Forget data cleansing, start data cleaning
Are you faced with asset data chaos at the end of each financial year? Struggling to prepare for valuations with incomplete or inaccurate data? View this webinar to learn:
- Why good data is so important and the typical data issues faced by local governments
- How to stop data cleansing and start data cleaning: throw out the huge data cleansing project and start daily cleaning
- Five key steps for cleaner asset data that are not time consuming or costly
- How to utilise what you already have for success: use existing dashboard tools and reports to alert you to data issues as they arise, making it easier to keep asset data up-to-date and accurate
Welcome to our webinar this afternoon. For those of you who have never met me, my name is Robin. I'm a senior services consultant based in New South Wales here at Riley. I've been in asset management for about 18 and a half years and I'm primarily focused on the confirm system, but I do also work with ascetic on occasion.
With me is Rebecca Harris, who's also a senior services consultant with over a decade of asset management experience. Becky focuses on them in Victoria, but like me, she's also a woman of many systems and works with our predicted solution when the workload demands it. I think there's a slight lag with the slide, so I'm just going to give that a couple of seconds.
Okay. So for this webinar, we're going to focus on how using data cleaning principles in your asset management system with a longer view of the data and its potential issues, can offer an overall improvement to the quality and completeness of the data and save you valuable time, money and resources when it comes to utilizing that data for larger projects.
We're going to do this by focusing on the issues that you're probably already aware of, some of which you might already have halfway solved, you know, temporarily. So. So if you happen to be a concern or an esthetic user, all of this can be achieved using the tools and resources and functionality that you already have and ideally will be leaving you with plenty of tips and solid methods on creating a less stressed, simpler and more proactive approach to data quality.
So I'm going to start by talking about some of the typical issues that are faced by local governments. This felt a bit like Family Feud when we were putting this information together. You know, we surveyed 100 people and the typical data issues are we didn't survey 100 people. What we did is go through all of the problems we've helped sold over the last few years when our customers are performing valuations and the financial year exercises, carrying out that kind of long term planning and doing some of their larger asset projects and the sorts of activities that cycle around every year.
And we came up with six of the most common issues that are faced by local governments at the moment. So the first one I'm going to talk about is incomplete data. So incomplete data is most often caused when the asset data has been entered into the system because we know there is an asset there. So we need to put it in there.
We know the entry is valid and necessary, but we don't know all of the correct details about it. By the time we notice it's been too long and we don't know what we can't remember to ask. All of the immediate knowledge about that asset has gone stale. And then by the time we notice that it's incomplete, we've got a big project on the go.
It's become urgent and we're running out of time to find those details. The second thing that quite often goes hand in hand with incomplete data is mismatched data. So this example encompasses things like finding a gravel road asset with a material type of asphalt, for example, or perhaps a light coat with a gate type that that particular type of light pole would never support.
They just wouldn't be able to go together. Mismatched data causes mismatched actions, you know. Is the light pole correct? Is the gear type correct? We don't know which one is right. We possibly don't know what sort of maintenance should apply to the asset. It can get caught up in the wrong kind of report. It can get caught up in the wrong kind of inspection.
That then builds problems out of a problem into our long term planning. If we're using data extracts to run through a planning system or any sort of homegrown algorithm, we could potentially end up with skewed results. And then that has knock on effects to our budgeting as well. So that can be quite, quite a serious issue. The next thing I want to talk about is out of date data.
This is where, you know, the data is complete and we know it was accurate once, but now it's still an out of date and we don't really have confidence in it. This can be the result of data collection exercises that were carried out for specific project data collection, you know, takes a long time. And then as soon as it's been used, we move on to the next big project and it sits and goes stale.
There's no upkeep carried out because we've moved on to that next big thing and we're busy, right? So the data is used once and then left to go to waste and the next time we want to do anything with those assets, the data is out to date again and we need to repeat the whole data collection process, which again is expensive, as I've already said, takes ages.
Then we use it, then it's left to go around and set about and go stale and around and around. We go in that cycle over and over again. So the next thing I'm going to talk about is overwhelming issues. The problems here aren't massive, but they are just so many of them. This is actually the biggest issue that was told to us by customers that was given to us to help deal with and the problems that we have with the data here tend not to be big showstopping ones, but there are just so many of them and it becomes overwhelming.
