Thanks to research performed by Duedil (https://www.duedil.com) we learned this week that there were 5.5% fewer business startups in the UK in 2017 compared with 2016. This may not sound like too much of a drop however it signals the end of a steady rise in startups that began with the financial crash in 2008.
The important piece of information however is that, despite the drop, there were still almost 650,000 startups last year, down from 686,000 in 2016.
This is an extraordinary total. It indicates that, over the last few years, millions of people have opted for a living based upon their wits, determination, courage, and sheer hard work.
For most it’s not a bed of roses. The average working week is about 40 hours. And average profit is just under 16k. In other words, take home pay is slightly less than that earned through the minimum wage.
Sadly, the great majority of these business will not last more than a few years. The obvious reasons relate to a tightening in consumer spending and uncertainties around Brexit.
This is despite a myriad of schemes available that provide business and technical support as well as low cost startup loans.
But here lies the critical problem for many of these fledgling businesses. There are plenty of people willing and able to ‘push’ them along – providing free advice, financial and legal services, office space and office facilities, as well as a range of loans and grants.
But there is a severe lack of what is perhaps the most important element of all – information that will ‘pull’ people into business e.g. about their potential customers, the people out there who just might have a need for whatever the business is selling.
Imagine the famous Mrs Miggins Pie and Coffee Shop starting to trade today (as featured in the Blackadder saga). Thanks to being a startup she gets a discount on her rent and can therefore get a decent location on the high street. And thanks to a mix of grants and loans she can get all the capital equipment she needs.
But Costas is only a few doors away, so it’s going to be tough to get going. She might soon find herself in business, but without many regular customers to her name, and already in debt.
What a difference it would make if, in addition to receiving the push that she needs, Mrs Miggins can also benefit from the ‘pull’.
E.g. data about the people who are likely to be close by at some point in the day but who are unlikely to make it into Costas. Data that also suggests when in the day those people might welcome one of Mrs Miggins’ coffees and a pie. And data that suggests the best way for them to take delivery.
Suddenly she can envisage a business that doesn’t have to compete with Costas in order to survive – because she has figured out how to reach a completely different set of customers, and a set that may be greater in size than the one to be found walking along the high street.
The data is there. It just needs to be accessed, shaped, analysed, and made totally relevant to the next person dreaming of being an entrepreneur.
If Mrs Miggins had had this information right at the start, she might have decided that she didn’t need a place on the high street and a rental agreement lasting a whole year, or a loan for enough equipment for her to make the whole day’s pies in one go.
She might instead have opted for a completely different business model, one that would give her a fighting chance of lasting much longer than the 3 year average lifespan for a startup – and a fighting chance of growing her business – exactly the outcome that the UK economy needs.
We at Develin specialise in helping businesses like Mrs Miggins to use all the data available to them, in particular to discover where tomorrow’s customers are to be found. To find out more about the role that data can play in your organisation’s growth, please give us a call.
When we read about using data in business, whether it’s to find better ways of working, or to locate the next customer, it generally refers to the bigger businesses.
But every business has data that it can use. And smaller businesses in particular can gain great advantage by making smart use of their data. This is data that’s already there in the business, and data they can collect in order to answer critical questions that might just give them the edge over their bigger competitors.
Here are 10 different sources of data that can help to boost your business, in particular when combined together rather than used in isolation.
- Customer conversations
Asking customers what they think of your service can be helpful. But you risk hearing what the customer thinks you want to hear. The real value comes from finding out why a customer chose you in the first place, what’s behind their purchase choice, and why they are purchasing now. This can help you to make offers that will entice them back, and help to find your next customer.
- Google analytics
If your website has any role to play in your business then this is a must for understanding whether, and how, your current and potential customers are reacting to what you say and what you offer.
For example, I can see immediately whether my on-line offer has created interest, how customers are finding their way to the offer, how long they stay, and where they go next. All of which can be broken down by a range of demographics.
For businesses in regular email contact with their customers, emails can indicate whether, and how, those relationships are changing. This can be measured not just in terms of the number of emails, but also their timing, and what sort of sentiment they convey. If you detect a change in the pattern of communication with a customer it may be an early sign of a problem starting to develop, or of something that you are doing well and that could be of benefit to others.
