Recently, I took on a new challenge that has stretched my boundaries a bit. If you’re into product management, I recommend it to you, too. I opened a store.

Product management is one of those gigs where it just takes time and experience to build your skills. But like a lot of things in life, I’ve found that I’ve improved and gotten better not when times were easy and good, but rather when I’ve struggled. I’ve long said that the best training ground for becoming a product manager is not to hop right into working with a first-class, dominant product, but instead, to start in the bottom/middle of the pack, where you have to fight and scramble. My first PM gig was perfect in this way: we were fighting tooth and nail against a much larger, aggressive competitor on one end, and commoditizing, “free” solutions on the other. Interesting times.

Any product manager, but especially in enterprise software, has to try to build empathy with their users. We usually do this through tried-and-true methods like customer interviews, job shadowing, questionnaires, and many other forms of research. These are all effective when done well, but they’re really attempts to make up for the fact that most of us just don’t have the same experiential background that our users do. The career paths that lead into a software vendor’s product team and a retail brand’s digital marketing group are usually quite different. I’ve never run a digital analytics group for a major corporation, or a brand’s data science center, or a company’s digital marketing division, and nor have the vast majority of my peers. I know most of these folks’ pain points pretty well… but do I really get it? That’s been a nagging doubt that has always bugged me.

I was pondering this not long ago when an idea struck me. Why not open my own ecommerce store and give it a go myself?

I’d long had this idea for a simple product for developers that I wished existed. I’d even created a prototype. I did a little research, found a fulfillment vendor, and then figured out a basic business model. Nothing fancy – just COGS plus margin. I set up a very simple store with an ecommerce platform and boom – I was in business!

I’ve placed a high premium on automating as much of my store’s operation as possible. In case you’re new to my blog, I’m a pretty busy guy, with a demanding job and a baby at home. I don’t have time to be mucking around with an ecommerce store every day. So I don’t. I’m at a steady state now where I mainly just check the analytics and ads report and fulfill an order every now and then when one comes in.

The hard part that I face now is the same one any online vendor does: acquiring customers economically and converting them. This has been the focus of most of my experimentation.

(By the way, before you ask: no, I’m not linking to my store here. The reason is mostly because I’m doing this as a learning exercise, and don’t want to taint my results. Also, it’s a pretty basic affair.)

Working with a tightly constrained customer acquisition budget has meant experimenting with a couple of different ad platforms. The most effective, by far, has been Facebook. AdWords was a total waste of money, and reddit is approaching that point as well (which was a bit of a surprise, to be honest). I may investigate Pinterest in the future too, but it’s clear that Facebook is it as far as digital ads are concerned. As I’ve spent more time with Facebook’s ad products, page insights and reporting, it’s really very obvious why they seem to eat global advertising spend at will. Their reporting leaves a lot to be desired, but the level of precision of their targeting and, more importantly, the traffic the platform is capable of sending just isn’t rivaled.

On-site conversion has been a challenge. I’ve made changes gradually based on my analytics, and even gotten an idea for a few product tweaks. One thing that I’ve noticed gets lost in a lot of the digital marketing hype out there is that most things you try do not “work.” It’s really very rare to discover those tweaks that produce a meaningful result you care about (like a conversion, product view, clickthrough, etc.). When you find one, though, it’s important to move fast. When I discovered with Facebook’s ad manager that two of my ads were responding well (they got high relevance scores, which is a key metric Facebook’s algorithm uses in deciding ad placement), I replicated them and created similar ones. This produced a demonstrable traffic boost. Cool.

I’ve also plunged into the community of ecommerce forums, which is a whole world of people doing the same things (except that most of their stores aren’t just learning projects). You hear the good, bad and the ugly there, but most people are struggling with the same issues: how to best use the same ad platforms to bring customers to their sites. You see people with $250 of ad spend producing thousands in revenue, and others with thousands in ad spend and about $250 in revenue, and both parties feeling like they’re doing everything wrong. And now, I get it.

What does this all mean to me as a product manager in marketing tech? Well, first, I get the thrill, and the panic, of seeing your traffic numbers spike and then plunge, and waiting expectantly for those order notifications to roll in. I get the disappointment and then surprise at seeing which platforms are delivering reach, and at what cost per impression or click, and discovering which are worth my time. Most of all, though, I get the frustrating mystery – what’s driving my converting traffic? Where did that order come from? What sub-segments could I better target? And so forth.

Oh, by the way – I haven’t made a ton of money from this experiment. I’m hoping to break even by next month. Customer acquisition is a component of my business model that I’m still learning how to incorporate profitably. I don’t feel so bad about this, though, given that even the biggest retailers on the internet are trying to figure out the same thing.

Learning by doing. Learning through struggle. I highly recommend it.

[mc4wp_form id=”185″]