The Business Idea That Passed Every Test and Still Didn't Get Built

A comment from a vet, made in passing, is where MyLocalVet came from.

The complaint wasn't dramatic. Independent vets were losing the retail side of their business, the food, the flea treatment, the accessories, to the big pet retail chains. Woolworths owns Absolute Pets. Shoprite owns Petshop Science. In September 2025, Spar launched its own pet retail chain, Pet Storey. Three of South Africa's largest retail groups now compete directly for the products a vet used to sell over the counter almost by default. That was the whole comment. Not a pitch, not a problem statement someone had thought through. Just someone naming something that was bothering them.

Most people let a comment like that pass. I didn't, and not because I'm more entrepreneurial than the person who said it. I have an enquiring mind. Ideas turn up constantly, more of them than I could ever act on. That was never my problem. My actual problem, and I'd argue it's most people's actual problem, was never a shortage of ideas. It was knowing which ones deserved my time.

Frustration is not inspiration. It's data. Most advice about finding a business idea stops at "notice what annoys you," as though noticing were the hard part. It isn't. The hard part is what you do after you notice, whether you run the thought through a filter before you fall for it, or whether you go straight to building because it felt good in the moment. MyLocalVet is a business that cleared every filter that matters and still didn't get built, and that's a more useful story than one that ends in a launch.

The numbers

• Three of South Africa's biggest retail groups now run national pet store chains: Woolworths (Absolute Pets), Shoprite (Petshop Science, expanding rapidly since 2021), and Spar (Pet Storey, launched September 2025) (News24, May 2026).
• South African households own an estimated 7.4 million dogs and 2 million cats, the country's first officially reported national pet count (BrandMapp/TGM Pet Care Survey, via News24, May 2026).
• Poor product-market fit is the single biggest cause of startup failure, responsible for 43% of the 431 venture-backed company failures CB Insights analysed in its 2024 update (CB Insights, 2024).

Where the sparks actually come from

Most articles on this topic point you to two places: your own daily friction, and other people's complaints online. Both are real, and both are worth building a habit around.

Your own friction is the easiest to access and the easiest to get wrong. You notice something because it's costing you time or money, which tells you the problem is real for at least one person. It tells you nothing about whether it's real for anyone else, which is why "I built this for myself" is a fine starting point and a terrible finishing line. Log it anyway. Write down the specific moment, what took longer or cost more or felt clunkier than it should have, not the business idea it might become. The polished idea comes later, if it survives the filter.

Social media works the same way, if you do it properly. Scrolling isn't research. Searching is. Go to where a specific complaint already clusters, a subreddit, a Facebook group, a review section, and search the phrases people actually type when they're venting: "so annoying that," "why is there still no," "I wish someone would." Then count. The value isn't in how strongly one person feels. It's in how many people say a version of the same thing without being asked.

There's a third source that gets skipped in most of this advice, and it's the one that produced MyLocalVet. Other people's throwaway comments, in rooms you're already sitting in. A colleague mentions a broken process in a Monday planning meeting and everyone nods and moves on. That nod is market research most people let evaporate. You don't need a personal grievance for this one, and you don't need a social media habit. You need to actually listen when someone names a problem near you, instead of waiting for your turn to talk.

For a corporate professional specifically, this third source is probably the most useful of the three. You're sitting in more rooms with more different people than you realise, and some of them are naming real, specific, recurring pain in front of you every week. It's one of the corporate scaffolds worth using deliberately while you still have it, the access disappears the day you leave the building.

The three filters

An idea earning your attention is not the same as an idea earning your time. Here's what I actually ran MyLocalVet through, and what each filter is for.

Is it commercially achievable? This asks whether the pain is real, widespread, and attached to people you can reach. The vet's complaint passed this cleanly. The pressure wasn't a hunch, Woolworths, Shoprite, and Spar were all extending retail muscle into a channel vets used to have mostly to themselves. And I had something most people chasing a first idea don't: a specific vet who trusted me enough to pilot it. That's the market test and the access test passing at the same time. If you can't name a specific person who has the problem and might say yes to your solution, you haven't cleared this filter yet, no matter how sound the logic feels.