If your business is advertised on Facebook, your downloadable data includes data about all of your connections, followers, as well as the likes and emojis posted. With all of it in one place you can gain insights about the timing and content of anything posted or advertised in terms of reader reaction and sentiment.
Twitter is a great resource for early insight into emerging trends and newsworthy events that could affect your business. If you know what they are likely to be, Twitter data will allow you to spot them ahead of your competitors ensuring that you are there for your customers with the right offer, and at exactly the right time.
- Orders and sales
The pattern of your orders and sales will clearly indicate whether there has been any change in demand. But this may be too late to be of practical use. By the time there’s enough data for it to be noticeable, the factors causing that change may have come and gone.
However, you may be able to correlate that change with indicators from the other data sources mentioned. If so, you will know more about the factors that are affecting demand for your business allowing you to capitalise upon them in the future.
- Google Trends and Google Alerts
Google Trends allow you to see how frequently a search item is mentioned on the web, and Google Alerts will alert you whenever new content on that item is discovered.
This helps you to further understand what might trigger demand for your products or services and how these trends correlate with others e.g. interest amongst Facebook Followers.
- Your competitors
You will already be monitoring what your competitors are selling. But when it comes to how well they are doing and what they are planning to sell in the future then it’s important to track their announcements and any responses to those announcements. If you know who their key suppliers are, it is worth doing the same for those businesses.
A combination of Google, Facebook and Twitter data will provide a way to correlate any announcements they make, as well as the reaction to them, with general sentiment across the market place. This will help to indicate whether your competitors have stolen a march, or whether they have gone where others wisely fear to tread.
- Public data sources
The number of open source data sets relating to our towns and cities is vast. You need only go to data.gov.uk to get a sense of the breadth and depth of data available.
It is easy to become overwhelmed by data. Therefore it’s important to be very clear about what you are looking for before diving in.
However, contained within the wide range of available data may be information about your customers e.g. how their location explains particular delivery issues, or whether it influences the demand for and uses made of your products or services.
- Your own ‘ad hoc’ data
Last but not least. Imagine you want to get more trade from the people on the street outside. You could, for example, employ low cost sensors to detect the mobile phone signals that ‘walk’ past your shop front and those that stop awhile.
This might indicate when best to post certain offers in the window and what type of offer will attract the most attention.
Mobile technology is providing a myriad of ways to collect additional data about your customers as well as the movement of your people and vehicles. We would always advocate respecting the privacy wishes of everyone involved. However ‘ad hoc’ data can sometimes give critical insights about how best to serve your customers.
We at Develin specialise in helping businesses to make best use of their data. To find out more about the role that data can play in your organisation’s growth, please give us a call.
In their report ‘Where machines could replace humans—and where they can’t (yet)’ the consulting firm McKinsey & Company looked at the extent to which human activity in the workplace could be replaced by automation.
The somewhat sobering results are shown below. E.g. anyone involved in Data Collection and Data Processing can expect to lose two thirds of the time spent on those tasks to automation.
For those not quite convinced.
Predictable physical work in a car plant 10 years ago.
Predictable physical work in a car plant today. (Can you spot the two people in this photograph?)
Some other professions are also facing the same transition in work practices. The outsourced Accounting Firm may be one of them. It is threatened not only by automation but by the client themselves, as DIY Accounting programs such as ‘QuickBooks’ become more popular. And it is under threat from certain authorities. E.g. HMRC who want to gain direct access to company databases and records making the need for tax returns a thing of the past.
As these unstoppable forces take hold Accountants will experience a decline in two critical areas – their income, and their relevance to their client. How do you stay relevant when many of the tasks that allow deep insight into a client’s business are being taken away?
This might be just another story about a profession falling victim to unstoppable change.
However, a huge opportunity is presenting itself. Yes, clients may be seeing less and less relevance in the more traditional accounting services that they have received.
But they are increasingly thirsty for intelligence that will help them to drive their business forward. This might be a simple insight into their customers’ buying habits. Or a heads up about a specific risk heading in from left field.
They want this information because they are increasingly aware of the growing volumes of data becoming available to them. This is data from within their own business and from outside, with the potential to yield valuable intelligence that will help them to grow.
Trouble is, they don’t have the means to get it. Their own people don’t have the skills. And those with the right skills cost a fortune to hire.
Yet, this demand for information is only going to grow. The government has made a vast amount of data available about all our towns and cities and the people who live within them.