We can't categorize the issues properly. We can't sort them from each other. And we have so many small little things to deal with. That prioritization then becomes too hard. We end up with a massive backlog that we can't keep track of what individually is showstopping groups together and becomes a huge block and completely showstopping. And now, however many of those problems we solve, we then have the problem of Re-occurrence to deal with.
This is where we've identified issues. We fix them, but things fall through the cracks, right? So they reoccur. And the thing with this is it starts very small. It starts sneaking in over time and then all of a sudden you realize you've got a big reoccurrence of the same issue without anyone noticing until it becomes a really big problem, that's going to stop your project from moving forward.
If we've already solved this problem once, what we end up doing is a lot of time, effort and money solving it a second or sometimes even a third time. Now the last thing I'm going to talk about here is data cleansing and collection. I've put these two together because they so often do go hand in hand. Some people might look at the first five issues here and say, well, number six is actually a solution to those problems.
Well, cleansing and collection can also be an issue on its own. You know, money is a focus the local government everyday, but we don't often like to talk about money. But if we're using cleansing and collection as solutions to any of the first five issues, more often than not, that involves spending a lot of money. Data collection is really expensive and we have to balance those budgets every single year.
And it also takes a long time. So it's expensive in resources and time as well. If we're having fresh data collected before a large project and sometimes before small ones too, usually a large amount of data cleansing has to take place. So we then do a data cleansing exercise. We give the data collection people the extract of our assets.
When we're finished, they collect them. We can finally get on with the long term planning project or whatever work we had planned data cleansing in this case becomes a project in its own right, and then the collection project takes place. The whole thing time, money, delay, the start of planned activity. So where we thought we had one project, we end up with three or four all stacked up and taking time and effort that we couldn't really afford.
Now all the while that's going along, long neglected day to day workload is just sitting there because we're having to concentrate on all of those other things. Our day to day tasks have to be put to one side and just pile up while we're concentrating on the cleansing and using even more of the issues that we've already talked about.
Two to stack up and cycle around. So before I carry on, I've just got a little poll question here, so if everybody could grab that that mouse before we move on and see if you can tell me how much confidence you have in your current data, you should be able to just click on that. Now it's up on the screen.
I've got my time around. I'm going to give everyone about 45 seconds to answer that question. So you've got a couple more a couple of more seconds if you need it. Okay. So I'm going to move on to the next one. Now, let's have a look so we can see that there's quite a bit of confidence there, which is really nice.
But there are a few people out there who don't have a huge amount of confidence in their data. So what we're going to talk about then is, well, all right, I've said I don't have much confidence in my data. What do we do about that? We've looked at the problems. How do we solve it? Well, the way we're going to solve this today is by taking a long term view of our data and to do that, we need to stop data cleansing and we need to start data cleaning.
And that's my main topic of today. The huge difference really between cleansing and cleaning is that we're data cleansing. We have a specific project for a specific purpose. You know, I am cleansing this limited asset set because I have a specific project. You take that targeted focus on a very small area of your asset register, but for the whole of those assets, it's very short term and it comes with all of the issues we've just been talking about.
It cost time, it costs effort, it causes day to day workloads to be neglected. It's great for 5 minutes and then the data goes stale, especially if no one is keeping an eye on the new assets that are coming into the system. After the project is completed. With data cleaning, you swap those middle two points around. So we take a whole of asset register view, but we think about the specific issues that we are having and we do it long term for just a few minutes every day, not just when we have a specific project coming up.
We do it all of the time and the benefits of that are that when we get that specific project, either, you know, one of those special things that comes along that we have to do or an end of financial year or a valuation situation, so on and so forth, the kind of activities that cycle around every year we end up doing less time and effort to clean up any issues that we sign because we've been quietly and consistently chipping away at the entire time.
So how can we achieve this? So I've put here we've done this with a couple of customers now and we've distilled it down into five steps, five steps to set everything up. And then once you're up and running, it's 5 minutes a day to carry out. Okay, Now I know we've all been promised a flatter stomach and brilliant abs in 5 minutes a day and we bought that thing go off the TV and at best it went rusty in the garage, potentially after being used as a plate source for a while.
But I'm not that woman who can't open a simple milk carton. And Becky is not that woman who can't just put slightly less food in the plastic container. We're not selling anything here. So do you remember when I said at the beginning, if you've got one of our asset management solutions, you've already got the tools and the functionality to make this happen without spending any money or that you need from me is a couple of ideas and a few tips to help make it work.
