Get more trading volume with cTrader
The basic model that lots of brokers now run is to add a bunch of products, use less profitable ones for marketing, and ultimately switch people over to CFDs.
You get less revenue per customer, but you (1) pay way less to acquire a customer and (2) get so many more customers coming through that you ultimately make more money.
Robinhood made this model popular in the US and now Trading 212 and XTB have copied it in Europe. It’s also what IG appears to be moving towards doing.
XTB released its results for H1 last week and you can see how this works.
Manage your dealing desk with Centroid’s risk technology
Almost 80% of first client trades were in cash equities, ETFs, or ‘investment plans’ (similar to roboadvisory funds).
Those products generated a massive…3% of revenues.
The remaining 97% was from CFDs.
Lol
XTB is also on track to potentially have its lowest ever annual ARPU of approximately $750.
By comparison, the broker’s best ever year was 2013 when it made, using the current PLN to USD rate, ~$3,400 per client.
The difference is that in 2013 the company had fewer than 17,000 active clients. This year it will have more than 800,000.
Get a Match-Trader server today
But the weird thing about XTB’s results this time around is that the firm saw…
- Record trading volume
- Record active client numbers
- Record new client numbers
- Same CAC
…and yet it made LESS in profit than it did in the first half of last year.
Maybe I’ve read through the accounts too much but I found this weird so figured it would be fun to look at how it happened.
- Trading volume not commensurate with active client growth
The first point is that XTB saw just shy of 854,000 active clients in the first half of the year, compared to 502,554 in H1 of 2024.
That is a 69.9% increase.
CFD trading volume rose 41.5% in the period.
So the first point is that CFD volume did not rise in line with active client numbers – in fact, it was much lower on a relative basis.
Active clients at XTB traded 5.95 lots on average during the first six months of 2024. This year it was 4.95 – almost 18% less.
- CFD per lot profitability lower
The next point was that the money XTB made on CFDs was far lower this year than last.
Profit per lot last year was PLN 289 ($78). This year it was PLN 251($67.8). That means last year XTB made 15.1% more per lot than it did in the first six months of the year.
That then filters through to average revenue per customer.
This was PLN 1,900 ($513) in H1 of 2024 but was ‘only’ PLN 1,400 this year ($378).
In other words, the average customer at XTB made 35.7% more in revenue in the first six months of last year than they did in H1 of 2025.
- Higher operating expenses
The next point was that XTB did a lot of ad spend in the first half of this year compared to 2024.
That operational expense was up to PLN 264.3m, compared to PLN 156.3m last year. A big chunk of that was probably due to the partnership with Zlatan Ibrahimovic the firm has done, as well as some large scale branding initiatives it has done across Europe, which have involved a lot of out of home efforts.
Employee expenses also rose by 30% to PLN 192.7m. That is probably in large part because the company continues to spend so much money on technology hires. XTB now has 570(!) people it classifies as tech employees, up from 498 last year.
These were the two largest operational expense increases but, aside from ‘taxes and fees’, every operational expense rose in H1 of this year compared to the first half of 2024.
The result was that total operating expenses this year were PLN 608.7m compared to PLN 410.4m in 2024 – a 48.4% increase.
- Weird finance expense
Another thing that impacted the bottom line was a huge increase in what it terms ‘Financial Expenses’.
This is probably some kind of FX loss on converting into PLN but we’ll only know at the end of the year when full results are released. For context, in 2024, USD was pretty much flat against PLN in the first half of the year. This year it was down almost 12%.
To put it in some context though, for the first half of the year XTB’s loss from this finance expense was PLN 85.3m ($23m). Last year it was just PLN 513,000 ($138,497).
The bottom line
You can’t draw that many conclusions from all of this because ultimately the revenue per user and P&L per lot of CFD trading are interlinked.
More volatile markets tends to mean higher P&L and – one would assume – more active clients.
Consequently the quieter markets we’ve seen since liberation day are the most likely cause of lower net income than last year.
However, what you can potentially see is how this model can go wrong.
You are ultimately relying on a lower CAC that gives a much higher volume of overall clients, who still increase overall revenues (eg. 8x as many customers with 0.25x ARPU = double revenues).
But if you are seeing…
- CAC stays same but ARPU decays
Or worse..
- CAC increases and ARPU decays
This would be bad. eToro has seemed to suffer from this problem from time to time.
I think XTB and the others will probably continue to work but it will be interesting to see if it can continue long-term.