All the major cities across the UK have plans to become smart cities and amass extraordinary amounts of data that will be available to the public.
The Internet of Things is about to explode into life as low power communications networks spread across the country providing means for zillions of sensors and detectors to collect data.
All those businesses who could use this data to create relevant and timely information just need someone who already knows their business to show them what’s possible, what it could be worth, how to go about it, and where to go for help.
Accountants might not see this to be something that is for them. But if they want to, they can play a pivotal role in helping their clients grow within the new digital age.
They are already uniquely equipped in three key areas. They know their clients, they do numbers, and they know how to explain to clients what their numbers mean.
The element that may be missing for many is the ability to build business intelligence out of data.
But those skills can be acquired. And if they are, the size of the win win that will follow will be substantial.
The relationship between client and Accountant can be transformed, and the Accountant’s position as a trusted, knowledgeable and highly relevant strategic advisor will continue to be secure.
And all those client businesses will have their eyes opened to the opportunities available to them for better business intelligence, strengthening their ability to trade in an increasing digital world.
Executive Vice President, Academics
Charted Institute of Management Accounting
from his presentation ‘The Future of Finance’.
According to the American Bureau of Labor Statistics, in 2015 American men had, on average, almost 6 hours of ‘leisure time’ per day. Unsurprisingly perhaps women had less, a little over 5 hours. We haven’t found any comparable figures for the UK but they would probably be similar.
The figures are undoubtedly correct, but they just don’t feel right.
This is probably because, at the end of the working day, although we may have left the office to engage in what would appear to be a period of leisure, we don’t feel that it really is one. We may still be checking our phones or thinking about the emails that need to be sent in the morning. Mentally, we haven’t left the office at all.
The root of the problem is the dramatic rise in the amount of information that we have to process, not only about our work but also about our personal lives and the lives of those close to us.
Also, all of this information is reaching us through the same channels, and to an increasing extent through the same devices. How many of us use our own phones and tablets at work? The moment that we do, our work information becomes available to us 24 hours a day.
Separating out the different streams of information and then spotting what is important is becoming increasingly difficult.
In the workplace, ‘visual’ has been the watchword for some time. More recently this has been followed by ‘.. and keep it simple’.
This means: provide people who are struggling in an ever-deepening sea of complex data and information with something that works like satnav i.e. clear, simple, visual images that show the road ahead. They should ‘at a glance’ trigger critical questions, point to key issues, and highlight the right courses of action without the viewer needing to be aware of all the data that sits behind them.
The task of creating these images is far from easy, and it takes confidence. One of the reasons we are reluctant to pare information down is because of concern about what we might then miss.
But an increasing number of people in the workplace are doing just this. They are to be found in Finance, Operations, Sales & Marketing, Customer Services – pretty much every part of the business that depends upon data to function.
These people are mastering the data, collaborating with each other to figure out what the critical questions are, probing for the key issues, and building the skills and technology with which to create something clear, simple and that gets the vital points across.
We know that this is happening because we are there too, equipping these people with everything that they need for the task.
And they must continue to be supported. The better we are at seeing just the information that we need, and the more confident we are that we aren’t missing much that matters, the easier it will be to leave the office behind at the end of the day.
But what of all the other complex information in our lives. Can we apply the same watchwords about visual and simple to that as well?
The answer of course is ‘yes’. But how this is done depends upon how much of it there is and what matters most.
The process might have to start by switching off as much of that information as possible, perhaps for a week or two. In the quiet that would follow, it will be easier to focus upon just the few things that matter most, and to then let imagination come up with answers to the question ‘how can I turn what I need into something visual and simple?’.
Create something that works for just those few things, whether it involves technology or just notes on the fridge door, and then wait awhile before switching anything else back on. All being well there will be little inclination do so, allowing for the additional leisure time that has been allowed back in to remain in place for good.
We have long experience of helping organisations to make things visual and simple. If we can do the same for you, then please give us a ring.
Much has been written about Santa’s achievements on Christmas Eve in terms of the distance and speed that he travels, the weight of gifts that he delivers, and the number of mince pies that he consumes along the way. But without an impressive array of Artificial Intelligence algorithms underpinning his efforts, none of it would be possible.
The logistical challenges are, of course, formidable. In order for Santa to do his job a customer database is needed, one that contains many billions of addresses and is 100% accurate at exactly the same time every year.
