take a look at this example of what the acquisition channel landscape looks like these days.
To put “AARRR” into motion, here is an quick example of the customer lifecycle and conversion behavior. Clearly, there is a lot going on here and it nearly impossible to track what is going if your marketing metrics are not clearly defined.
The Progression of Analytics
A very quick history lesson here. Bear with me… you'll see why it's useful in a sec.
The Internet has come along way since it became popular in the 1990s and ALOT has evolved. Remember web counters? Well, that was the start for analytics.
We all had them and who ever’s numbers was the highest was the “baller of the month”.
The 2000s were far more advanced with analytics companies like Urchin who provided not only page view data, but time on site, session time and country. By the way, Urchin was acquired by Google and that is what is now Google Analytics.
Let’s fast forward to 2015…
Today we can literally track probably more than you are comfortable with (cough, thanks Edward Snowden)…. with dozens of analytics tools that you can see here.
BTW, there are so many analytics companies now that other companies have been created to aggregate all of this data (like Segment).
The 4 “Crystal Ball” analytics tricks
Now as I promised earlier, I going to share with you 4 crystal ball analytics tricks that will help you be the oracle of your business data.
Please read this next section closely as this is probably the most tactical, yet important part of what I’m trying to convey today.
1. Create a Tracking Plan
This is an essential component of your team docs. Everyone on the team should understand the growth objectives for the organization and be onboard with what to measure in the tracking plan.
Here is an example of a tracking plan that we use with our 500 Startups Portfolio companies.
As you can see the core business objects are outlined PLUS the actual developer code is here.
It really important everyone on the team is on the same page and this is a great exercise that includes both the businesses and your technical team.
It’s also helpful in the following cases:
- New hires can understand the explanation behind these events
- You can plan the analytics implementation schedule with your developers
A great question to ask in this meeting is:
“What would make the biggest impact on our growth curve in the next 30 to 90 days?”
2. Carefully Select Events
Q: What is an event?
A: An event is a description of the action that is taking place.
A common mistake is that people want to track everything.
But, how do you know what is important?
The problem with this is that you don’t know what is important when you go in to do your analysis. There will be too much noise versus signal.
Always ask “why?” you are tracking each event that you choose to track.
Events should directly reflect your business objectives over the next 90 days.
Choose 3 events. YES, just 3 events.
Here are a few examples from e-commerce, SaaS, and a marketplace business.
Put the detail in the properties
What is a property?
Properties are traits that make up the event.
Here’s an example:
Ask yourself, “How many new sign ups did we have today?”
The event in this case is event = signedUp.
The properties related to this event would be userId, userName, email, type (organic, referral, paid).
Remember, Data First! Always be sure to prioritize theintegrity of your business data.
4. Define a naming convention
We recommend our portfolio companies name properties in camelcase (lowercaseCapital form) for easy reading.
This an example of a poor naming convention. I’m not sure what is what here.
We recommend the naming convention “Object” + “Action” (in past tense).
Here is an example of a correct naming convention:
As you can see it’s much easier to read and understand from the one that was done incorrectly.
As you consider putting your crystal ball to work for you, imagine how lean your company will become after you trim the fat of unnecessary spending and how much more attractive your company will look to potential investors.
It’s time to make your growth story an exceptional one by taking action with the 4 “crystal ball” analytics tricks I shared with you today.
PS — if you want more Crystal Ball Analytics, see the slides (lots more images) here.
Special thanks to Diana Smith at Segment, for her contributions to this idea and post.