Tech | Startups
Barely 3 years old, Chowdeck is arguably Nigeria’s leading food delivery company — depending on who you ask. Unlike many startups that keep metrics private, Chowdeck’s leadership has publicly disclosed some numbers including the total value of orders and number of registered users. This article attempts to assess the validity of the latter.
How does a food delivery business see a 5,000% increase in registered users when food inflation has grown by 52% over the same period (2022–2024)? Is it possible for most of this growth to happen within the same period Nigeria has seen a significant currency devaluation and an increase in the cost of energy?
The Numbers Do Not Add Up
An over 187,000% growth in user base in 3 years is insane. But, the problem begins when you check the number of downloads on the Google Play Store, as every registered user should first download the app.
By October 19 2024, Chowdeck’s Playstore downloads stood at 100k+. Technically speaking, 1m is 100k+. It can be argued that the Play Store only displays milestone numbers with 1 as the leading digit such as 10k, 100k, 1m etc. This would mean that until Chowdeck reaches 1m downloads on the Play Store alone, it would still display 100k+.
Play Store Downloads = Total number of unique users who have ever installed this app on one or more of their devices
However, Play Store records numerous milestone numbers before 1 million. While 1 and 5 are the most popular leading digits there, it still records other leading digits — rare, but still recorded.
Notwithstanding the rarity, Chowdeck’s download numbers should display 500k+ at least. What could be the issue with Chowdeck’s reported 1m registered users and the 100k+ download number on the Play Store?
Here’s a list of possible theories.
Theory 1: More Downloads from the iOS Devices
Chowdeck is not only available on Android. This makes it possible to theorize that the company may have more downloads from Apple’s App Store. Since the App Store does not publicly share downloads for each app, it is easy to lean into this theory.
But hollup…
Nigeria is an Android market — a massive Android market. Statcounter puts Android’s share of Nigeria’s mobile Operating System (OS) market at a whopping 87.85%, with iOS having 9.22%. Can Chowdeck realistically have the majority of its users on iOS?
Per Semafor, 7 in 10 Chowdeck orders are from Lagos. We cannot say with certainty which parts of Lagos account for the bulk of the orders, but if the number is largely the bourgeoise and not the comrades, it may be feasible to have more orders from iOS devices.
But orders are not app downloads. The gap between Android and iOS market share in Nigeria is too big for Chowdeck to have its iOS downloads significantly larger than Android. There’s a different conversation to be held on downloads vs active users but we’re sticking to the former as every registered user would first download the app.
Nigeria’s smartphone penetration — % of the population with a smartphone — stands at 10–20% (Statista). Doing some tablecloth maths, using an estimated population of 220 million and the more positive estimate for smartphone penetration (20%), we would have 44m smartphones in Nigeria. Plugging in the OS market share;
- Android (87.85%): 38,654,000 million smartphones
- iOS (9.22%): 4,056,800 million smartphones
For Chowdeck to have 500–800k users on the App Store alone, it would need 12-20% of Nigeria’s iOS market and this 20% will need to be concentrated in the cities in the eight states it currently operates.
Going by smartphone penetration and OS market share, this theory fails to convince. Chowdeck will need to be an outlier app for this theory to hold.
Theory 2: Web Users Are Significant
Chowdeck is available for use on the web. I never used the Piggyvest app in my short period with them and used Cowrywise via the web for a long time before downloading the app. For these companies, their users don’t need their apps, as usage is not frequent. This makes web use reasonable.
However, for web users to make a dent in Chowdeck's total user base, they have to be a lot. A LOT. If this were the case, Chowdeck would be seriously bucking the trend. Per eMarketer, of 4 hours Americans spent using the internet, less than 30 minutes was on the web. That’s a whopping 88% of the time spent with mobile apps. The odds that Nigerians exhibit a different behaviour are slim.
This theory also fails to convince.
Theory 3: Chowdeck is Counting Differently
The number of downloads and registered users are not the same. Ideally, downloads should exceed registered users if we account for those who install and don’t create an account. This is based on the assumption that Chowdeck counts registered users as only those who create an account.
Chowdeck currently allows for ordering without creating an account, meaning downloads should be higher than registered users, even if it counts such users as registered.
Then there’s Chowdeck’s ecosystem of apps. It operates three apps — vendor hub, rider and food delivery — for the three groups it works with; restaurants, riders and customers. Could it be that Chowdeck is counting all three and reporting the sum?