And then a bit of a demo from Becky to see how it actually can work in a real sense. So this slide step, that's all we need. If you follow the five, you're going to end up with a fair amount of data freedom in this. So next, talk about what the five steps are. So first things first, get everyone on board.
Ideally, this is a group situation for both the identification of the issues that you're going to solve and for the solutions that you're going to come up with. If you don't have everyone involved, then you're not going to get a good outcome. This is simple change management principles, really. Remember, we're talking about following five steps to set this up and then only 5 minutes a day to carry it out.
Then much further down the line, it's going to save people hours and hours of pain because we're not going to do all of the solutions flat out in the two weeks before the end of the financial year. We're going to do this every day for the ten months before the end of the financial year, for example. So it's much, much easier.
So get everyone on board, sell that idea and get their input. The other side of this is that in the beginning they're going to feel ownership. And when it comes to spending that 5 minutes a day, they're going to be much more likely to make it part of their routine every day if they've been involved right from the get go.
The second one is probably the most important one. And I think if you came to illuminate last year, 2022, you'll have heard Becky speak about this, still trying to make it perfect. Perfect data is the enemy of good data. We have to stop aiming for perfect data because it's getting in the way of us having good data and it's getting in the way of us getting things done.
It's a bit like, you know, when I try and do my hair in the morning, if I strive to protect perfection, I'm going to spend the entire day feeling very disappointed and I'm happy and I'll never be satisfied with the result. At a certain point you have to say, Good enough, let's move on. Good data is really important when you're looking for good results.
And the only way we can get good data is to let go of the idea of perfect. Okay, now on to more practical things. It's time to make the issue list and give each issue a priority. So this is step three. If we've managed to let go of the idea of perfect, what we then have to decide are what the issues are and what the priority of those issues should be in order to make your data good.
The way we did this, for most of the customers we've worked with on this so far is to go back to our last end of financial year, our last set of valuations, our last big data cleansing project and make a list of all of the issues we had with the data. So this might be a really, really long list.
It might be a really short list. It doesn't matter if there was a problem in the data, if there was something there, just write it down on the list. Make a list of every single problem you can remember from those last few projects that involved your asset data and then prioritize it. This is the bit where having everyone involved is important because what I might think is important and what holds me back every day might not be the same thing that holds Becky back every day and that she uses important causes her pain.
So have a big get together with biscuits that's really important. And prioritize the list. Sometimes we have to think about prioritizing what's important to us as an organization and to us for the specific project we know might be coming up in 18 months time. It's not just for us. We have to make sure we include the needs of stakeholders above and below us.
When you choose your priorities, they're not set in stone. Make them for now priorities. You might aim to solve the top five Big four now issues and they can change and be added to over time. And I'm going to come to that in a short while. But step four is I have to now decide how we're going to report the assets that are affected by each issue.
So I have to make decisions about whether each problem is better reported on the dashboard or reported on an automated email or a data extract to a network folder and so on and so forth. Different data cleaning problems might need to go to different people within the organization. So I might decide that assets with a missing foot path length should go to Becky, and she prefers the dashboard.
But I'm going to tackle the assets with mismatched light poles and gearing, and I prefer a straight data dump to a network drive so I can get stuck in. So you might end up with several different reporting methods that send out several different issues to different people. For the non system. Uses a little PDF in their inbox each morning listing the three assets that I need some extra information about might be best for them.
So be prepared to take responsibility for populating the system with data for those non system uses. It's all about the system you've got Being versatile enough to let everyone work, how they're comfortable and confirm and ascetic are absolutely best of breed at that. I do say that the dashboard as an overall master control I suggest you absolutely make a dashboard is your overall master control because it's so simple to use.
It delivers everything straight to you and it's very visual. I can very quickly see if I left a dashboard, if the issues are building up again or if we are solving them, if they're if they're moving downwards. Step five is absolutely critical. Automate it, automate everything. This will ensure that nothing falls through the cracks. The dashboards in our two systems confirm esthetic are already automatically refreshing on the timetable that we choose, but every report, every day to extract every email that are missing footpath links needs to be automated so that you can just set it up and then concentrate on finding the answers you will.
