Skip to main content

The Data Gap in Women's Football & How FIFPro and Databricks Are Helping to Close It

Summary

  • FIFPro, representing over 65,000 professional footballers, had years of survey data on women's working conditions scattered across fragmented spreadsheets on individual laptops, blocking evidence-based advocacy.
  • Blend360 built a data platform on Databricks using a medallion architecture, AI/BI Genie for natural language querying, and Power BI dashboards to consolidate survey data spanning multiple tournament cycles and surface cross-confederation trends.
  • The project revealed shifting exit drivers in women's football replacing earlier maternity concerns, and the team is now targeting injury prediction and pay equity analysis for the 2027 World Cup.

The Data Gap in Women's Football & How FIFPro and Databricks Are Helping to Close It

Watch: The Data Gap in Women's Football & How FIFPro and Databricks Are Helping to Close It
As the FIFA World Cup kicks off in the USA days before this summit, billions will follow the men's game, but women footballers still fight for visibility and basic protections. FIFPro, representing 65,000+ professional footballers, is using data to change that. For over a decade, FIFPro has surveyed women footballers worldwide on working conditions, welfare and employment rights. But that data lived on laptops in fragmented spreadsheets with critical insights locked away. In this video, Blend shares how we built a data platform on Databricks to power evidence-based advocacy. We'll cover designing a medallion architecture for messy survey data spanning multiple tournament cycles, configuring AI/BI Genie so non-technical researchers can query data in natural language, and building Power BI dashboards that surface cross-confederation trends. When women's football data sits on laptops, it's easy to overlook. In a lakehouse, it's impossible to ignore.
Talk By: Alice Hodgson, Client Partner, Blend360 EMEA ; Mairi McKinnon, Data Consultant, Blend360 EMEA ;

Chapters

FAQs

What is FIFPro and why is women's football data important?

FIFPro is a global union representing over 65,000 professional footballers, including women athletes who face significant disparities in pay, prize money, and career length. Centralizing and analyzing their survey data enables evidence-based advocacy to improve working conditions and welfare for women footballers worldwide.

How did Blend360 build the data platform for FIFPro?

Blend360 designed a medallion architecture on Databricks to handle messy survey data spanning multiple tournament cycles and standardize it into a single governed source. They configured AI/BI Genie as a conversational layer so non-technical researchers can query player data in natural language, and built Power BI dashboards to surface cross-confederation trends.

What insights did the FIFPro women's football data platform uncover?

Analysis of the consolidated dataset revealed that maternity concerns, once a dominant theme in player exit surveys, have been replaced by new exit drivers in recent years. The platform also enabled salary trend analysis for injured players and free-text theme extraction at scale across the full survey corpus.

What is next for the FIFPro women's football data project?

The team is planning to build injury prediction models and conduct pay equity analysis targeting the 2027 World Cup. The lakehouse architecture means that data previously locked on individual laptops is now impossible to ignore for advocacy and policy decisions.