Even though many of Santa’s wish lists are posted on Christmas Eve, next day delivery is guaranteed, with a delivery window of perhaps two hours between midnight and 2 am. And none of the packages are ever left with a neighbour, or just dumped on the door step.
But, not everyone provides Santa with a wish list. Therefore the first, and perhaps the most important of the algorithms is for ‘gift preference prediction’.
For this algorithm to work the customer database needs to include a record of all previous gifts, purchases, likes, dislikes and browsing histories, not to mention a record of behaviour (specifically incidences of being ‘good’ or otherwise).
A suitably named ‘logistic regression’ algorithm will produce the necessary gift preferences, ranked in order of their ‘fit’ with the each of the customer characteristics contained in the database.
All the gifts however need to fit into the sleigh. Therefore, running in parallel with this algorithm is one that adjusts the preference list if the total volumetric weight of all the gifts steps outside set boundaries. The sleigh may be extensive in size but it is assumed to be of finite volume.
A suitable choice is the ‘Knapsack’ algorithm. It figures out the maximum number of objects that can be fitted into a set volume. This algorithm will adjust the choices made by the gift preference algorithm in order that the maximum volumetric weight from all the gifts can be accommodated within the sleigh.
These two algorithms need be run many times over. Last minute wish lists have to be accommodated as do last minute fluctuations in behaviour, the latter being used to rule a potential recipient into, or out of the gift delivery schedule. This means that, once the final list is confirmed and signed off, there may be only a few minutes within which to load many millions of tons of gifts on to the sleigh in exactly the right delivery sequence.
The next algorithm concerns the routing of the sleigh. Given the extraordinarily tight delivery window the priority will be to find the shortest path through all destinations.
Dijkstra’s algorithm is perhaps the most suitable for the task. It is used widely in routing protocols to find the shortest path through, for example, a complex road network.
Lastly, there is the need for a load balancing algorithm similar to that used with aircraft. During the trip weight will be lost from the back of the sleigh as gifts are distributed, but gained at the front as Santa eats several million tons of mince pies.
There are few references to Santa’s sleigh requiring ballast movement, but this will probably be essential for stability.
The IT implications are significant. If limited to a single machine, and given the size and scope of the task, the preference prediction algorithm would require many days to run. The Knapsack and Dijkstra’s algorithms would both require many years. All the algorithms would, however, need to be run many times over to capture last minute dips in behaviour as well as the last minute Santa wish lists.
In order to reduce processing times down to a few seconds, an array of parallel quantum processors would be required that could only be assembled in space. And the data that would be needed, about our choices, behaviours, thoughts and wishes would be vast, dwarfing (or should it be ‘elfing’) anything that any of the world’s intelligence agencies could be capable of collecting.
This data would of course be a gold mine to any intelligence agency or commercial enterprise.
But there hasn’t been a word about any form of cyber attack or leak, so the data seems to be totally secure. And every year the whole enterprise works like clockwork (apparently), not to mention on time.
If only we could all operate like Santa.
Happy Christmas to all our readers.
We have long experience of helping organisations to operate like Santa, and all the year round – as analytics are not just for Christmas. If we can do the same for you, then please give us a ring.
This week the Prime Minister and Communities Secretary announced a £40 million homelessness prevention programme, with an emphasis upon ‘innovative methods and seeing what works’.
The Homelessness Reduction Bill is following close on its heels. This will place a duty of care upon Local Authorities to prevent homelessness. Housing Associations haven’t yet been mentioned but Local Authorities will need their support if they are to fulfil this remit.
This matters to us.
Using machine learning methods we have been able to pinpoint people living in Social Housing who are at high risk of homelessness, either through eviction or abandonment. We have piloted our approach within one major Housing Association, we are about to put it to work within another.
However, if the Government’s vision is to become reality, approaches such as ours need to operate across different organisations and a large number of locations.
In principle, this shouldn’t be a problem. We place a mobile app at one end of the operation, for use by Field Workers on the ground, and cloud based machine learning algorithms at the other. Sitting in the middle we have specialists who understand what it takes to spot people who are heading towards homelessness.
The reality however is likely to be different.
For example, from 2010 to the present day, the government has invested £500M in homeless prevention. As a result 1M cases of potential homelessness have been avoided.