Is it operationally achievable? This asks whether it can actually be built, by you, with what you have. I built the platform. A real practitioner agreed to pilot it. The model exists at real scale overseas too, US veterinary groups run near-identical platforms, individual vet practices with their own branded storefronts, coordinated backend, so this wasn't a fantasy I'd invented. It worked. This is the filter people obsess over and the one that's easiest to pass with enough time and money, which is exactly why it isn't the one that decides most ideas.

Is it sustainably achievable? This is the filter almost nobody writes about, because most idea-generation advice quietly assumes you have unlimited time. It isn't asking whether the idea can be built. It's asking whether it should be the thing you build, given everything else already competing for your hours. My pilot depended on a close relationship with one vet, access to her systems and processes that made an "easy" pilot possible. When she emigrated, replacing that wasn't a small task, it meant finding a new partner and effectively rebuilding the access the pilot depended on. At the same time, I had other opportunities that were clearly worth more of my limited capacity. I chose to put my time there instead. That wasn't the idea failing. That was me weighing it correctly against everything else on my plate and making a call.

The line in the water

Here's the part that matters most, and it's easy to miss if you're only counting launches as wins. I got MyLocalVet to a real answer, a working platform and a real practitioner willing to put her business behind it, without spending money I couldn't recover. That's the actual goal of testing an idea. Not proving you were right to have it. Getting a genuine answer, cheaply, before you're committed.

There's an old line that fits here better than anything I could invent: you need a line in the water if you want to catch fish. Most people never put a line in. They keep the idea in their head, safe from ever being tested, which also means it never earns a real answer. I put a line in. It caught something real. I chose not to keep reeling, because I had other lines in the water that mattered more right then.

"I tried and it failed" is a common story and a mildly useful one. "I validated it properly, for nothing, and still walked away because something else earned my time more" is a rarer and more honest one. If you're sitting on two or three ideas right now, the second story is probably the one you need, not the first.

What this means practically

• Start a running log of friction, yours and other people's. Capture the moment, not the polished idea. Most of what you log will go nowhere, and that's fine, that's the log working.
• Treat comments made near you, in meetings, in passing, as data. Don't wait for a formal problem statement. A throwaway line is often the whole signal.
• Before you build anything, name the specific person who has the problem and would say yes. If you can't, you're not testing an idea yet, you're admiring one.
• Ask whether the model has been proven anywhere else, even in a different market. It doesn't make the idea yours, but it tells you the physics work.
• Run the sustainability question honestly, on a deadline. Not "could this work" but "is this worth my next six months more than what I'm already doing."
• If you walk away, write down why. That reasoning is worth more the next time a comment lands near you, and it will.

Building this habit matters more the longer you stay employed, not less. The pressure inside corporate careers isn't easing off, and the earlier you get good at spotting and filtering real opportunities, the more options you'll have when you eventually need them. Ideas like MyLocalVet are exactly what Stage 2 of the Launchworks stress-test is built to interrogate properly, the market, the operations, the money, and whether it fits the life you actually have, before you spend a rand more than you need to.

Frequently asked questions

How do I find a good business idea?

Most good ideas aren't invented, they come from noticing a real frustration, your own or someone else's, often in an ordinary conversation, and then filtering it properly. Test whether it recurs, whether you can name and reach the people who share it, and whether it's genuinely worth your time before assuming it's worth building.

What is the most common reason startups fail?

Poor product-market fit is the leading cause, responsible for 43% of the 431 venture-backed startup failures CB Insights analysed in its 2024 update, ahead of running out of cash (29%) and having the wrong team (23%). The lesson is to validate real demand before you build anything.

Do I need to spend money to test a business idea?

Not necessarily. A real pilot conversation, a working prototype shown to one genuine prospect, or a single committed partner can validate both demand and access before you spend a rand. The test is whether a specific person says yes, not whether the idea sounds good to you.