Will I report earnings accurately, to the best of my ability in this post? Of course. Will I disclose all instances where I need to use “fuzzy math” or can’ t make a true “apples-to-apples” comparison? Of course. Did I delay reporting the earnings until several strategies I had hoped would work started to bear fruit.
Ok, let’s get to the numbers. We’ll start with the best Apples-to-Apples numbers I can get.
The 80/20 Drummer – Sales from March 15 – April 15 – $621.67
Ad Spending over (roughly) the same period – $471
Here’s where it’s a little fuzzy – I calculated the ad spending form the Facebook invoices, which are billed on fixed dates, so unlike with Gumroad, which allows me to specify an exact period of time, I have to take what they give me. (For some reason Facebook disabled their feature that used to let you track your spending over the last 28 days right on the front page. I think I know the reason…)
To give you an idea of the trend lines, I’d like to zoom out, then zoom in.
Total facebook ad spending, last 6 weeks – $602.59
Total Revenue, last 6 weeks – $782.58
So, last 6 weeks, that’s profit as 23% of total revenue.
Last 4 weeks, 24% of total revenue.
Last 2 weeks – $322.83 revenue, $232.52 Ad spending, for 27% profit.
Am I splitting hairs a little? Yes.
Let’s also not forget website hosting at $20 per month.
Now let’s talk about why I’m not particularly concerned.
First and foremost, I’m earning profit. So while my brand is gaining thousands of views (and surpassing 1000 facebook likes), I’m still making money. In the last six weeks I’ve done a ton of “scale” experiments – let’s see what happens if I double spending on this ad, and cancel spending on this other ad completely. I got “slick” with it, installing “conversion tracking” from Faceboook on my pricing pages to track which ads were sending the most people to the pricing pages.
It’s a pretty simple application of the Pareto Principle, and one that mirrors the theme of my series – find what’s working and do more. Find what’s not working and do less.
Here’s the problem. “Working”, when you’re dealing with Facebook ads, turns out to be difficult to pin-down.
It was the darnedest thing. I would see that one particular ad was “converting” extremely well, so I would triple spending on that ad, and I would see that another was apparently performing “poorly”, so I’d cancel spending on that one. And guess what happened? Sales dried up.
What was going on? To pin that down I had to get away from Facebook’s analytics and use an independent source: Squarespace itself. On Pat Flynn’s advice, I created six parallel conversion ladders. Each of six different ads sent people to its own landing page, each an exact duplicate of the others save the urls. If you landed on page 5 and clicked “learn more” you’d land on More Info 5, and if you clicked “see pricing” you’d land on “pricing 5”, etc. If the ads were converting in the traditional sense, you’d expect to check back in a week and see which pages were getting the most hits. For instance, if “add to cart 5” was making it rain, you’d know ad 5 was converting well, regardless of what Facebook told you.
Are you ahead of me? What do you think I found? Well, here’s a screenshot of the analytics for the last week.
That’s a bit of a lot to digest, so let me break it down-
Home 1 – 45 hits. Checkout 1 (I count this as a “sale” because it means someone has clicked “add to cart”) – 8 hits.
Home 2 – 107 hits. Checkout 2 – 1.
Home 3 – 31 hits. Checkout 3 – 3.
Home 4 – 8 hits. Checkout 4 – 0
Home 5 – 94. Checkout 5 – 1
Home 6 – 19. Checkout 6 – 0.
Now, before we ad in ad data, let’s understand what’s happening. Hardly any checkout page has any hits except for checkout 1. So that means Ad 1 is converting the best, right? Since 8 people of 45 respondents clicked “add to cart”?
Here’s why. When somebody lands on one of the home pages, he or she will only reach the checkout page for that conversion ladder if he/she clicks directly through to the checkout page, meaning they’re an instant conversion. And respect to those folks, because that’s no how I buy.
To get a more nuanced picture, consider this number. People who landed on the main homepage – 45 as we said. People who landed on the main “learn more” page? 68. Which means they’re reaching it from other places than the home page. The site has many detours, like “about me”, “browse free videos”, etc., and the button to return to the chapter previews from each redirects to the main chapter previews page.
Ergo, very few people are buying right away. They’re clicking around the site to get the gist, before some of them land on the checkout pages, and others leave the site…for now.
Consider a second data point. No matter what changes I made to my ads, if I cancelled one ad and replaced it with another sales would drop off for about a week. Whatever the reason, that means that people who saw most ads aren’t converting until more than a week later. Think about it – you see a product pitch, you click through to get the gist, but maybe you aren’t ready to buy. A few days later, maybe you’ve seen some videos, and decide the time is right.
The bad news about that is it means it’s almost impossible to get granular about conversion performance. If someone sees an ad, then comes back after 3 weeks to buy, who knows which ad it was. So I have to be modest about my ability to pinpoint exactly which ads are most effective.
There is an outlier, though, if you look again at the numbers. Ad 3 is producing an out-of-proportion number of instant conversions. 31 homepage hits and 3 checkouts. $25.82 total spent in the last week, for a little over $8 per conversion. But remember, that’s only instant conversions. Here’s what the ad looks like –
From a ground-level perspective, what does this mean?
Well, most ads are succeeding in generating interest, which gets people to the site, then the videos and the sales pages make their pitch, and they either buy later or don’t. But the “smart people” ad is different. Somehow, it’s creating a synergy between expectations and delivery – selecting for the exact people who will eventually buy, and giving them just the right expectations so that they know what they’re getting into. It’s a pretty flattering picture, actually, that people see an ad that says drum lessons for smart people, then, after visiting the site, decide, “yea, that’s about right.”
It looks like the trend line on conversions for this ad have gone down. Still, It’s probably worth my while to up the spending on this one to $6 daily.
Ultimately, the reason I’m not sweating the ads so much these days, beyond the fact that I’m making money from them, is twofold. First, the current incarnation of the product is only the first. I’m releasing the final chapter soon, and I may commit to a $10 price increase, which I think will more accurately reflect the value of the product. I’ll probably lose a few buyers for whom price alone is the major qualifier, but I’ve been looking at my data, and very few of my buyers are that. Higher price is not in-and-of-itself a guarantee of higher margins, but it’s a step in the right direction. Mandarin-language versions of the series, and higher production-value/updated versions are both on the horizon.
The second reason I’m “at ease” about things is that I’m playing the “long game”, inspired by Seth Godin’s wisdom in Permission Marketing and The Dip. Eventually I want ad spending to be only a small portion of my traffic, as most people find me through YouTube. I’m also pitching the mailing list almost more than I’m worrying about short-term sales, because I know “1000 true fans” whose lives my free content has enriched will both be a valuable resource in deciding what new products to launch, and also willing to buy those new products.
And if you’re an 80/20 Drummer customer, or prospective customer (or even prospective employee) reading this because you googled me, welcome. I was inspired by Pat Flynn’s example to make a lot of this stuff transparent. I’m also continuing to conduct an experiment in entrepreneurship to see how viable Tim Ferriss’ ideas are for “regular folks”.
I have a lot of thoughts about the direction I want my life to take, and some outsourcing anecdotes to share, but I think 1320 words is plenty for one post, so I’ll leave it there and be back soon with another.
One thought on “Belated Earnings Report – Mid March to Mid April”
Still super cool! I like how you are calling it an experiment to test the theories you have read about. That makes good sense to me!