What can we expect for the future of marketing analytics?
It’s time to ditch your Marketing Dashboards.
My first true, long-lasting, work love story happened a few years ago when I used Tableau Software for the first time.
I was working at a marketing/data agency and we were handling media budgets on behalf of our clients. For each client, we set up real-time, multi data source dashboards, allowing us to track all of their marketing KPI’s and metrics through tables and data visualisations. I was truly amazed — and still am — about the power of such a tool.
It enabled us to gather and easily manipulate the data into one central platform and to slice and dice it, analyse trends and uncover insights in order to make informed decisions.
But this was 2014.
Time has passed, technology has evolved and now is the time to ditch your marketing dashboards.
Ok, maybe not all of them, as tools like Tableau are still useful in tons of situations — but there is a better solution out there for many common day-to-day needs of marketers.
1. As a performance marketer, you're probably working with dozens of dashboards
Even if you have consolidated your tracking data into one single source of truth dashboard, you probably still jump from this one to all the others where you actually optimise your marketing campaigns.
A Winterberry Group report quoted by Hubspot highlights that “Marketers are on average using more than 12 different tools, and some are using more than 31 tools to manage campaigns and data.”
Can we admit that just isn’t sustainable?
2. So much data, so many questions… and so few answers
Performance marketing has been quickly changing thanks to the granularity of the data you can track and learn from. But it’s overwhelming just how much data marketers have access to.
Metrics from dozens of different media channels, hundreds of campaigns, thousands of keywords… it’s hard to keep track. According to a McKinsey report, “25% of knowledge workers’ time on average is spent searching and gathering information.”
You can, of course, organise your data to highlight KPIs, key trends and add a colour system / heat map / smart data viz to help you read through it — and it works.
At the end of the day, even if dashboards give you access to ALL your data, it still makes you do all the heavy lifting. taking data-driven decisions, either to scale highly performing campaigns or to fix bad ones as quickly as possible.
Where should you start?
3. … and (equally important) where should you stop?
It’s easy to get carried away as a performance analyst. There will always be another KPI breakdown, another insight, or another trend to look at. It can be just too much data.
And who doesn’t like to analyse for the sake of analysing… right?
Terrified by the idea of missing out on key information, we have all spent hours looking at data, creating pivot tables and over analyzing instead of drawing a conclusion.
Lots of literature has been written around this, with tips & tricks to face it — a London Business School Professor gave his in Forbes.
So what tools actually exist to make performance marketers’ lives easier?
1. Automatic optimisation
These solutions have been around for a while but the technology keeps improving and can drive fantastic results for marketers.
Based on complex machine learning algorithms, lots of tools enable automatic campaign optimisations, sometimes even across platforms. As soon as marketers start to scale their marketing budget and spend on multiple platforms and countries/markets, allocating budgets and adjusting bids on a daily basis can be extremely time-consuming.
A little bit of transparency can help give peace of mind.
2. One-stop channel management platforms
Tools have emerged recently that gives you the ability to manage multiple platforms within a single tool. You can create, modify, optimise and pause campaigns on Facebook, Adwords, Apple Search Ads etc. through one unique console.
It gives you control on your ad spend — as you’re in charge here -, and automatic rule setup saves time. The common console spares you the learning curve when testing new channels.
3. AI-driven recommendations
Marketers are lucky enough to benefit from direct applications of AI. If bigger teams would probably have invested in in-house data science teams, tools are also already in the market for medium to smaller teams to use.
A few applicable use cases available are:
- Conclusions, and not only insights: based on historical data, the algorithms can suggest ROI-enhancing recommendations
- Anomaly detection: detect any suspicious change or pattern, to enable quick corrective action
- Predictive analytics: predict user behaviours based on previously observed patterns
4. Natural language processing
Natural Language processing is defined by a workgroup at Stanford as “algorithms that allow computers to process and understand human languages.”
NLP will enable marketers to ask for specific data points or more complex questions as simply as if they were searching in Google. Similar to query for information, marketers will be able to ask a software what they are looking for, such as “how could I increase my ROI?” or to generate complex reports on-demand, such as cohort analysis, funnel reports etc.
Instead of looking at the same dashboard over and over again, wasting your time, your software will soon be able to tell you how, where and why you should look at a specific asset,
Poorly performing campaigns and creative issues won’t disappear, but at least, you won’t be alone to face them.
This story is published in The Startup, Medium’s largest entrepreneurship publication followed by + 381,862 people.
Subscribe to receive our top stories here.
