25 October 2016


Retired Member

1,977Posts 6,439,663Views 2,303Comments

Finding risk from Facebook: Friendlyscore

21 August 2014  |  3890 views  |  0

"We're glad we're not the smartest people in the room"

It's a little strange hearing these words come from a man with a masters degree in Automatics and Robotics. A man probably as comfortable with rule-based programming and natural-language parsing as I am with brushing my teeth. But Emilian Siemsia, CTO of Friendlyscore speaks genuinely, as does his partner, Maciej Dolinski, CEO.

Maciej explains himself "We feel we are not the smartest guys in the room - it's a positive thing - if we were the smartest, something would be very wrong."

When Nektarios Liolios, MD of Startupbootcamp, told me about the application process for the 10-week programme, I was taken aback. Skype interviews, pitching days, one-on-ones, hundreds of applicants whittled down to 10. But when Emilian, a man who's clearly very intelligent, tells me how inspired he is by the other teams he shares the space with - it made sense. Startupbootcamp trusts its teams as much as the teams trust Startupbootcamp. It needs to.

"We're not saying banks are doing it wrong, we're saying times are changing"

Credit scoring is one of those sectors ripe for disruption. It's archaic, it's discriminatory, it's generally a bit short sighted. Essentially, Friendlyscore looks at your social media footprint: the way you interact on Facebook, who you talk to and what you say - to build up a different picture of your credit risk. They look at the demographics of your Facebook friends, how many different circles you have, even the way you type (Here's something I learnt: more exclamation marks = you are more dominant).

Maciej tells me a story about how he's technically a credit bureau nightmare - he's had several different jobs - is self-employed and has moved from country to country, apartment to apartment - are banks wrong to mark him as a credit risk?

He thinks they are.

Going back to that discrimination thing - I thought 'Hold on. Aren't Friendlyscore just as short-sighted as the credit bureaus? They're just judging me because I use too many exclamation marks or occasionally swear. Ha! Not that friendly, are you?'

But that's not exactly how it works - they've teamed up with psychologists in Poland who've created profiles of people. They've determined the correlation between the words we use and our creativity, our openness, our dominance and other competencies that can be connected with reliability. The sample is big and it's growing. Teaming up with a payday loan company in Poland their data grows more accurate every day.

"I believe in doing things from my heart; I've worked for a big corporate. For a bank. For a loan shark. I lacked doing something positive in my life"

I learnt the most about Friendlyscore when they spoke passionately of their favourite project - Kano. A do-it-yourself computer, cheap and cheerful, that opens up the world of programming and development to new markets. It breaks down barriers of class, prejudice and money. Is this what Friendlyscore are hoping to emulate?

So I asked them about money. "Would you sell Friendlyscore for £10 million?" "Right now?" he asks with a slight smile. We all laugh.

Having worked at Getin Bank and Procter & Gamble, he's familiar with the corporate life. The cliché of being a small cog in a big machine. The agonising lack of job satisfaction. But now his focus is on bringing about some genuine change to this world.

'Is it important to you that people who couldn't get a loan are able to get one now - because of Friendlyscore?'

Stupid question. Quick answer. 'Yes.'



This is the first part in a series of features taking place at the Startupbootcamp in London. Part 2, same time, same place, next week. 

Maciej and Emilian: 'We're not the smartest guys here' TagsRetail bankingInnovation

Comments: (0)

Comment on this story (membership required)

Latest posts from Retired

Fintech innovation in the B2B space has only just begun

12 September 2016  |  10853 views  |  1 comments | recomends Recommends 0 TagsPaymentsInnovation

Protecting Data with DLP

23 August 2016  |  5045 views  |  0 comments | recomends Recommends 0 TagsSecurityBrexit

How to end what ails online commerce

22 August 2016  |  4598 views  |  2 comments | recomends Recommends 0 TagsPaymentsTransaction banking

What internet retailers need to know about Google’s recent webspam report

08 August 2016  |  8291 views  |  0 comments | recomends Recommends 0 TagsPayments

Modelling fixed income: Why realtime analytics are key

29 July 2016  |  5271 views  |  0 comments | recomends Recommends 0 TagsPost-trade & ops

Retired's profile

job title
member since 2014
Summary profile See full profile »

Retired's expertise

What Retired reads
Retired writes about

Who's commenting on Retired's posts

Hardeep Singh
Ketharaman Swaminathan
Graham Seel
Gerard Hergenroeder
Konstantin Rabin
Matt Schofield
Anna Robert
Ian Davis
Steve Patel
Aparty Behera
Karim Maalouf