Higher education institutions have a large pool of data!

Student records. Staff information. Assessments. Applications. Alumni. Research. Regulatory reporting.

And now, data being used to power AI! - We all know this. But having lots of data isn't the same as having data you can TRUST.

In this episode of the Data in 10 Podcast, we take a look at the data challenges facing higher education and what universities can do about them. The session is hosted by Andi Jarvis, the conversation brings together Ian Tilsed, Director of Digital Technology and Operations at the University of Surrey, alongside Binary10's CEO and COO James Blake, and Steve Smales, COO.

They discussed data quality, integration, data governance, standardisation, security, AI and the challenge of delivering a joined-up digital experience for students and staff.

In this episode, we cover:

• Introduction and current state of Binary10 in Higher Education 00:16

• Specific data types and objects in Higher Education 02:02

• Regulatory reporting requirements 02:48

• University of Surrey scale and transnational education 04:19

• Data standardization across Higher Education 05:55

• Biggest data challenges in Higher Education sector 06:53

• Data quality issues and student experience 08:07

• Future direction and information management improvement 10:36

What makes data in higher education different?

Higher education has a particularly broad data landscape. A university may need to manage data relating to:

• Students and their entire student lifecycle

• Applicants and admissions

• Staff and other users

• Assessments and attainment

• Courses and modules

• Alumni

• Research data and research outputs

• Regulatory reporting

• International and transnational education

• Student and staff services

Some of this data also needs to be retained for extended periods because of regulatory requirements or research grant conditions. And as universities increasingly operate internationally, the picture becomes even more complicated. Different countries can introduce different regulatory requirements, meaning the way data is managed and used cannot always be treated as a UK-only problem.

What are the biggest data challenges facing higher education?

For this, answer you definitely should listen to this podcast! However, I will give you the main challenges here, in case you don't have time.

1. Fragmented data across multiple systems

2. Poor or inconsistent data quality

3. Integration between systems

4. Data governance and ownership

5. AI readiness

6. Regulatory reporting

7. Data security

8. Student experience

9. International operations

So, how can universities improve their data?

There is no single technology that will solve higher education's data challenges. The conversation points towards a more joined-up approach.

🎧 Listen to the episode!

Do you need help for you upcoming data projects?

Whether you're preparing for a data migration, starting a digital transformation programme or simply trying to get a better handle on the data you already have, we can help. From data quality assessment and profiling to cleansing, enrichment and governance, we'll help you understand your data and make a plan to improve it.

Get in touch with our team: CONTACT

Questions answered in this episode:

1. What are the main data challenges in higher education?

2. Why is data quality important in higher education?

3. How does AI affect data quality in universities?

4. How can universities improve data quality?

5. Why is data integration difficult in higher education?

6. What is data governance in higher education?

7. Is university data ready for AI?

EPISODE TRANSCRIPTION:

Disclaimer: This transcript was generated by an AI tool that did its best, but it's never met different British accents it could fully decode. Expect a few funny mistakes. Enjoy!

[00:00] - Andi J

Hello and welcome to the Data in Ten podcast from Binary Ten. Today, I'm joined by Steve Smiles, the Chief Operating Officer; Jamie Blake, Chief Executive; and Ian Tilzad, who is the Director of Digital Technology and Operations at the University of Surrey, based right here in Guildford. I'm Andi Jarvis; I'm the marketing guy here at Binary10. We're on location in this amazing room at the Guildford Institute Library. Steve, Jamie, how's things in Binary10's world at the moment?

[00:29] - James B

It's good. It's good. It's busy, and specifically, obviously, we're here a little bit with Ian to talk about higher education, and and that's where a lot of our efforts are at the minute. You know, a lot of transformations, a lot of upgrades to cloud solutions, and of course, as with everywhere as well, you know, really looking at data quality. You know, as as the AI horizon continues to you know grow on us. But um, but yeah, exciting times.

[00:52] - Andi J

Fascinating sector, HE, isn't it, Steve?

[00:54] - Steve S

Absolutely. I mean, it's it's one we're passionate about. We've done, you know. Quite a few projects

now in higher education. It's something we always enjoy, and but you know it does have its specific

data types, data you know entities, data challenges as well.

[01:07] - Andi J

And timing as well, and rhythm to it. Given the holidays and breaks and things, and when students are recruited, it's it's different to all the sectors.

[01:14] - Steve S

Absolutely, I mean all sectors are different, but yeah, higher education's got its very much its

specifics, and that's obviously something that Ian will be other sort of talk us through.

[01:21] - James B

Yeah, and I was just to add to that, like I just love how this sector. It's so important to our country, so important to the UK for the next generations, and making sure that you know we empower them to go on and innovate and do great things. So it kind of holds a bit extra to you know a bit of extra

responsibility to get this right and to to make sure we do best by them.

