Tomatoes

Editor’s Note: This is a guest post from Helen, our VP Product & Design.

Do you ever get that feeling, as we all run around vibe coding and sticking every piece of software in the world together, that we're just playing out the scenario where you decide its time to start a garden, spend $150 on supplies, and end up with eight edible tomatoes at the end of the season? 

Sometimes that's fine, because it's about the tomato journey and the squirrel friends you make along the way.

But if you were a chef at a restaurant, and you had to feed 500 people per night tomato soup, starting a garden just to put tomato soup on the menu is probably far from fine.

AI and no-code tools are giving us a lot of “tend your garden” moments right now, and it’s had me thinking a lot about the real cost of those choices and how maybe having the ability to do seemingly anything is highlighting that fact that you really can’t. At least not if you want to be good.

I don’t want to yuck someone else’s yum, and I don’t mean to suggest that people should not learn or explore new things. But there is inherently a deeper problem around strategy: Most companies aren’t that good at it.

If you start with the problem you’re trying to solve and then define which parts of that solution you’re going to be insanely good at (and, by extension, which parts you’re OK with being insanely bad at), you can end up deciding very quickly what you should spend your time on and what you shouldn’t. 

Let’s say you’re an ecommerce brand. You probably have a list of choices of what you want to be insanely good at: 

• Product

• Business Operations

• Logistics 

• Brand

• Direct Response Advertising

• Customer Experience

Software/digital product might be on the list, but Shopify has mostly taken that off the table. Everyone seems to agree that being insanely good at homegrown ecommerce isn’t a worthwhile endeavor anymore.

And while that is true for Shopify, it’s not always as true for things that happen on top of Shopify. You might improve the core product with additional tools or homegrown solutions to make you better at something else, like business operations.

As a very Intelligems example, you might want to use a free shipping threshold progress bar to drive AOV. You could build that yourself. But should you build that yourself? 

If the core thing you’re aiming to do with that solution is enable better outcomes via the offer (because you’ve decided to be insanely good at business operations), your homegrown solution would have to grow in complexity. It would have to support testing the offer. Or personalizing that offer to a specific customer segment. Or changing the offer based on a number of other variables.

Would that really be the best use of time?

Having a DIY mindset might save a few dollars on software, but the true cost is often hidden in the time it takes to manage that solution and the opportunity costs from not being able to spend that time on the area you’ve decided to win. In this case, where you’ve decided to be insanely good at business operations, you need to know whether the shipping threshold is correct, whether promoting the threshold is correct, and whether gamifying progress against the threshold is correct. So, the core thing to solve for isn’t actually the progress bar, it’s the data the progress bar creates. And that’s a lot to build yourself.

(This sort of feels like a tie-in to Drew’s newsletter last week about menu costs; if you’re building the execution, the menu costs are higher than if you’re buying the execution.)

The question of where and when to invest often comes back to what is your zone of genius? What are the core competencies that make an impact for your brand and/or in your role? 

If you haven’t spent the time thinking about that yet, it’s probably worth it. Otherwise, you might end up digging for small potatoes.