Full transcript

[00:09] Welcome everybody. Good afternoon and thank you for joining us for the presentation this afternoon. Our presentation is on the data gap in women's football and how thief pro blend and data bricks are helping to close it.
[00:22] We've got very good timing with the World Cup being on at the minute um and coming to talk about women's football. Um you may you may have seen us walking around earlier. We actually came in our football shirts. Um so you'll see maybe some of our colleagues dotted around.
[00:35] We're kind of representing many different countries. Um but yeah, it was quite fun uh to be it was quite noticeable this morning kind of having the shorts on. Great. Um we got 40 minutes for our
[00:47] presentation. We're going to talk for probably about 30. So we'll have about 10 minutes at the end if there's any questions. We're going to start with some quick intros. My name is Alice Hodgson. I'm a
[01:02] client portfolio leader at Blend. Blend is a specialist data and AI consultancy. We work across industries to help our clients get real value from their data.
[01:16] Uh we're based in the US. our headquarters are here but we've got uh offices globally within LATAM India and Amia and Mari and I are here today from Scotland so it's great to be in San
[01:29] Francisco it's our first time here so we're taking the opportunity to have a bit of an explore as well so we work with our clients from everything from foundational data and technology projects all the way through to um kind
[01:43] of more complex AI ML Aentic we're when you Think about us. We're the builders of data solutions. We get our hands dirty. We're not just the theory. We kind of get in get in the code and
[01:56] solve the problems. Uh and we do that with data bricks. Okay. Perfect. My role my role is um I in my role I sit next to our clients and
[02:09] the ones that probably work with them closest to really understand their business, their challenges, the problems they're trying to solve. and we work together to figure out how we can use data and AI to achieve their goals.
[02:23] So, one of our clients is FIF Pro um and we're here to talk about the project that we've delivered with them today and it's extremely transformation for women in football. Unfortunately, FIF Pro haven't been able to be here today, but
[02:36] Mayer and I are going to hopefully do a very good job of telling their story. >> Perfect. Um hello everyone. Thank you for joining us today. There's lots of you in the audience, which is fantastic. My name is Marie. I'm a data and AI
[02:48] consultant here at Blend. So over the years I've worked in numerous industries, the likes of energy, finance, pharmaceuticals, but when the project with VRO came along as a big sports fan, I was super excited. Um I'm
[03:01] sure a lot of you in the audience as well are sports fans by the fact that you're here at this talk. So to bring data and sports together into a project seemed like the perfect one for me. Great. So to kick us off, as you already
[03:13] know, Mari and I are from Scotland. We call it football. I'm sure we've got a lot of people in the audience who don't call it football. Uh so a little bit of audience interaction. Can you just put your hand up if you call it soccer?
[03:27] Okay, great.
[03:40] that. Okay, I'm going to start. Um, I've got a very short quiz. I promise it's only three questions and it's going to be easy, especially if you know the answers. I've already given you the answer to this first question, but I'm
[03:53] going to ask it anyway. There's a pretty big tournament, international tournament happening right now. H, could you please raise your hand if you have any idea what that tournament is?
[04:06] I'm trying to see if anyone hasn't raised their hand. Um, right. We've got a bit of an interactive element. Mary's going to throw the ball. And if she throws the ball to you, can you tell us the answer? >> And you get to keep the ball, so that's exciting. Just I guess everyone kind of
[04:18] knew it. There we go. >> Yay. Well done. >> Fantastic. Okay, great. Second question. The last World Cup for the men's football was in 2022. It was hosted in
[04:31] Qatar. Can you put your hand up if you know who won that tournament? Are you at the front? >> Correct. And just to remind myself, they beat France in the final.
[04:44] >> Uh, right. You can maybe see where my next question is going. In 2023, uh, the women's World Cup was hosted in Australia and New Zealand. Can you please raise your hand if you know who
[04:56] won the Women's World Cup? >> Uh, what you not the best. >> Spain. >> Yep. and they beat England. My colleague here, Sav, is in an England shirt.
[05:08] Sorry, Sav. Uh, correct. I don't know if you had the opportunity to look around and notice the difference in the number of hands that went up between the answers for the men's World Cup and the women women's World Cup, but there was a
[05:21] a pretty big gap in the number of hands went up. And that is what we're here to talk about today is the gap uh the visibility gap in women's football.
[05:35] And I'm sure none of you have kind of ever thought, "Oh, I really don't want to follow women's football." It's, you know, that it's not been a malicious or a deliberate thing. It's just the way it is. It's just the default has always been the men's game.
[05:47] And there is a book that you will have seen on the table um called Invisible Women by Caroline Crio Pere Perez um and she has a phrase for this. the way that