This required collaboration between a myriad of different agencies, government departments and Charities all of whom interact with the people of interest for different reasons, at different times, and in different ways. Although we have technology that can operate pretty much anywhere, to make it work we still need the involvement of all those different organisations. The bureaucratic hurdles are likely to be dizzyingly high.
That said, the government’s latest approach could give major grounds for optimism. In short, funds will be allocated to Local Authorities (LAs) to pilot new initiatives to tackle homelessness and the LAs will be left to get on with it. But if an LA doesn’t get to grips with homelessness within its patch, penalties will apply.
Hopefully the combination of a ‘hands off’ approach and devolved funding will encourage new local initiatives to emerge. They will operate under the governance of the LA, but be run by entrepreneurially minded people who will want to hear about innovative methods and smart use of technology.
This is the right approach to take, because it should allow things to move fast. Homelessness has doubled since 2010 (Homeless statistical release Feb 2016 – Dept Communities and Local Govt) and it is only set to grow further. If LAs don’t make progress quickly they will have a big problem on their hands. Yes, if the Bill is passed they may incur the aforementioned penalties, but more importantly they will have to provide emergency accommodation in very short order for an ever increasing number of people with nowhere to go.
Not only will smarter technology help but, once it is up and running, it can be used to find those people who are not yet on anyone’s radar, but who will eventually end up on the street. This group includes young single people who may be sofa-surfing today, but who will soon be riding the night buses because they have nowhere else to go.
It will be a major challenge getting it all to work, and we will inevitably stumble into bureaucratic barriers along the way. But if, as a result of some astute predictions, more people can experience a friendly knock on the door just at the moment when all seems lost then it will all be worthwhile.
More data can mean fewer words
One of the great things about the steady growth in data held by businesses is that there are ever-more imaginative ways to illustrate the problems that need to be fixed.
If we just use words to describe a problem, they become just another set of paragraphs in a report – to be skimmed over and immediately forgotten.
Pictures of course catch the eye. If they are good they might then prompt the question ‘how did you create that image?’. And if that happens, the next question is almost certainly ‘what’s it telling me?’.
Job done! They’re engaged.
Here are a few of ours that have recently caught the eye. Some words were needed, but in each case we set ourselves the challenge of using no more than 50. See what you think.
Customer retention strategy
These are clusters of customers. Circle size shows what each is worth. The strands are features that tie them together.
Blue customers have been poached by the competition. The closer the yellow are to the blue, the more likely they are to follow. Unless they can be persuaded to stay.
Service efficiency improvement
Each dot is a visit to a customer’s home by a home repair service.
Each line is one day’s journey by one person. Line thickness indicates distance driven.
Better scheduling of visits can reduce the number of separate lines, extend their length, and make the thick lines become thinner.
Shifting customer profile
Each line is a different customer channel and each dot is a customer. The further to the right, the more they have spent in one month.
Switch between months and you can see densities shift: As their spending changes. Or they switch between channels. Or they just disappear altogether.
Back office efficiency
Month-end is a busy time for Accountants.
Each circle is an activity performed at month-end. The lines show how much traffic there is between the activities.
The heaviest traffic goes between the four activities that deliver the least value: Making changes and reworking the outputs.
(45 words – but not including those in the picture)
Develin Consulting Ltd
If we can describe the issues that your business experiences, in 50 words or less, please drop us a line.
Land of hype and glory
A recent conversation with an IT analyst was fascinating. She suggested that, as far as big data analytics were concerned, if we want to put our exciting new skills to the test, the sectors we need to focus upon are retail, financial services and higher education. The hype and the smart money are (still) with predicting which film we might like to buy or spotting when our credit cards have been cloned.
She agreed that other sectors also have plenty of data, and opportunities to put it to good use. But they just don’t have the money for this sort of thing.
These others are, of course, sectors such as our health and education services, local government, housing associations, legal practices, and some of our larger charities. And the data in question might just help to reduce the costs of primary or secondary care, or of housing the homeless.
But, to a certain extent, she is right.
She was pointing out that those who decide to employ big data analytics are prepared to invest a lot of money. Given that sectors such as Health and Social Housing don’t have much spare cash they are unlikely to join the party.
So, we could follow her advice (and the hype) and seek the riches and glory that will come with yet another prediction engine for holiday destinations.
Or we could do something more useful. We could continue to use these techniques to make a difference where lives matter.