[01:39] - Andi J

We are recording this just after the UK has got its I think it's eighty third prime minister in the last six years or something like that. So I know Ian, you won't make any political statement, but hopefully the new government does see the value in the higher education sector because you're absolutely right. It's an economic driver. It's an innovation driver. It's a huge, crucial thing for the future of the country.

[01:58] - Steve S

Absolutely, no, absolutely.

[02:02] - Andi J

After you.

[02:02] - Steve S

Oh, okay. So, I mean, Ian, what are the I mean, what are the specific sort of data types, data objects that you encounter within higher education that may be unique to your sector?

[02:14] - Ian T

Yeah. So, typically, in a university or in higher education, we have the expected stuff like student data, data about staff, but we also have data around assessments and about the the student

lifecycle in terms of applicants and then also alumni. But then that's not. Forget research as well. We have a lot of research data, primary sources, research outputs, all of which need to be retained, often for extensive, extensive periods, either for regulatory reasons or for for for the grant, you know,

conditions.

[02:48] - Steve S

Yeah, I mean, I was just going to say. Oh, you mentioned regulatory. There is there a lot of data that is regulated within higher education sector.

[02:24] - Ian T

Yes, we the universities are generally sort of asked to report to various regulatory bodies on a regular frequency, year in year out. The Higher Education Statistics Agency has a particular onerous sort of task on us, where we have to report on student numbers, completion rates, levels of attainment, and so on. So, yes, there is a huge emphasis on getting data right, so that we can report effectively for those.

[03:20] - Steve S

Yeah, and HEIS. Sorry, I was just going to say on HEIS. I mean, how often do you have to report to HEIS? Is that an annual thing, or is it something that's that's done on a more regular basis?

[03:28] - Ian T

It's it's an annual thing at the moment, but it's changing soon. So it's going to become a continuous reporting regime. So the emphasis on getting the data right first time is now moving to more cases of getting it right all the time.

[03:41] - James B

That's just even to me. That's just like you know that we think of that word trust with our data, don't we? You know that you can trust it that you don't have to further manipulate it to ensure that when you're sending it off to regulatory bodies or other that you know what you're giving them is accurate and true. You know I think of other sectors. You know like retail. You know, you have that pressure of data in terms of related to stock, and the fact that if you don't get it right, if you don't transition, that can have such a crippling effect to the business. But we almost need to think of that equally. You know, in higher education, that if that data isn't right, if it's not trusted, then that that can just sort of snowball, lead to so many issues. You know, of distrust and you know failure in ongoing processes. So yeah, no, that's a really good point.

[04:19] - Andi J

Do you want to, Ian, give us a sense of size and scale of the University of Surrey in terms of? How many students you have, how many people there are, but also what level of data you need to keep for how long? Because I think back to my university days, which were a year or two ago. Yeah, well, it started in the last century, sadly. Back when TVs were black and white. A little bit after that, but you need to keep data for a long time, don't you?

[04:45] - Ian T

Yes, so the University of Surrey is about sixteen thousand students, and I think is about four thousand staff. But I think we're currently managing about thirty thousand users in total, because we have other people that engage with us, like honorary members of staff and so on. We're also substantially based in Guildford. We're beginning to, like many universities, go out abroad to develop and have campuses in other places. As we as Less students can come into the country. We're going out to them now to go to where they are and delivering services to them instead. So, what's called transnational education is a particular challenge across the sector at the moment as well.

[05:26] - Andi J

And that must create more challenges for data in terms of do you have different laws and regulations in different countries as well to to you know abide by?

[05:35] - Ian T

No, absolutely. I mean, we obviously have our retention policies and regulations in this country. But then we also have different regulatory regimes in in other countries that we're working with, and only this week we've had to adjust the provision of AI to a particular set of students because of their location, based on the regulatory requirements in that country.

[05:55] - James B

Wow. And and I I see a lot as well, and I know it's a big push in central government around data standardisation. You know, all different departments, but all trying to sing off the same hymn sheet. You know, the same datasets, the same reference information. Is that the same within higher education? Do you have like a central Body that kind of governs the universities to use certain, you know, reference datasets, and does that expand wider than the UK as well?

[06:19] - Ian T

Yes, obviously we're heavily mandated by the Office for Students and HEIS, so we we have definitely got standards that we need to comply with in terms of the data submissions we make, and there are definitions for things like what constitutes a course or a module, for instance, to be so that universities understand what they've got to deliver. Yeah, I think there's Definite, clear standards we need to work to, but when we come to integrating different systems, then that kind of falls away a bit

because then it becomes supplier-based and and less standardised.

[06:52] - James B

Yes, got you.

[06:53] - Steve S

Okay, what would you say the the biggest sort of challenges you're facing in terms of data within the higher education sector?