5 minutes a day is all about finding those answers. Then also stick a priority one untouchable appointment in your diary every day for your 5 minutes and automate that as well. So that's it. That's our whole five steps. That's what you need. I also, though, have secret step number six, because this is a cycle. We have to make sure that once we've set this up, we do put aside a little bit of time to shine it design and refine it.
If you identify extra aspects of investigation, once you're up and running, just add them in, just add them in and then spend 5 minutes a day on them. Now, the key takeaway to all of this that I keep harping on about and I will harp on about it forever, is that with confirming esthetic, you already have all of the tools that you need to achieve this.
This is also true with other systems as well. If you are not a confirm or esthetic customer yet and you've got an asset management system and something like cowboy or even good old Crystal Report, then you also have all of the tools you need to do this. This is all about don't reach for the shiny new toy that promises to solve all of your problems.
This is about use the resources around you to their fullest extent. So before we move on, I've got another poll question. Who amongst us already utilizes dashboards and custom custom reporting to monitor your data So you might use it to monitor jobs, you might use it to monitor your inquiries, you might use it to already monitor some of your data quality.
So setting a little timer, 45 seconds, again. Okay, So let's have a look. So half of you already that's outstanding. So half of you are already utilizing some of the tools that you can then adapt to do all of the things that I'm suggesting here. That's brilliant. So this is the esthetic dashboard. So for those of you who aren't using dashboards at the moment and are assessing users, this is the esthetic dashboard right here.
Both systems do include, as I've said before, configurable dashboards and reports is very, very easy to start identifying some of the issues that we've talked about today and the kinds of issues that you would be putting on your list. You can see the esthetic dashboard has got a range of ways to display the data. We've got pie charts, we've got bar charts, we've got data extracts there, and it's very configurable to me.
So I can see just the stuff that I'm interested in because this is my dashboard. Okay? And when I'm ready, I can start drilling in to this dashboard here to the data on this one. And this is just a simple report of pipes that don't have a length. So I might send a couple of emails right now to go and find out the length of the pipes.
And then what I'm going to do is I'm going to come and drill down even further. I'm going to fill that in. Kristian gave me these screenshots and he highlighted for me the bass. So it's interesting. It's and when I'm ready, I've filled those in. And then the thing is, now that I set will disappear from that data extract, it's no longer a pipe with a missing link.
So with the dashboard, this is why I advocate for the dashboard so much. I'm only ever looking at the data that I actually need to react to. Along with this, we have some report templates that we can set up. This is quite an old confirm reporting template. Okay. And this is one that tells me I've got features that haven't been inspected properly.
This has been created using the confirm report writing tool called data Miner confirm reports can be configured in data mining and then schedule using the report scheduling. So again, just like a dashboard, I can automate this to be sent out at required intervals. So daily, weekly, whatever I want to a distribution list, or I can have this as a data dump, export it on to a network, drive to be access, and then you can easily adapt this for a data cleansing purpose Instead of features that I'm inspected in an allocated month, I can have my, you know, my PSB of all of the pipes that don't have an appropriate length, it's very, very easy to
set off is very, very easy to start using. So what I'm going to do is I'm going to stop talking and I'm going to hand over to Becky and she's going to give us a bit of a live demonstration of how concerned can be used in a dashboard driven way to clean your data.
Thank you very much, Robin. So stand by for a very smooth transition. So that's a country I was like, Oh, there we go. Brilliant. Okay. Yeah. So thanks for that. Robin. I'm as Robin has already said, you already have all the tools you need to kick start this process. So what I'm noticing is a simple dashboard that can be used to help me monitor my data is essentially a digital to do list, and that will update in real time as I work through it.
So I've chosen to keep this as a completely separate dashboard as I don't want to be looking at it when I'm complete my day to day work. It would just be like a dark cloud hanging of my head, waiting for me to action it. I want to be able to drop in and out of this when I have a spare 5 minutes for the dashboard I've got set up monitors, several different asset classes and look specifically for my problem areas.
As I mentioned, if you already know where the funds lie within your data, then that's a good first step or it's a good third step. When you're creating a list of problems and priorities and you can then choose Prisma information on an easy to use dashboard, much like this one. You can also make that dashboard available to any colleagues who may be asset owners.
And you can, of course, work through that data together. You'll notice my dashboard is specifically looking at where I'm missing information regarding valuation data. So I've got several widgets on here and they're looking for things like assets that have never been valued or assets that have value. This year, assets have no active from or active to date assets that have exceeded their useful life in the system, assets of no category.