The unhelpful answers are sometimes the best
But, there is the money problem. If there isn’t much of it to spare then the perception is probably that nothing much can be done.
Well, let’s nail this perception.
As a start, the analytics involved should be planned to simply deliver a healthy Return On Investment (ROI).
This may not sound very helpful. But the rigour that this will involve will impose a tight and much needed discipline across the whole exercise. In short, it will force everyone involved – to think.
It means for example:
- Being rigorous in defining the problem to be solved;
- Designing analytics that will help to solve just the problem at hand;
- Being very creative about how to extract and process the data needed;
- And knowing exactly how the benefit will be delivered that will make the whole exercise worthwhile.
Why should these make the difference?
Because this is the exciting world of ‘Artificial Intelligence’ and much of the hype has been generated by those who can explain what’s possible rather than by those able to say what’s needed. And there are many businesses that have jumped in with both feet (and a great deal of money) in their search for the possible that are now struggling to see any sort of return on their investment.
But, when you are forced by financial constraints to step back from what’s possible to examine just what’s needed, things can suddenly appear to be a great deal simpler, easier and cheaper.
It’s similar to a film plot in which a cop is tenaciously picking the lock of a property they want to sneak into. Their partner however, seeing how long it is taking, steps over them and just kicks the door open.
The problem to be solved wasn’t how to pick the lock, it was how to get the door open.
We are in a world of increasingly sophisticated lock picking devices. But at the end of the day, if you are trying to improve social housing provision as quickly and as effectively as possible, the best solution might be to just kick the door open.
Develin Consulting Ltd
We at Develin specialise in analytics that simply get the door open. Please give us a call if you would like to know more
We can be more confident
We get to meet a lot of Finance people. We have the privilege of teaching Finance Teams within the various Mastercourse Programmes that we present, and of working closely with them as part of our day job as Consultants. Based upon the conversations that we have had along the way, in particular about the demands being placed upon the Finance Function, we believe that a significant problem is starting to emerge.
The amount of data that is available within businesses is growing, and yet Finance Teams seem very hesitant to exploit it.
More data should mean finding more patterns in customer behaviour, trends in demand, insight into risks, threats and possibilities. It should mean knowing more about what’s ahead, and increasing your confidence in what you know. As a result, forecasts should be looking further ahead and be packed with options, scenarios and alternatives.
But, on the whole (and there are exceptions), this is not the case. Nor, we believe, will it be anytime soon.
Here is just one example of what should be possible:
- Financial forecasts that look well beyond the end of the financial year;
- Forecasts starting with projections of history using regression methods;
- The use of ‘confidence intervals’ (e.g. a 95% probability that it will be between A and B);
- Refinements to the forecast using new information – plans in place, trends in demand, economic factors, competitor intelligence, consumer preferences and purchasing behaviour, regulatory impacts .. and so on .. all of which are accompanied by confidence intervals.
Some industries have taken a probabilistic approach to forecasting value for many years – oil companies, for example, when valuing their reserves. But for many businesses their forecasts are still single numbers. They will almost certainly be wrong. We need to know by how much.
But data rich and time poor
Of course this all depends upon data, and most businesses feel that they don’t have enough of the right sort for this to be possible. But that position is changing.
Businesses typically have an increasingly rich seam of historical data to draw upon as well as better tools with which to access it. Businesses are also closer than ever to powerful sources of information that cover all the drivers of demand – from customer preferences through to the economic and political climate.
What is missing is the time, and the skill set, to put these to good use. For example, from a recent survey of staff time in a client organisation, Finance Managers had less than 5% of their time available for ‘value adding analysis and decision support’. That’s a couple of hours on a Friday afternoon. And this result is consistent with all the other surveys that we have conducted.
Also, based upon entirely unscientific and ad hoc surveys of the delegates to our Mastercourses, there seems to be a major shortage of skills. E.g. only about 5% of delegates have so far been able to explain what a confidence interval is and how it might be applied to a forecast. Extrapolate this out across all sectors of industry and we appear to have a problem.
In short, businesses are missing out on the value contained within their data. To see how, you need look no further than the weather forecast.
Imagine being the organiser for a church fete. If the forecast says that it is simply going to rain on the day, what do you do?
Most would probably assume the worst (i.e. heavy rain) and shift the venue indoors. This would guarantee everyone stays dry but may lead to less people coming.
