Honestly, it shocks me that businesses are still not actively utilising their data. I wonder whether people harvest their data expecting it to jump up one day, collect itself into a strategy and present itself to you, ready for profit. Data is one of the most valuable assets your business has, enabling you to back your decisions with science within moments. However, few seem to actively be doing it. Those who are have the data (ironically) to show its success.
Why test new marketing strategies, sales strategies and even products and services when you have such a large pool of data to validate your ideas with? This blog will guide you through a step by step process to get the most out of your data.
But before we proceed, I do need to add a disclaimer. In general, there are two types of people who feel they aren’t adequately utilising their data. They are:
Those who aren’t collecting much data at all.
Those who collect data but don’t get around to utilising it.
This blog is for the second individual, with the first needing to focus primarily on data collection and aggregation.
I’ve found locating your data to be most effective when using a departmental approach. For instance:
I’d like to collect marketing data to create more effective marketing campaigns. To do this, I will collate web traffic and conversion data from Google Analytics, email marketing data from Mailchimp and social data from Facebook and Instagram.
Let’s break this statement down:
Identify the department: Marketing
Confirm the reason why: To create more effective marketing campaigns
Isolate each platform and the corresponding data retrieved: Google Analytics (traffic and conversion), Mailchimp (email marketing), Facebook/Instagram (social)
Upon understanding this, the process of collecting and aggregating your data becomes incredibly straightforward.
Once you’ve identified the source of your data, it’s time to collect and aggregate this data. Naturally (but for those who need clarification), the steps are:
Collection: Importing the data from external sources to a local database.
Aggregation: Collating information from tools offering similar data (such as Facebook and Instagram).
For collection of the data, you need to start by exporting data from the respective source. For instance, in the case of Facebook, you can export advertisement related information using Ads Manager (for those interested, here’s a document detailing the process to follow).
For aggregation of the data, you need to identify which sources offer similar data, and collate this through a normalising process. By similar data, I’m referring to topical data. The different categories in our example would be social, email marketing, traffic, and conversion. While Google Analytics would be the only tool within our traffic category, naturally we have Facebook AND Instagram data to cover social. To normalise data from different sources, you want to ensure you determine where there is crossover between the data captured (in this case, both platforms offer “likes”, “comments”, “shares” and “views”). Upon identifying this crossover, you can then collate data from both sources into a single database (or spreadsheet) with an additional column identifying the source. This allows you to determine which platform is harnessing the best results for each advertisement type.
Next is the fun part. Although, that does depend on your definition of “fun”. This is where you really start to see the benefit of data. I will flag that data scientists can do much more with your information than we’re about to. However, we will do some high level analysis to start driving direction.
Start by referring back to your end outcome. In this case, we wanted to create the most effective marketing campaigns. So we need to understand what is presently working for us. Let’s do this as follows:
Collate all marketing data across various sources (you should now have this finalised from the above steps, but you can’t proceed without it).
Identify success metrics within each part of the marketing funnel. In the case of Mailchimp this would be conversions. For Google Analytics, this would be a combination of web traffic and bounce rate.
Group the data by campaign. If you’ve setup event tracking in Google Analytics and lists in Mailchimp, this becomes a very straightforward exercise.
Calculate a unique metric to rank this data. If you’re collecting data from Mailchimp, conversions are conversions (unless you look at sales too). However, in the case of Google Analytics, you’d want a standardised number that combines these two figures to provide a meaningful insight. Simply, let’s multiply traffic by the inverse of the bounce rate. So, 100 visitors and a 100% bounce rate would become 100 multiplied by 0, resulting in 0 (a 100% bounce rate doesn’t help us). Conversely, 100 visitors and a 25% bounce rate results in a score of 75. You can take this further with new compared with existing visitors, source, and other data. I’ll keep this light for the purpose of the blog.
Sort the data according to this new metric. Ultimately, if executed correctly, this scoring metric will show the highest performing campaigns (in the third stage we grouped the the data by campaign) at the top of the database/spreadsheet, and the lowest performing at the bottom.
You should now have a series of databases (or spreadsheets) that accurately portray the most effective campaigns you’ve executed across a number of different tools and metrics throughout the marketing funnel.
With this data, you’re now in a position to start making data driven decisions. I will identify that the purpose of this blog is not to encourage you to only use marketing data. However, as a department every business has since its creation, it is certainly relatable. These practices can be used across any business department, with the bold terms referring to what you need to do at each stage to most effectively utilise your data.
If you have any further questions, or require clarification, please send me a message. I’m always eager to hear from our readers.