And of course, assets have no replacement cost or valuation quantity. All of that missing data will cause you problems. When I come to undertake my end of financial year valuations exercises, you of course, can set up your dashboard to show anything you need. For example, you may have an asset class that requires a specific attribute, so on your dashboard you will show all assets that are missing a value for that view.
And of course, as we mentioned, while mismatch data is about on a dashboard to show you anything you need to validate uncheck, you can display any kind of any sort of whom you want. So the charts and things like that, I've got mine on, so I make sure bar charts, column graphs and of course a couple of pie charts to make it look pretty.
And you can also, of course, present that information on a map. If you say cheese. So I'm going to zoom into my data, hence I have it closer so you can see a bit better. So as you can see, I have a large number of assets with no active from date, so they are displayed here in my columns.
Mine are separated out by asset groups so I can click into a column to drill down to the data behind this. So it's now showing me I've got four records that are missing actually active from date within the Bridge feature group. And so I can overlap all of those records at once, which will allow me to navigate through them using the navigation buttons at the top and go flick through.
And so now, depending on how I have my data set up when I first went live, my system, I may have utilize the assets start date and that could be my year of construction. I could have also record that information as an attribute. There's some construction data information in there. So there is data. I've already got it in the system so I can click onto my valuation type.
I'm going to see that my start date is my construction. I verified that's true. So all I'm going to do is use that start date and populate my active from date down here. So there is type of and on your notice for as soon as I move out of that field, it has also automatically populated my active to date.
So that is of course based on background rules in the system. Looking at the useful life for my asset category for my bridge here and I can save that and move on. So I'm essentially killing two birds with one stone here. I'm populating two missing fields of data with one action. And of course I can then move on to the next record.
I in the next day. And it's really that quick and simple says I go here and in that short space of time I've now updated four records. So you see how quick that was to me. And there we go. So a populated for records all at once and as I said, as I go through, you'll notice that they're very quick to do.
I also try not to get too stuck into it, though, as it can become weirdly addictive. And you could lose more time in here than you want to. They set that timer 5 minutes a day is really good as a starting point. And so once I showed in that set, I can close out here and close on this.
And so now I can choose to open up another section here, start populating those as well. Or of course, if I finish my asset updating exercise for the day, then move on to my day to day tasks. Alternately, I could also use some of the compounds batch updating methods to update this data. That's what I was saying. We can put it in a data dump.
I can also select this data out the system here and say that as a spreadsheet using the copy button or the Save saved button and I can use that and work for that data in a spreadsheet format if that's something I prefer, and then import the information that I can. So as you can see, data cleaning can be a really quick and easy task that we complete in small batches.
And that means that over time our asset data will gradually improve and we'll be in a better position when we actually want to use it. So you won't have that kind of big panic and quick sprint towards an unobtainable deadline. And when we're trying to update all of our data at once, we can work through this methodically and we have clean data for when we need to use it.
These dashboards are also available in the Web format, so that means that you can share them with upper management really easily and send me a link and say, Hey, look, look how good I'm doing it. Clean the data. That's how clean our data is. And I'm sure that we really impressed with your proactive this OSI. Some other data will require more intervention and potentially some data gathering as well.
But at least this way you will have visibility of that as you manage it within your tasks accordingly. So as long as you have visibility, you know it's there to do, it's not going to get lost in that big pile of, oh yeah, I'll get around to at some point. So yeah, too easy. Right now we go back to, you know, some sharing.
Thanks back. That was awesome. And so you can see that the confirm can really be used in a dashboard driven way asset. It can be used in the same dashboard driven way to really start making a dent in some of those issues. Well ahead of time, where they're going to actually make a huge dent in your project. So I'm just going to sum up real quickly.
You know, it's five steps to setting this up and then 5 minutes a day. So go and get everyone on board, talk them into letting go of perfection. It's a useless concept anyway. It's like when people say, that's not normal. Normal is use this concept, make your list and prioritize your issues. Decide the best method of delivery for each one of those issues.
Don't just execute the plan. Automate the plan. And then don't forget secret, super secret special step number six, which is when you finished, make sure you refine it shiny and redesign it because this is a cycle. Okay? It's a cycle. And the most important thing to remember about today is that you already have all of the tools you need to achieve this.
