By David Maitland, General Manager EMEA, Couchbase
How well do you know your best friend? When you consider that deep meaningful friendships are usually built up slowly over the course of several years, it would be no surprise to learn that you know exactly who your best friend is, what they like, and what they dislike at any given moment. However, what if, instead of forming this knowledge over the course of several years, you had to to reach the same level of knowledge about somebody in just a fraction of a second? It might sound like an impossible task, but it’s a challenge that those in the advertising industry have to face every day.
The reason for this is that an increasing number of ads today are uniquely tailored as a result of taking into consideration the current behavioural, demographic and psychographic characteristics of consumers. What this means is that, in order to create the most relevant content for their audience, advertisers have to have an almost microscopic understanding of end-users, their behaviour and interests. In many ways, they have to get to know consumers as if they were their best friend, in a fraction of a second.
To achieve this, many ad-targeting services are using NoSQL technology to analyse billions of user-related events, and to process huge amounts of data, some targeted ads are delivered as quickly as 40 milliseconds, and doing this effectively can be a huge revenue driver for advertisers. Now, although this may sound like an easy thing to achieve, the truth is that many companies are still grappling with legacy technology that is not designed for this purpose. After all, getting a 360-degree customer view is very important, but should it come at the expense of infrastructure efficiency and performance? Remember that if the process is slowed down and the right advertisement is not displayed to the right customer at the right time, it can have a significant negative impact on revenue.
In order to achieve a successful outcome, advertisers have realised that they need data management technology with the ability to analyse, process and execute billions of user-related events. They also need to be able to source these events from different devices and platforms and have the ability to process both structured and unstructured data. User profiles have to be scanned and updated immediately and then decisions have to be made about what advertisements to show and when. This requires data management solutions that combine both the ability for high-throughput analysis as well as low latency.
This is where NoSQL database technology comes into the picture. In contrast to relational databases, NoSQL databases have the ability to scan user profiles and query datasets with sub-millisecond latency to get the data needed to make optimised decisions about real-time ad placement and tailoring. Essentially, it can ensure that the right data is sent to the right person, at the right time, and make delays and slower processing times a thing of the past.
Advertisers who still think that they can cope with their data management system as it is, or that they can simply upgrade existing systems will also quickly realise that this is not a viable option. Let’s not lose sight of the fact that the amount of data businesses have to deal with will increase exponentially over the next few years. Apps are a good example of this, and it’s clear that they are becoming an increasingly popular platform for advertising. By way of context, in 1975, an interactive software system with 2000 users represented the pinnacle of scale. Nowadays, an app can grow a user base of over 2 billion — over night. Today’s applications are online 24/7, 365 days a year — and so are the users.
Flexibility to add and remove user profiles, to update them at any time, with the ability to scale upwards and downwards is becoming a key consideration. The ability to add structured or unstructured data without first defining a rigid database schema are also invaluable advantages in the race to enhance user experience by offering tailor-made advertisements. In short, businesses need a database that is there for them, understand where they are coming from and what they are trying to achieve — just like you might expect your best friend to. And just like a best friend, they should be able to grow with you, keep up with you and be flexible enough to fit your way of life — or even your way of business!