[05:59] women are routinely left out of the data, the way they're routinely left out of the research. She calls it a kind of not thinking. And it's not that it's malicious. It's not that it's deliberate. It's just the way how things
[06:12] have always been. And the way things have always been in football is the men's game. It was around first. It's set the standard. It's set the calendar, the contracts, the conditions, and the
[06:26] women's game has just conformed to it and not thought about what the women's game might need differently. And that's what this slide is setting out here. How can you drive change when you're visible in sorry, when you're
[06:39] invisible? >> Perfect. As Alice has touched on, um I'm from Scotland, so I would like to bring a personal example in here. And again, you can probably tell by my accent, so I
[06:52] will try to speak as slowly as possible so you can understand my Glasgian accent. It's a struggle. Um, but until this year, the Scotland's men team had not qualified for the World Cup since 1998, which I mean, fair enough, we're
[07:04] not particularly known for our football on the world stage, but we did win our game last week, so bring on our game this week. Who knows what'll happen. Um, but when we did make the World Cup this year, it was front page news for weeks.
[07:17] Headlines like this. Scots celebrate reaching first World Cup in nearly 13 year 30 years. It was a big deal, but that's not quite accurate. The women's team made the World Cup in 2019 on the
[07:29] exact same global stage and almost nobody remembers. And as Alice touched on, it's not malice. It's not deliberate. It's exactly what Perez talks about in that kind of not thinking. The default is men's and
[07:41] nobody questions it sadly. And for the women in football, what does it feel like? We've got a couple of examples just to highlight some of the disparity. Uh so as you can see on this
[07:55] slide here, uh there is a fairly significant difference in the prize pot between the women's football world cup and the men's. So in 2023, the prize pot for the women's World Cup was $152
[08:08] million. And for that same cycle, the prize pot for the men's game was $444 million. This is the same same sport, same tournament, players that have the same dedication and passion, but the
[08:21] women are receiving a third of the prize pot as the men are. Another example is it shows up in career length as well. Men in the professional game typically play till around age 35
[08:33] and for women is about 33. And two years may not seem like a big difference, but when your career in in professional football is maybe only 16 years, two years probably feels like quite a lot.
[08:45] And one of the main drivers of that is injury. When a woman has been injured, she doesn't typically have the same uh financial support as what the men's game do. Um, and quite often when they're
[08:58] injured, they don't receive their salary. uh they don't also receive the same support for rehab and recovery from injuries. So for men what may be a career setback can be career ending for
[09:11] the women. Uh and a great example of this and how um kind of equipment in sport is designed is not designed for women. ACL injuries which is a ligament injury that
[09:23] happens in your knee. It's much more common for women. I feel like someone knows this in the audience. Yeah,
[09:36] common in women and that is because the football boots have been designed specifically on men's feet. So women are playing in these shoes which are just not even designed for them and it's leading to injuries which are
[09:49] potentially ending their careers. I'm going to jump in again with a personal example. So you can see our touch rugby team up on the pitch. Um number four and number 25 at the front. You'll recognize here. Um so for those
[10:02] that don't know um touch rugby rugby but with no contact so I'm interested in it. Um we play against other businesses in Edinburgh over the summer. And if you know anything about a Scottish summer,
[10:14] the pitch is really dry. So we've started the league and I'm in my trainers. We're slipping about and I think probably should invest in some football boots. So order them. they arrive and I can confidently say they're the most uncomfortable shoe I've ever
[10:27] put on. Um, and it kind of got me thinking, I'm only wearing these boots for an hour or two each week, but think of these female athletes which are wearing them day in day out hours at a time. It makes sense why they're getting injured and it seems like a really easy thing for us to prevent going forward.
[10:42] This photo is also not representative of the Scottish weather. Just don't get any illusions. It was a one-off.
[10:56] around reason that women's um their career length is typically shorter and it's due to maternity. Women are having to choose between having a family and their career. The maternity provisions
[11:08] for women in football are non-existent because as we've already laid out the game the kind of contracts and protections were designed for men and men don't go off and have babies. they don't need to have the same maternity
[11:20] provisions. H and with FIF Pro's own research, they're reporting it's about 47% of women are leaving football early because they're making the decision to have to have children.
[11:32] Okay. So, what are we doing about it? Uh so, sorry, I've lost my train of thought for a second. So, but what if we had the data um that was actually tell
[11:47] us what women's football actually needed? Not just measured against the men's game, but from the voice of women telling us what they needed and what they wanted. And not just data that was going to lay out what those challenges