So with that in mind, can I just ask again who intends to put some of these tips into action? Okay, I'm just going to move this slide on. That's amazing. 96% of you think that Becky and I have not audited today. I'm really impressed. Thank you so much for that. That's amazing. So with that, then, does anybody have any questions that they would like to ask?
We've had a few dropped into the chat already. Okay, hang on. So can I know more about the tools you're using? Is it part of. Rightly so. Thank you, Matt, for that question. That's a really good question. Yes. So if you a confirm user or if you're using a static, then yes, you have the tools already. They are the standard dashboards and report writing tools that are part of your software.
And if you need to know a bit more about them, if you don't know where to find them, get in touch with us. Get in touch with your account manager, book some time with one of us and we can talk you through finding those without any problems at all. Yes, We're quite happy for you to have a copy of the presentation.
That's absolutely fine. No worries to.
And as a question about using a discrepancy data source. So yes, if you have confirmed, you can use a discrepancy source. It works in a similar way to the dashboard with a validation process. The idea being you have a set of rules set up against your asset group and then you validate your asset against those rules and it gets the time and dates that they this asset has been validated and matches the rules at this time are dated by this individual.
So yeah, that's those are really good idea of making sure your other data is completely updated. Correct.
I think I've got a question from Jeff out here. Are we aware of any I type tools to clean some of the issues rather than manual methods? So at the moment, yes, we are aware of I tools that can do this sort of thing. There are currently non included in a static or confirm systems, but hey, get on the suggestion portal because I'm absolutely up for that.
That sounds brilliant at the moment to use the tools that we've been talking about here. It would involve a manual update either via manually populating the data, importing data, or using some of the batch update tools in both systems.
So question from Scott here. So is there a logical way to capture information about a failed, i.e. the age or construction date of the asset as estimated calculated versus actual? And say, Scott, that goes down to how confident are you in your data and how it collected originally. Obviously if your record keeping. Yes. Has not been completely accurate, it is going to involve some manual exercise of data gathering to ensure that your data in the system is correct.
A logical way going forward would be to have a great asset handover process on capture process to make sure you are so keeping track of your news on your updates and your maintenance and things like that and yeah.
Okay, I'm going to answer two questions at once here. Is it possible for Bradley to show us how to create dashboards? Is that kind of training other than on my knowledge page and all the standard suite of admin cleaning dashboards available that Becky was demonstrating? So I'm going to ask them both together. Yes. For both systems, we can do reporting and dashboard training.
That's absolutely no problem at all. Get in touch with your account manager and put yourself in with one of us. We can do some standardized reporting and dashboard training or we can do some specialized towards what we've been talking about today, or what your personal needs off the dashboards and reporting. And the other thing is we don't have a standard dashboards for this that we can give you.
However, the stuff that we've demonstrated today, we can lend it to you for inspiration. And that's probably the best way to put that. Oh, I've missed the question. I'm so sorry, Claire. You've asked if we have a service agreement with Breitling. Would this be covered under that agreement? So your service agreement with broadly the tools that you have are already there.
If you're a consumer or an esthetic user, then you already have dashboards. There included in the software that you have, depending on what it says in your service agreement about providing services or providing training it may or may not be covered. So get in touch with your account manager and have a chat about that. That's my best advice there.
Okay. I think we've about covered of all of the questions there. I'll leave it open just for another minute or so, just to let everybody that hasn't asked a question that anybody that wants to open can put a question there for us or our standard dashboard training for them. A day and hour dashboard training for a I believe is half a day.
It is included in your system admin training. I believe. So you should have hopefully someone on site who's already have the training. So it might be time to seek out who's got that knowledge and make best friends with them. At this point.
So the guys on Google.
You're going to take it.
I was going to say, Scott So can the dashboard reflect data from external sources? The data would have to be put in to confirm to be able to display on the dashboards.
And just a thank you to anybody who's dropped a compliment into the chat there. That's really appreciated. We don't have any more questions popping up. So from here it's just my job to say thank you so much from Becky. And I thank you so much for coming. Everyone. I do hope you enjoyed this session. Please feel free to contact your account manager, Get in contact with Becky or I if you want to discuss this further.
That's absolutely great. We'd love to talk about this with you. And as I say, I do hope you enjoyed the session and I do hope you enjoy the rest of the afternoon. Thank You so much.