Starting a startup is rough. If it’s your first time, the likelihood of failure is high — around 90%.
But, this is a known fact, and everyone goes into this line of work knowing this is the case. Early-stage venture capital firms have their entire strategy built around this fact. They know that most of their investments will be failures, and they prefer outright failures to companies that just doodle along and consume time but never have a chance to provide any outsized returns.
So, where are all the failures?
Well, they’re certainly not hidden. In fact, you can find them pretty easily if you just look around. They’re the startups that have been shuttered, the ones that have reorganized or sold themselves for a fraction of their former valuation, and the ones that have just faded into irrelevance.
The venture capitalists are betting that they can hit a home run for every ten or twenty failures. This means that they invest in a company that will become so large that it not only pays for all the failed investments but can return many times the value of the entire fund to the partners.
Again, everyone knows this, and this is why when venture capitalists are looking at a potential deal, the key thing on their mind is whether the startup that they are looking at has the potential to become really really big — otherwise, it just simply isn’t interesting.
Startups and the types of technology companies we are talking about have some advantages, especially when it comes to tracking metrics and being able to value and understand the outcomes.
There are a few key things that, as an investor, you would want to look out for, and this is the same for any technology startup, regardless of the specifics of what they are doing:
- MRR — Monthly Recurring Revenue is the amount that all customers have contracted to pay the organization every month, regardless of whether they use the product. This is a good metric to track because it gives you a sense of the business’s health, and whether it is growing or not.
- MoM % Revenue Growth — Month-on-Month growth as a percentage. This is because high monthly growth can translate, even from a low base, into a surprisingly large company over a few years. A company making $10,000/month growing at 30% month-on-month can expect to make over $5,000,000/month after two years.
- LTV:CAC Ratio— A customer’s Lifetime Value (LTV) divided by the Cost of Acquiring a customer (CAC). The LTV is the total revenue a customer will generate for the company throughout their relationship. The CAC is, well, self-explanatory. This ratio is important because it tells you whether the company is making more money off of its customers than it is spending to acquire them.
- Monthly Churn Rate — What percentage of the customer base leaves each month? This is important because even seemingly small monthly churn rates can have an outsized impact. Imagine if 4% of your customers leave each month; that means you need to grow 50% per year just to keep your head above water — ouch.
- DAUs/WAUs/MAUs — Daily, Weekly, and Monthly Active Users. How many people log in and get value out of the platform daily, week, and month? This is important because it tells you whether people are actually using the product, and it’s a good leading indicator of whether the company will be able to grow its customer base.
If you see a company doing well on all of these fronts, that’s a good sign that they have a chance to become really big.
These numbers are elementary to measure for any technology startup and can automatically be placed into real-time dashboards to be monitored. And more importantly, they are a standard measure across any startup, regardless of their vertical or industry.
If you sell geo-location APIs to Fortune 500 companies, or whether you sell project management software to small and medium businesses, you’re measuring and evaluating the same core metrics. And that’s what makes it possible to compare and contrast different companies and make investment decisions accordingly.
So, the question that gives this essay its title is not about startups but about projects in the NGO sphere, which we can equate to startups.
NGOs are difficult to run because they are solving problems that are not only complicated but are complex. There are no easy answers, there is a lot of potential for negative second and third-order consequences, and there is no standard metric that you can use to compare the effectiveness between projects within an NGO, let alone between NGOs.
This is not to say that there are no success stories or that it’s impossible to have an impact. But it is much harder, and the path to success is much less clear.
I’ve attempted to grapple intellectually with this last point in my essay on Cost-Per-Transaction Model, but that was just a high-level conceptualization of how such a cross-project and cross-organization monitoring system could be set up.
The other key difference between a private sector startup and an NGO project is that it is much more difficult for a private sector startup to fudge the numbers because, in the end, it all links to the amount of cash that the startup is bringing in, and you cannot fake bank statements. With an
Or, more precisely, you can’t fake bank statements for long until you eventually get caught and arrested for fraud and spend a significant portion of the reminder of your life in a concrete box.
NGO projects, on the other hand, are not funded by their beneficiaries, the people consuming or getting value from the products or services being offered. They are funded by external donors, who essentially act as venture capital, making bets on which projects are likely to work and which aren’t — and this is where things get problematic.
With NGOs, the final metric is often something much more qualitative and difficult to measure, such as “number of people reached” or “level of awareness raised”.
There is a broken element here which is the feedback loop of what is and isn’t working and what receives or doesn’t receive funding. This, generally, works very well in the private sector. Venture capitalists are pretty happy to let a startup die if it does not grow as quickly as it should.
But, because often the people working inside an NGO project are also responsible for setting the success metrics, we can quickly see how this can become self-serving, as these can be chosen to lessen the possibility of failure by picking easy-to-game metrics or my miscounting or misrepresenting.
My question “Where are all the failures?” Is that one way to statistically see if this is happening in any given organization is to ask to see the failures. If they amount to less than the equivalent in the startup world, i.e. 90%, then some questions should be asked. If the failure rate of an organization is an order of magnitude less, say 10%, then there are only a few conclusions to be drawn:
- Individuals in the organization are gaming the monitoring and evaluation systems and are essentially defrauding donors who could put funds to better alternative uses.
- The organization is picking straightforward projects, which are likely not that impactful. I am sure there is a strong correlation between how risky a given project is and how impactful it can be. After all, if things were easy, they would have been solved already.
- The organization is genuinely achieving these incredible success rates, and an in-depth study should be made to understand how this is the case.
Regardless of the conclusion, an investigation should be made if failure rates of projects are significantly lower than the private sector equivalent.