Can the Workforce insights shed light into the FTX scandal? Also what kind of insights can we see within Coinbase, Binance and BlockFi?

Julian Herzog
8 min readNov 22, 2022

Roughly five months ago, Osterus took a dive into the crypto land, looking to decipher the workforce code in the world of mysterious unicorns, entrepreneurs and engineers who are trying to change the landscape of fiat currency and upset the long standing status quo in the world where finances, security and technology intersect. Back then, we never imagined we would be looking at the data again from the current perspective, yet, the unfolding of recent events provided us with a unique opportunity to learn about the red flags that were raised, and the ones that could have been in hindsight.

On a request from a family office, at that time we analyzed some of the biggest crypto players: FTX, Binance, Coinbase and BlockFi. We gained interesting insights and formed certain expectations from these companies, based on which we created a ranking and recommended a go-to company for an investment, which turned out to be Coinbase — we argued, backed by data, that their people stats were simply better. Today, amidst the turmoil in the entire market, we are proud of the results of our analysis and are even more convinced in the capabilities of our software and the data that we provide.

As it stands, things have gone south, or more appropriately, have hit the rock bottom for FTX, facing a prospect of charges for fraudulent business practices. At the time, we did not raise clear red flags for FTX, at least in terms of being able to say if this place looks like a fraudulent company with no real human capital. We did, however, raise big flags that the profiles are much less sought after on the market in comparison to Coinbase, giving them a low competitiveness score. Furthermore, we quickly realized that the retention rate was a mess for FTX, resulting in a very low loyalty score among other things. Not to bare teeth on FTX alone, we did note that Binance did not fare all too well in the loyalty department either, but still better than FTX. One observation we also deduced was that, there were not as many profiles to be found on FTX, which could be truly associated with the company. Early on, we built a Machine Learning model to help filter out “fake” profiles or at least profiles, where people mention a company, but in fact are just trading with their software or apparently worked there short term on a consulting project.

Cryptocurrency was a booming industry, well depending on your perspective for some, this was a necessary market correction and is only the start.. With studies showing a 187.5% growth in market cap, the rate of acceleration in the past 10 months was astounding. This opens the door for new entrants to capitalize on the opportunity to house crypto transactions.

FTX and Binance are two crypto-based exchange platforms that have similar foundations, yet dramatically different success rates. FTX was started in the Bahamas in 2019 and quickly became the third-largest crypto exchange by volume in 2021; however, now in 2022, the company is in bankruptcy proceedings for its liquidity crisis, suspected of fraud and running a full on Ponzi Scheme, setting the cryptocurrency market back for years and possibly not leaving space for a full recovery.

Binance is now coined the world’s largest bitcoin and altcoin exchange after being founded in 2017, with recently closing a $200 million seed round for their U.S. company, with a $4.5 billion valuation as of April 2022.

Comparing the human capital behind the scenes can help us understand where companies usually went wrong and offer us indicators, if the company is not healthy. Hence, we took a look at how Binance is relatively successfully navigating the crypto world. The data we will be using from Osterus was taken right before FTX’s downturn in economic performance. We have three different snapshots in time of FTX, one from July, the other from the beginning of November and another one from November 20th. We decided to use the data from July, as a lot of profiles have removed their work experience at FTX, for obvious reasons.

As you might know, Osterus is a software that revolutionizes company comparison through strategic insights into employment specifics. During this analysis, we used it extensively to understand and decipher the workforce DNA of these companies and its impact on their current standing in the market.

The first telltale sign something is off, or at least deserves some scrutiny, is the fact that roughly 30% less profiles are to be seen with data on FTX work experience, when compared to our data from July — one of the hallmarks of Osterus data signals, not only showing obvious insights, but pointing to significant deviations and breaks in typical patterns you might expect from industry peers, as it happened here. During that period, we also examined the BlockFi acquisition via FTX.

