Chris Conway
Chief Architect, Quantiv
There’s an old political adage to the effect that, “If you’re explaining, you’re losing.” It’s usually credited to Ronald Reagan and is typical of his style, though as with many of his quotes it could have been borrowed.
It’s a reminder you need brevity and clarity for good messaging. And the statement could be considered a good example of the point it’s making. Short enough to be easily digestible, and pithy enough for it to seem as if no further discussion is needed.
The importance of ‘explaining’
Even in Reagan’s day, when the media didn’t feel as rushed or immediate as it does now, being able to use a simple phrase or image to express an idea was a good way to ensure a point could be easily understood. Today, when just a paragraph can feel like an overly long description, the advice is probably essential for ensuring ideas are not only understood but considered at all.
But while the sentiment may be well-founded for politics, its wider application has always bothered me, especially in the world of IT. This is perhaps not surprising given a lot of my time is spent analysing problems and then proposing, and explaining, possible solutions.
In short, sometimes ‘explaining’ isn’t an inherent weakness. Instead, it can actually be the central aim.
The AI car wash test
I was reminded of this again recently by the viral artificial intelligence (AI) car wash test, where popular large language models (LLMs) were asked whether someone should walk or take their car to the nearby car wash. It’s an instance of where an explanation would be extremely helpful. So, instead of the response being a confident, “For short journeys, you should definitely walk,” a more nuanced explanation could be, “I’ve no data about walking to a car wash but for short journeys walking is usually better.”
The results of the AI car wash test emphasise the gap between pattern matching and true understanding. In a situation like this, the words of the journalist and satirist H. L. Mencken might be an apt modifier: “There is always a well-known solution to every human problem – neat, plausible, and wrong.”
The benefits of deep understanding
However, that’s not to say all ‘correct’ solutions must be unfamiliar, messy and implausible. So, there may be occasions when having to explain does indeed show unclear thinking. But at other times providing an explanation could show the complete opposite, i.e. that a situation is understood. And while cursory statements might be sufficient in politics where the intent is to convey an overall impression, in many other areas a deep understanding is necessary.
Either way, the basis of any accurate description, whether superficial or deep, is information. Even a short summary must still be based on evidence, while thorough explanations need comprehensive details.
Data in context
But information doesn’t just mean a lot of data. AI answers like the one above show how answers based only on an enormous amount of data can still be misleading or just plain wrong.
Much of what turns data into information is context, in particular: the who, what, where, when and why of the data.
This is sometimes referred to as ‘metadata’ because it describes the characteristics of the data. We find the best way to start identifying that metadata is by considering the points at which significant events occur in an organisation. We call those points ‘interactions’, and they form the primary description of an organisation’s behaviour. As such, they also form the perfect basis from which further contextual data can be collected.
Working in this interaction-oriented way makes it easier to move from observed behavioural patterns to business process descriptions that include further details about the interactions: the who, what, where, when and why.
This approach allows contextual information to emerge naturally from your organisation’s operations, rather than being forced on to it with assumed, and often inappropriate, concepts.
In this way, the information gathered is more closely aligned with how your organisation operates, and so in turn allows explanations to be both more approachable and more accurate, even for quite small amounts of data.
How Quantiv can help
We’ve designed our NumberWorks method and NumberCloud service to support this context-gathering approach. NumberWorks allows the behavioural patterns that emerge to be described in a consistent format, while NumberCloud supports the collection of your metrics without the need for predefinition of the data to be collected.
It’s this context that makes your underlying data approachable, and therefore useful. When it’s present, your audience will have more patience than you’d usually expect and time for explanations. But without that context, your audience will often suspect (rightly) that even more detailed descriptions are too simplistic. So, all things being equal, clarity beats brevity.
To wrap up, perhaps the real adage should have been, “If you’re having trouble explaining, you’re losing.”
Learn more about NumberWorks and NumberCloud
Talk to the Quantiv team on 0161 927 4000 or email: info@quantiv.com