[11:59] were and what needed to change, but data that was going to be able to track those changes and see if what was actually being implemented was working and making a difference. And the good news is that data does exist. But just like the
[12:12] women, it wasn't visible until now. So this is a story of how thief probe, data bricks, and blend, the key players off the pitch, are making that happen.
[12:25] So Mari and I uh are super excited to be here today to talk to you h about women's football and considering the timing with the men's football, it feels
[12:37] uh suspiciously good. Um, uh, as we said, Frankie and Alex are the team from FIFA Pro couldn't be here today, but this is their story and we wanted to make sure that they had their voice in
[12:50] the room today. So, we've got a short video to play to you and then we'll tell you a little bit about FIF Pro.
[13:04] I
[13:18] >> not my sexual
[13:30] futures at risk? I dream
[13:42] become one. Protecting every dream, >> defending, >> demanding, >> compl. >> We are the players. We are the union. >> We are.
[14:02] Great. pretty powerful video. So, FIF Pro are a trade union. They represent 70,000 professional footballers globally across 70 different player associations. And
[14:15] FIF Pro make sure that the players have a a voice at the table around decisions that are affecting their lives and careers. Uh, so without FIFA Pro, the players
[14:27] within that video, household names like Mia Ham, Lucy Bronze, Lionel Messi, Killian Mbappe, they would have had much less power over their careers without an organization like FIF Pro. Uh, they Fif
[14:42] Pro have been driving change for 60 years, but as we've laid out at the top of this talk, there is one area that is definitely still needing progress, and that is women's football.
[14:57] Perfect. So what challenge did Fifro come to us with? At the heart of this project is nearly a decade of women's welfare survey data which has been collected from players all across the world. So the data comes from two main survey types. So we've got the global survey which is completed from by
[15:10] players at all levels and it gathers their perspectives and experiences in the women's game. And then we've got the confederation survey. So that focuses on post tournament experiences. Um so think
[15:22] of maybe more like your national teams. Um so that kind of higher level how they found the experiences and any um operational feedback. So FIF Pro have done a fantastic job of collecting all this feedback over the years. But the
[15:35] challenge was that it lived on multiple files, multiple laptops. We've got two different survey structures and as time's gone on these survey structures have evolved and questions have changed. So we asked ourselves a simple question.
[15:49] How do we make all of this data accessible to the team at FEPro in a way that's trusted, scalable, and doesn't require them to be a data analyst to use it? And that's where data bricks came in.
[16:01] So why did we choose data bricks? Uh and I'm pretty sure everyone the question here would be why didn't you why wouldn't you have chosen data bricks considering we're at a data bricks conference. Uh so when we started working with FIF Pro and understanding
[16:14] their needs and requirements and designing the solution, it was clear that they were going to need a single platform that was going to be able to do everything from the data preparation, the data quality through to the AI
[16:28] powered analytics and that's what data bricks gave us. So uh with data bricks um with having the unified platform we're able to give everything that we needed in the one
[16:40] place which was then helping us simplify the build the deployment and then through to the maintenance of the solution afterwards. We were always designing this solution that we would be handing it over to FIFO at the end. We weren't going to wasn't something a
[16:53] solution that they were going to be uh locked in to us as a vendor. So that was really important that we designed something that could be easily handed over to them. And we also wanted to make sure that we're thinking about the future. So how do they extend beyond the existing use case? And again that's what
[17:05] data bricks would give us. Perfect. So let's get into what we actually built. The solution comes down to two key layers. And the first one is a data foundation. So we took all of the
[17:17] surveys, all these Excel files that were on these varied laptops and we transformed them into a unified governed model within data bricks. So we took the global ones and the confederation ones and combined them together. And if
[17:30] anyone's worked with survey data before, due to its nature, it's never quite as clean as one would hope. Um, so this involves mapping questions across the two different survey types. And it also involved standardizing the questions as they evolved and resolving any
[17:43] inconsistencies that had built up over the years and any of the responses. So to put it simply, data bricks became the home of all of this previously disconnected information. But organizing the data was only half
[17:55] the challenge. The goal of this project wasn't just to create a trusted data asset. We needed it to be accessible to the people who needed to view this insight and that's where our
[18:07] conversational AI layer came in our data bricks. So once or data bricks sorry data bricks genie um and I mean we're all at the keynote this morning I imagine there was a lot of talk on Genie. So we'll be interested to see