Comparing the education degrees of Binance and FTX, both companies yield similar results. We can see that the majority of employees, roughly 60% of the workforce at FTX, hold Bachelor’s degrees, while Binance ranks slightly lower by a margin of less than 5%. The Master’s degree category shows comparable results in the range of 12–13% for both FTX and Binance.

Both companies were looking to hire educated individuals to work behind the scenes. Since the distribution is relatively similar for both companies, we will need to look further into why FTX is a complete failure vs. Binance with the similar education levels and generally speaking familiar patterns.

Digging deeper and looking at data points such as ranking of universities and generally going through previous companies, we can definitely see that Binance and BlockFi have more reputable university graduates and company backgrounds in comparison to FTX. In fact, the university backgrounds from FTX are quite random and do not offer the patterns you would expect to see — again easy to spot using the Osterus app.

Moving onto the average work experience in the current company, we see that Binance is able to hold employees for a longer duration. Both FTX and Binance have a high number of employees that have been with the company for less than 3 years with 50% and 57%, respectively. This makes sense given that the companies are relatively new.

The 3–5 and 5–10 year categories are when Binance begins to outpace FTX with 11% to 3% in the 3–5 year category. Which makes sense, because FTX was founded in 2019, however we did find a lot of “wrong data”, when it comes to profiles claiming, that they have been with FTX before 2019.

We find most of these profiles but sometimes there are a few which are legit and possibly an error occured, when entering the year.

Binance is able to keep employees for longer durations, reducing turnover and increasing profitability per employee. FTX doesn’t seem to be able to keep employees long-term, indicating issues behind the scenes, either in retention or poor hiring.

The total work experience is also important to analyze and helps you to understand, over time who was there from the beginning and remained, vs who is relatively new. At Osterus we try to use the data and decipher, which employees have had the most impact on a company during their time.

Although Binance slightly outperforms FTX in their employee work experience, the results aren’t drastically different. If FTX has all the same tools as Binance,why did employees not raise red flags?

The loyalty and diversity scores can help answer that question. We can see that both FTX and Binance have very low loyalty scores. This score analyzes employee sentiment toward the respective company by factoring in current employee tenure, the boomerang ratio, which analyzes employees leaving and returning, the employee ratio and more.

We created a ranking at the end of the article, which might seem obvious now, but happy to read the comments and hear what you say.

FTX has an extremely low employee tenure ratio and employee ratio, with 0.04 and 0.06, respectively. We can see that Binance slightly outperforms FTX in those categories with 0.06 and 0.08, respectively. Be mindful that this is very low, on a scale from 0 to 1! Although this is a minimal difference, some of this data will help shed light on the health of a company, especially once we have more historic data.

FTX has a high diversity score, while Binance has a very high score. The data shows us that FTX falls short on education diversity, language diversity, work experience group diversity, age group diversity, and location diversity compared to Binance.

Having said that, FTX has a lot of random profiles, also our Machine Learning system picked up 300% more fake/broken profiles within FTX, than Binance, which we did not factor in.

All in all when doing a score on the human capital we currently have the following results:

1) Coinbase (Highest competitiveness and loyalty score out of the group)
2) BlockFi (Second highest, however with the highest rank on university backgrounds in terms of ranking)
3) Binance (Better than FTX)
4) FTX (no surprise here, but our data always indicated a weak score on its workforce)

The data provided by Osterus gives us key insights into why FTX may be struggling, however it is now clear, that it was incredibly flawed and most likely a lof the employees left early as they did not understand the way this “ship” was being steered. Just recently, we noticed that a lot of profiles took FTX out of their LinkedIn.

We will continue to monitor the data of FTX and Binance, updating you if any new developments appear in their employment specifics.

The ability to derive these insights wouldn’t be possible without using Osterus. From simple workforce analysis, to major disruptions, disasters and success stories, you can too be equipped to make better predictions and data driven decisions!

For information on other companies or to schedule a demo, reach out to contact us directly or on LinkedIn.