[07:00] - Ian T

I think there's a there's a couple of challenges. I think one is that historically universities have not felt a reason to manage data properly until recently, and I think. The advent of AI and more sort of stringent reporting has really now driven that culture. But I think we could have invested in this a lot earlier, and now we're reaping the benefits of that lack of attention. I think AI is obviously putting a lot of light on our data quality. So many people are now beginning to worry about what that will show

in terms of our data quality, and and also can identify data sources we didn't even know we still had. So that that's a challenge, and I think. More generally, protecting the security, the data, and the security component is a big challenge because AI will expose data in a way that we don't want it to. So we need to have a good approach in terms of security as well.

[07:53] - Steve S

Yeah.

[07:54] - Andi J

The AI point's really crucial, and I think because I know Steve and I talk about this all the time. With companies think, great, we've got data, we'll just launch AI. But it doesn't work like that because if the data quality is not right, if you can't access it, it just causes more.

[08:07] - Steve S

Just amplifies it absolutely. And I mean, what would you say in the biggest sort of data quality issues you've sort of come across in the sector?

[08:15] - Ian T

I think it's it's within the university. I think standardising the data, making sure that we have good reference datasets and not read sort of duplicating datasets, is one. I think when it comes down to moving data between systems, having a very sort of pattern based approach to integrations, I know. That the integration is generally is a major pain area for the sector, and I do also just making sure that we have the right data in the right place at the right time and incorrect, you know, correctly protected, so that we can really leverage the value of AI.

[08:47] - Steve S

Okay, so it's getting that sort of common data model. Then is is the you know move towards that is going to solve a lot of the problems?

[08:53] - Andi J

I know in our in our work in the sector as well, we've heard from people who are student facing who say that the students feel as a challenge with data as well, where things like Name changes or address changes they might change it with one part of the university, but find three other parts of the university haven't updated that. Is that something you see as well in the sector at the University of Surrey?

[09:14] - Ian T

Yeah, I think it's absolutely right. It's a challenge for the sector. It'sI've seen cases where there might be a change in a module for a student, and it takes 24 hours for it to percolate through the systems. So we're always looking at things like that, always making sure that we can move data quickly. Because ultimately, we're delivering a service to the students and to the staff, and they're quite rightly demanding an Amazon-like experience. Although we may not have the full money that Amazon's got to try and deliver that.

[09:40] - Andi J

That was my next question: Do you have an Amazon-like budget?

[09:43] - James B

Well, that's what I was going to say. I mean, you know, everything you know, God, it resonates with me. Everything you say, and and it's common across multiple sectors as well. But I guess, how do we grasp this challenge? You know, is anything being done now? You know, to turn this around to improve data quality and and sort of how you know we think of. Yes, you know there's tools you can use. There's technical techniques you can use to improve data quality, but governance and other things that need to be done, you know, is kind of the culture of the organisation to improve data quality. And then, of course, yes, you know, funding it all has a cost associated. So, are you starting to see any of this traction within the University of Surrey? You know, is are these conversations now becoming more serious? No, I think those conversations have floated for the last 10 years, but they've always fallen down the priority order, and therefore. Very rarely do get picked up unless you're a big, you know, sort of organisation. So yeah, what do you think? You know, is can you start to see the change coming, or do we still have a lot more to do to?

[10:36] - Ian T

No, I think the change is coming, and as you said, I think it's not really been done in the last ten years because there are other more sexy things to do. I like that word. You're right.

[10:45] - James B

I hear that a lot.

[10:46] - Ian T

But yeah, we're now we're now starting an information management improvement program. We're starting to to build out the governance and and the ownership by the business of the data sets. And we, you're right. It's about not just the data, but it's about ownership. It's about processes and procedures and real sort of an enterprise approach to data.

[11:04] - Andi J

Well, gentlemen, the ten-minute klaxon is going off in my ear, so I'm going to have to bring the data intent podcast to a close. So, Stephen Smiles, James Blake, and Till said, "Thank you for joining me, and see you next time..

[11:17] - Steve S

Yeah.

[11:17] - Ian T

Thank you.

FAQs:

1. What is data quality?

   
   

   Data quality refers to how accurate, complete, consistent and reliable data is. High-quality data helps organisations make better decisions, improve reporting and gain greater value from technologies such as artificial intelligence (AI).

   
   

2. What is data governance?

   
   

   Data governance is the framework of policies, processes and responsibilities used to manage data across an organisation. It helps ensure data is accurate, secure, compliant and used consistently by different teams.

   
   

3. How do organisations prepare their data for AI?

   
   

   Preparing data for AI typically involves improving data quality, removing duplicate records, standardising information, strengthening governance and ensuring data is secure and accessible. Organisations with trusted, well-managed data are generally better positioned to adopt AI successfully.