[18:19] what we can do after that. Um but we're all on the same playing field to what Genie is. Um, so as once our data foundation was in place, we introduced that conversational AI experience on top. So the team at FIF Pro who aren't
[18:31] technical, they didn't have to go into their Excel spreadsheets or ask their external data analysts. They could just ask a question in a way that was simple for them. So questions like how do
[18:43] younger players experiences differ to those who are maybe later on in their career? Or what tournament had the highest satisfaction score and more importantly why? or simply what themes are we seeing in recent surveys that
[18:56] weren't there before. The list is endless. And what Genie does is it pulls these answers from our governed model underneath. So there's no bespoke eye build. It's a simple datab bricks native capability doing exactly what it's
[19:09] designed to do because the data underneath is in good enough shape to let it. And that's what I want to be really clear about here in the context of where AI is right now. And I imagine some of the presentations that you guys have been to today. This isn't a
[19:22] technically complex build. We're not fine-tuning models and we're not engineering custom agents. We're simply applying AI in a really valuable use case and that's what makes it valuable.
[19:35] So, let's get a little bit into what we've built. We've got a few demos here that we want to run through. And the point we really want to get across is one, how easy this is for our non-technical stakeholders to ask the questions and two, how easy it is for
[19:49] them to use and understand the responses. So to kick us off here, we've started with how has player salary from clubs changed from 2017 until now, which sounds like a really simple question,
[20:01] but historically it really wasn't. In these exile files, we had salaries which were in local currencies from multiple countries all across the world. And then we also had different response formats. So we had maybe salary range options
[20:14] people had selected or it was just a simple free text input. So for the team at thief pro to make any form of meaningful insight from this, it was virtually impossible. But what our Genie does now, Genie does
[20:27] now is it pulls a unified compatible picture really quickly and it really unlocks that ability for the users to ask a question and get a simple answer back. And that's really what's valuable
[20:39] here is what uh the T thief pro can do is they can then go into conversations with decision makers and they can back up that what they're doing is working. So we can see here that the proportion of players salaries has increased and
[20:54] the proportion of players who've not getting paid have decreased from 49.6% to a small 1.7% and that's a massive massive improvement for the team at FIF Pro over the last nine years.
[21:13] Uh the second demo. So earlier on in the talk, we spoke about women that are leaving the game early because they get injured, not because they're not leaving the game because they get injured. They're leaving the game because when they're injured, they don't get paid. And that is a massive reason to be
[21:26] leaving the um leaving the game when it can be prevented. So uh in this example uh we asked Genie uh what is the change between 2020 2017 and 2026 uh in players
[21:39] responses around whether they do continue to receive a salary uh even when they're injured. Um and the great news is is over that period of time we're seeing that players are reporting they're much more likely to continue
[21:52] getting paid. And the the standout stat for me that I really like is that there's no protection. So from 2020 2017 there's 30.7% that reported they did not get paid when
[22:06] they were injured. Whereas now that number I mean it's still higher than it should be but it's now only 7%. That's huge. That is evidence that the the pressure on the clubs the policy work is
[22:19] having an influence and it's making a change. And now the players are able to concentrate on their recovery and their rehab and getting back out on the pitch and they're not having to worry about where their next paycheck is going to be coming from.
[22:33] And for the final demo just wants to go a little bit beyond the numbers. So what one of the questions that the surveys asked was where do players see themselves in five years in relation to the sport? And historically making sense
[22:47] of these open answered questions, these free text responses was a significant manual undertaking for anyone um in the thief pro team. But now what they can do is they can ask a question and Genie can
[22:59] pull these trends and themes directly from it. And what it pulls here is really quite interesting. It one shows us the proportion of players who do expect to be playing. It also shows us obviously the proportion that don't
[23:12] expect to be playing. but most importantly why these players are expecting to stop. And that's where then FIFO can target what they want to do moving forward.
[23:26] it's not just um surfacing these previous numbers. It's actually transforming the time and the effort of the team and they can then what used to be days, weeks, however long can now
[23:38] just be asked in a simple question. So actually go back. Sorry. Um, so that's the platform in action. Super easy for the team at FIFA Pro who again not technical to ask these questions and
[23:51] get these insights in a much more understandable format. But I want to take a step back because the technology is only part of this story. Ultimately success isn't measured by the solution
