our views and our knowledge in analytics and other releveant topics

our blogs

What does “real-time” mean to you?


Many people ask for “real-time” data to drive their analytics or optimisation programmes. But this can mean different things for different people, or for different purposes. If you want to integrate your behavioural data with an email send that goes out once a day, or even once a week, then your “real-time” requirements will be different to someone who wants to integrate with display media buying programmes based on whether someone has bought a product within the last minute. Of course, the infrastructure and systems requirements, and indeed costs, will also vary considerably for these scenarios, so it’s important to pick the right level for you.

But the current myth seems to be that “real-time” is only really properly cutting-edge and innovative if it is “real-time to the second”; this myth is often spread by software and hardware vendors, who inevitably want to sell you the most sophisticated, and so expensive, version of their products.

It’s important not to get caught up in that myth, and to remember that innovation is not just about technology; certainly, it is often triggered by technology, but it is how you change the people and the processes around technological advances that really drives innovation that is valuable.

So, one of the first questions you need to answer is: “how real-time do I really need to be?”. Before your boss or board demand that all your data is provided to you in “real-time” without thinking about the organisational and cultural impacts this may have, here’s a quick rundown of some of the historical innovations and break-throughs in relation to time to put this in context. In the last millennium, there have been periods of major innovation around the measurement of time and time-keeping; all of these involved cutting-edge technology developments and were ground-breaking in their era. But most importantly, they all outline the cultural or practical applications that these changes brought about.

700 years ago – “Real-time” level: to the hour

If your town had a church tower with a clock on it at the beginning of the fourteenth century in Europe, your community was at the cutting edge of “real-time” innovation; it meant everyone could see what time it was. There was a development in horological technology between 1280 and 1320, so mechanical clocks could now be placed in prominent positions in buildings and towns; previously, water clocks had been used, but they were difficult to manage, and were, well, wet, so were difficult or dangerous to put in high places. And so the mechanical clock led to clock towers, and were part of the massive civic and church-building programmes which you can still see in British and European towns and villages to this day.

But like all things “real-time”, there had to be an application for this; clocks represented new levels of accuracy for civic or administrative purposes. They meant that, for example, your market could start at exactly the same time each market day, making it more reliable for customers and traders alike; they could all be there at the right time.

But, this was not timekeeping accuracy as we would know it; clock towers in this period only had an hour hand. There was simply no need to be more accurate than that; the villagers needed to know that the market started at 10am, or that Mass started at noon, or the festival started at 3pm. It didn’t matter if it was ten past or quarter past – the market was still on, the sun still shone and the crops still grew.

How this could apply to you: This innovation scenario is most relevant for organisations that are either just starting out on their digital transformation programme, where it is important to show the business case and potential of the programme, but without breaking the bank. But this also applies to “traditional” companies where customer buying processes are more long-term, and take place offline more than online; in other words, where the crops will still grow, whether or not digital transformation take place. These might be professional services or other B2B companies or where existing communications with customers don’t need to be more than weekly or daily.

400 years ago – “Real-time” level: to the minute

Minute hands only became relevant centuries later, when scientific advances, and a cultural desire to be accurate, grew with the Renaissance. In perhaps the most famous example of “real-time” cultural change in history, Pope Gregory XIII ordered the Gregorian calendar to be adopted in 1582. The traditional view of the Earth’s orbit of the Sun lasting 365.25 days had been shown to be inaccurate – by 11 minutes! Clearly, every minute was now important and carefully measured.

But again, these weren’t just real-time changes for the sake of it. This fractional correction meant that Easter could be celebrated at the same time – different Christian traditions had used different calculations, and this was a final attempt to unify when this should take place. But temporal accuracy didn’t just fuel religious observance; it drove the Renaissance’s thirst for scientific knowledge. Copernicus had published his thesis on heliocentrism in 1566, using accurate observations and advanced mathematics to argue his case and his ideas were picked up by Kepler and Galileo. Accurate (to the minute) clocks and timepieces were so commonplace by the early seventeenth century that two amateur astronomers, Crabtree and Horrocks, independently measured and timed the first recorded transit of Venus in 1639 – at 6 hours and 40 minutes. It was this observation that enabled accurate calculation of the size of Venus, and, by extension, the size of the Sun and the Earth and the rest of the Solar System.

How this could apply to you: This scenario pretty much describes most companies going through a digital transformation programme. Gregory XIII is the equivalent of the CEO who suddenly realises that the way his behemoth of an organisation operates no longer fits with the reality of the world, and so needs to initiate major cultural change. His focus was on unifying time measurement, but most CEOs now are interested in how to record and target customers that use new digital channels in new ways. Your company has probably tried to work with digital, but it hasn’t really worked because the old processes haven’t changed from before digital. But now you need to kick on…

300 years ago – “Real-time” level: to the second

In October 1707, HMS Association went down with all hands, together with 3 other ships of the line, when they hit rocks on the Isles of Scilly sailing back from successfully campaigning against the French. 2000 sailors died in total, in the worst maritime disaster of the age. What was worst of all was that it was avoidable; because the flotilla had thought it was further west than it actually was, they thought they were in open water and so sailed through a sea fog on a course that should have been safe. So, the Admiralty needed accurate ways of calculating longitude, and the only reliable way was to have a clock that could keep perfect time; if you knew the time in one fixed place – London, say - you could therefore establish how far west or east you were of that point. On a journey at sea that could last months or even years, a clock that was fast or slow by a second would add up over time, and lead to potentially devastating miscalculations. This led to the Longitude Act, which funded a competition for the production of a reliable, accurate and sea-worthy clock and led to the brilliant clockmaking of John Harrison, who won it. But more than that, this ushered in the modern era of the need for consistently accurate clocks, down to the second. This is the first time that there is a genuine, scientific definition of “real time” that we in the 21st century would recognise and a very clear and obvious application for this level of accuracy – clocks could be used to keep people safe.

How this could apply to you: This will be relevant if you have business-critical processes or systems where you may need to respond, react or intervene with a customer journey or transactional behaviour at that moment, or where you have a regulated activity where you need to demonstrate an exact audit trail. The most common examples of both of these are in banking and financial services, but this can also apply elsewhere. Increasingly, organisations are building data-driven predictive models to manage and optimise customer targeting and communication strategies which are based on and designed to respond to “up-to-the-second” data; these types of models are being used by retailers, technology and telco companies and others, as well as in financial services, so this could apply to any companies using data in this way.