But the Met Office can provide you with a percentage chance e.g. 60% chance of rain for no more than a couple of hours. This might convince you to leave things as they are. A bit of rain dodging might be needed, but not enough to keep people away. And with the fete outside more people might be persuaded to come. In other words, with a little more information attached the forecast can lead to a different and potentially better decision being made.
Put your team to the test
So here’s the question. How confident is your Finance Team at dealing with confidence? We have four simple questions for your Finance Manager.
- Do you have more than 5% of your time dedicated to improving the quality of decision making?
- Do you know what confidence intervals are and how you could build them into a forecast?
- Are you confident about joining data together that comes from different systems?
- Do you know what regression analysis is and how it might be applied to a forecast?
If the answer to two or more is ‘yes’ then things are looking up. If not then your forecasts and valuations will probably remain as single numbers accompanied by pages of notes explaining why they are probably wrong, but giving few clues about how much they are wrong.
Develin Consulting helps organisations to turn the value within their data into better financial forecasts
Using big data methods we have been able to predict which social housing occupants are at greatest risk of abandoning their home without warning. The numbers are very small but their impact is big – court action to secure the property, unpaid rent to chase.
We started by flagging those who had probably already gone – primarily to recover the property more quickly.
We were then asked to find those who haven’t yet departed but whose behaviour suggests they probably will. If they can be caught in time, it may be possible to turn things around.
We are talking about 10 a month, out of a population of tens of thousands. About a third of the population are behind with their rent at any one time, and about a third of this group owe many hundreds if not thousands of pounds. Needles in haystacks don’t come much better than this.
But we have just produced our 1st ‘top 10’ list of candidates. To give an idea about how it works, it might be worth telling a few of the stories from those on the list.
Lives with her teenage son in Derby. Back in November her working hours were reduced to 1 day per week. She was then told that she had been paid too much Housing Benefit. Her benefits would therefore be reduced until the account balanced once more.
Unsurprisingly Jenny’s arrears started to mount. Her mobile number then ceased to work, gas safety contractors failed to get access, and cards through the door asking Jenny to make contact remain unanswered. We think she is still there. But, perhaps out of fear, has she simply gone to ground?
Lives alone in South London. His rent payments have been sporadic and now and again he has fallen into arrears, albeit not by much. When this has happened he has received reminders to pay to which he has always responded well.
But back in December he did something not seen before. Whilst his rent account was still in credit, he phoned to check the balance. In January he paid his rent in full bringing his account further into the black. But he hasn’t paid a penny since. There are signs someone is there, some rubbish has been put out into communal hallways. But is it Michael?
Lives with her partner in Plymouth. She has fled the property due to domestic violence, but her Partner still lives there. He hasn’t been paying the rent, nor has he allowed anyone in to check for gas safety.
Alicia wants to return, but to live there alone. She therefore wants her Partner evicted for non payment of rent. Unfortunately the tenancy is in both their names, so if he is evicted then so is she. And if that happens she goes back down the ladder when it comes to getting another home.
Alicia can’t stay much longer where she is now. Indeed she has already been in touch with the local Homeless Persons Team. How do we help Alicia?
And there are many more to come
Those are the stories behind 3 of the 10, and to find them we had to work data from multiple systems hard. Each indicator mentioned above was there on a system somewhere. But it was only when all were brought together did the important patterns emerge.
There are probably thousands out there who look just like Alicia, Jenny and Michael – people who will remain invisible until it’s too late but whose numbers are likely to grow as Universal Credit bites.
But if they can be found in time there are systems in place to intervene.
Ali, for example, was on one of our earlier lists but he had already been flagged as needing help. This was because he was still a new tenant, and still under the watchful eye of the team that gave him his home.
Ali was entitled to Housing Benefit, but his claim had stalled. He was therefore falling into debt. Thankfully the local support team sorted his claim out, and the local Money Advice Unit found other benefits to which Ali was entitled. Local Volunteers also stepped in to sort out his house and garden.
Not everyone is going to be found in time, nor can everyone be helped in time. But if we make the best use of the data to hand then we can at least better understand the size of the problem. The next step will be to figure out how to respond.
This month we found Alicia, Jenny and Michael as well as the others on the list. I think we are going to find a whole lot more in the months to come.
Unlocking value through analytics