This is a plausible explanation but it would mean Chowdeck has been misleading the public with its reporting. No one reads 1 million registered users and thinks of riders and restaurants. Play Store downloads for Rider and Vendor Hub are 10k+ and 5K+ respectively. Hardly makes a dent to the numbers.
The count theories are not valid.
Theory 4: A Pot of Lies
The CEO is lying.
This is a hard one to lean into as it has numerous implications for its users, investors, and just about anyone who has bought its dodo metric. It is not new for startup founders to cook metrics. And no, it’s not a “Nigerian founders” thing. There is a long list of founders around the world who have cooked numbers—resist Nigerian exceptionalism.
While WeWork and Theranos are popular examples of companies that exaggerated their performance and misled the public, many others abound. Luckin Coffee got delisted from NASDAQ and received a $180 million fine from the US SEC after reporting misleading metrics to look more successful compared to Starbucks in China. Ozy Media — a Series D startup — inflated its audience numbers and deceived investors, leading to the conviction of its CEO.
Lottery.com, Headspin, Imagine Express, Outcome Health and Nikola are some less popular examples. Nikola released a video of their truck “in motion” when it was just rolling down a hill. Then there’s Dash which inflated user numbers and transaction volume leading to its shutdown in 2023.
As Chowdeck’s founders are all Great Ifẹ̀ alum, I am not inclined to hold on to this theory. Yes, I am biased. But cooked numbers are not a reality we can’t stomach. It will be disappointing but we have seen worse.
The theory holds but it is not palatable.
Theory 5: Play Store Updates Download Milestones (Very) Slowly
In conversation with three software engineers, it was clear that no one can say with certainty how frequently the Play Store updates its reported download numbers to users on the listing page. While the developer console has better numbers, the listing page tends to slowly display the updated figures for the estimated numbers.
A search on Quora and Google’s developer support emphasizes that there is no predefined duration for updating the download milestone. The question is, how long is too long? Two weeks? A month? Three?
The problem here is that if Chowdeck crossed 250k since 2023, this would imply that in almost a year, Play Store has not updated Chowdeck’s download numbers. Is this the same for other apps across the store? Nigerian apps? Or just Chowdeck?
Piggyvest’s website claims it has 5m+ users and its Play Store downloads stand at 1m+. This could be reflective of the 1 and 5 leading digit preference of the Play Store, keeping the number at 1m+ until they cross 5m on the Store. Then there’s Kuda’s reported 7m users with a 5m+ users download count on Play Store. The difference is 5x and 1.4x for Piggyvest and Kuda respectively, compared to Chowdeck’s 10x.
Diving into Reddit, the same complaints over Play Store’s download milestone update emerge again but this time with a solution. A Redditor recommended using an app called App Finder to get it.
Searching for Chowdeck there reveals 440k+ downloads in total, with 23k monthly downloads.
Très intéressant
If we assume Chowdeck has exactly 1m users and the Play Store download is exactly 440k, it means Chowdeck has a significant non-Android user base. Even in the scenario where it has a significant number of web users, Chowdeck will need to have a large iOS user base that deviates from the Nigerian smartphone market norm. With the 87.85% Android market share, Chowdeck’s total number of users should not exceed 500k. But in this case, Chowdeck’s Android base is around 44% of its total user base.
Recall that Chowdeck unlike pure software apps needs to be physically active in a city to be useful to the Nigerians there. This makes the significant iOS user base plausible, as Chowdeck is not only active in 8 states, but the urban centres in these states; essentially serving the bourgeoises. Chowdeck is an outlier app.
This theory is valid and convincing.
I thoroughly enjoy thinking aloud. It allows you to see holes in your thought process, disproving the crap and gathering evidence in favour of the most convincing one. Ideally, I’d trash all the other theories and write on the valid theory. But you can’t deny it made sense to walk with me to find a credible answer.
For emphasis, Play Store only reflects unique app download numbers. We can safely assume that a registered user is anyone who creates an account with Chowdeck, via the mobile app or the web app. There’s a separate conversation to be had on whether it is a relevant metric for on-demand delivery compared to average order value and active users — measured by order frequency.
Chowdeck’s likely outsized iOS user count does create new questions worth exploring. Is Nigeria’s iOS market share growing significantly? Especially in urban centres? Does the mobile smartphone market share skew significantly from state to state? Worth finding out.
Till then, Chowdeck retains its reputation — reporting valid numbers.
PS: All reported numbers are valid as of 26th October 2024.