[24:03] that we've built here. Success is measured by whether this actually helps the team at FIF Pro improve the life and well-being of these female players around the world.
[24:18] So, as we know, FIFA Pro have done a fantastic job of consistently conducting these surveys over the years and that commitment matters. But collecting data doesn't create change on its own. Many people start with the data and they
[24:31] hope that the value follows. But here at Blend, we like to start with the value. What decisions the FIFA Pro need to be making? What action do they want to take? And ultimately, what does success look like for this team at FIF Pro? And
[24:43] then we build from there. And that's where the real shift happens. When the team at FIFA Pro can move from data to insight, insight to action, and action to measurable value.
[24:55] That's the cycle this solution is designed to power.
[25:08] with their data. So rather than being historic Excel files, it's now a strategic live a asset where one that generates insight and provides action to where they want to go next. And whether
[25:20] that's for a certain demographic in the world or whether that's for the sport as a whole. And that's where the most important shift happens, that reactive to proactive insights. It changes the
[25:32] questions that the team at Thief Pro can ask from what are players telling us today to is what we're doing actually working and what do we need to do next to make a difference in these players lives.
[25:49] Ahead of this session, we were chatting to Frankie and Alex about what's that insight that this platform has surfaced that you couldn't quite see as clearly before. And they gave a really nice example which I wanted to share with you all today. Alice has touched on it a wee bit so it gave you a little bit of a spoiler. Um but back in earlier survey
[26:03] cycles so 2017 the key reason that players gave for leaving the game early was maternity concerns and the lack of adequate protections around that. In
[26:15] maternity regulations were introduced into players contracts for the very first time which is quite crazy when you think about it. But FIFO's legal team were pivotal in pushing FIFA to do this. And now when you look at more recent
[26:28] survey datas from our global 2026 survey, paternity concerns is no longer a key concern. And that's not a coincidence. That's evidence that advocacy baked based based
[26:41] in real player data can drive real and measurable change. But the data now tells us something else. Maternity concerns may not be the key reason, but we can see financial reasons are now a key reason that
[26:53] players are leaving the game early. So, the focus needs to shift. The work isn't done. It's just pointing in a new direction. And that's ultimately what makes this so valuable. It's been able to see where
[27:07] progress has been made, what issues matter matter most to the P the players, and where it's still urgently needed. This data set has always had potential to drive real world change. But the key
[27:19] difference now is the potential is accessible to those who need it most.
[27:31] Great. So what next? FIFAP Pro set us the brief to unlock the data and drive change. And hopefully over the last half an hour or so we've demonstrated how we've been able to do that. Uh but now our attention turns to what next? Now
[27:43] that they have the platform, have the data, it unlocks a whole host of other potential use cases for the future. This could be anything. We've outlined a couple of here. It could be using player
[27:56] data to predict injuries before they happen. It could be using salary data to build pay equity tools so that when players are negotiating their contracts, they've got the data to help them make sure that they're getting the the
[28:08] fairest deal. These are no longer distant distant ambitions. This is stuff that FIFA Pro could start doing tomorrow. But what we want to leave you with today,
[28:20] we opened the talk with our fun quiz. I hope you all enjoyed that. H which highlighted the invisibility gap in women's football. Mar spoke about the women's football team in Scotland that qualified for the World Cup and then
[28:32] kind of everybody at least in Scotland forgot. Uh, and we spoke about the prize pot for the women's World Cup, which was a third of the prize pot for the men's. Again, remember, it's the same sport, same tournament, players doing the exact
[28:44] same job, but a third of the prize pot. But now, armed with the data and the tools and the technology that they need, Frankie and Alex and the team at the team at FIF Pro, they're now targeting
[28:56] pay equality for the 2027 Women's World Cup next year. So when that happens, thank you.
[29:08] room today who've joined us will know that that was possible because of the players that have shared their data, because of the team at FIF Pro who have pushed hard and driven for these changes, and due to the technology that
[29:20] data bricks has allowed us to have. And that's what happens when you put the right technology in the hands of people that are driving change. Thank you very much for your time this afternoon. >>
[29:37] >> We were very much on time, so we have 10 minutes if anybody would like to ask any questions. Thank you very much everyone and we appreciate the questions.

Learn more about the Databricks Data and AI platform.

The information provided herein is for general informational purposes only and may not reflect the most current product capabilities or configurations.