Analytics in marketing is so overused. It is a popular buzzword alongside, big data, machine learning and AI (artificial intelligence).
Although the importance of data has never been more front of mind, the growth of the information we collect has ballooned so much, with some saying data is eating the world. This has resulted with a lot of us in marketing paralysed to do anything or just playing safe.
Here we will look at how you can get the most from your marketing data, to ensure the insights from your data see the light of day and that they get actioned.
Key takeaways - Analytics in Marketing
Top 3 tips for Analytics in Marketing
See our checklist
1. Ensure you have a clear marketing objective that aligns to business a goal to give your analytics in marketing purpose.
2. Resource your business with the right mix of people. Check you have 40% of your resource focused on interpretation and translation of findings into actions…not reporting.
3. Have clear owners in your business who can action and execute quickly on those interpretations, to move the business closer to its goal.
Analytics in marketing trends
More data, technology and skills, but less action
Mega data growth
With the huge growth in data we have seen a spike in information and the ability for marketers to know where to focus. This growth has lead to a lack of understanding about how to engage their visitors, prospects or customers.
We have all heard the growth stats like these:
- There will be 4.8 billion internet users by 2022, up from 3.4 billion in 2017
- 200 billion devices are projected to be generating data in the Internet of Things trend, by 2020
- 90% of all data in existence today was created in the past two years
However the growth of data doesn’t mean we are more able to understand and action this information. For most it just means we have more data.
So be really clear about:
- what data you want
- why and what action you’ll take on it
- align data captured to the skills and tech you’ll need to action it
Demand for technology & skills
So with the growth of data, there has been a big increase in the adoption of technology and the skills needed to analyse data.
Growth in Technology
“In The First Golden Age of Martech, the US marketing ecosystem grew 50% faster than GDP”.
Another example is the growth of AI (Artificial Intelligence). By 2021 the Artificial Intelligence software market is set to be nearly $35bn which is 245% growth from 2018.
Growth in Skills
So what about the growth of skills? There has also been a rush to recruit the right talent to help businesses wrestle the vast volumes of data.
So with the increase in technology and skills surely there has been an increase in productivity and use of data to create business growth?
Reports do not mean action
We have looked at data growth and investments in technology and people, but this hasn’t resulted in more effective use of data in organisations.
In one research study, 37% of respondents believe they analyze less than 20% of consumer data available to them.
So why is this?
Debra Bass talked about InfoObesity when she was at Johnson and Johnson and Kim Whitler. wrote an article on Why Too Much Data Is A Problem And How To Prevent It.
In an article “Companies Are Failing in Their Efforts to Become Data-Driven” by Harvard Business review, they found a lot of businesses struggled with their data.
Given the increased investment in Artificial intelligence it was amazing to see 77% of executives report business adoption of Big Data/AI initiatives is a major challenge, up from 65% last year.
In the findings from the research by New Vantage Partners, the reason for lack of adoption was not because of from technology obstacles, with 7.5% saying this was the cause.
The major issues was people (skills) and process (tech) as the barrier with 93% saying was the cause.
Don't forget to check out our book - Your Number's Up!
Getting a grip on Data and Measurement to accelerate your Direct to Consumer ecommerce sales
With over 280 pages our book is full of actionable material and Avinash Kaushik has kindly written our foreword.
We have created a series of tools to help you understand, measure and grow your ecommerce business.
Defining your digital measurement maturity
4 x Cheat sheets for improving your ecommerce measurement
A full glossary of over 60 metrics, understand what they mean and why they are important
Practical scenarios for diagnosing your ecommerce website
Common campaign measurement mistakes and challenges and how to overcome them.
Why analytics in marketing fails
See our video guide to resourcing your organisation for analytics in marketing.
Surely we just need to employ more data scientists then?
Given the research above, the answer might point to the recruitment of even more data scientists to analyse more data.
If you look at demand for skills in analytics in marketing, you can see there has been a huge growth over time.
Look at this example for “data scientist” in the job category on Google Trends
However the volume of data captured and stored has been outstripping the capability, the training and the recruitment of people required to take action on it.
The gap to inaction is growing ever larger
The big challenge in businesses isn’t the ability to handle or process the data.
It isn’t having the right skills in house to analyse the data.
The challenge is being able to understand, interpret and then action the information. Moving from analytics in marketing to ACTION in marketing.
Furthermore having an organisation that is informed by data is key. This means senior leaders need to understand the data they have available to them, what’s useful and what’s not. Organisations need data savvy leaders at the top table and If not available then data savvy senior leaders need to brought on board.
Why having 50/50 resources means your analytics in marketing will fail
In the image to the right, you can see why there is a challenge in actioning analytics in marketing.
A lot of organisations work on a 50/50 resourcing model. They may have plenty of analysts and an equal number of those who work in and on the business. But there is a gap.
The gap is the interpretation AND translation of findings provided by the analysts. The people working ‘in’ the business may not have enough experience of interpreting what the next best action to take is.
With resource sitting between the analyst and the business, these findings can then be turned into tasks that be reviewed and actioned by the business.
How to ensure your insights are actionable, using the 40/40/20 rule
To help combat the lack of action, we need to introduce the 40/40/20 rule for resourcing analytics in marketing projects.
As the complexity of business, marketing and sales increases, you need resource that can interpret the next best action(s) for the business, working with the analyst and the business/function owner. It might be channel, content, offer, product, bundle, partner etc. It might be campaign, it might be tactical. It might be brand marketing, performance marketing, lead generation.
The gap needs to be fulfilled by the role of the data interpreter. Not just an analyst.
The role of the data interpreter is absent in many organisations. Given the lack of data driven knowledge in executives, as highlighted above, this role is more critical than ever.
Need to ensure your analysis sees the light of day?
Download our 7 step how to guide to understand what you need to do to get your analytics in marketing actioned.
Each step will guide you through what you need to do and has examples to bring the tasks to life.
At the end you will understand how you can translate your findings into actions for each team member.
OVERALL APPROACH: The 7 steps
In the overall approach you can see there are 3 key owners of tasks from the 40/40/20 model above.
It all starts with the business defining the goals.
This then moves onto the analysts to collect, process and analyse the data.
It then heads to the interpretation and translation of the data by the business.
The output from here is then passed on to marketing, IT or operations to put the actions in place.
STEP 1: Set your goals (owned by the business)
Define the purpose of the analysis. What are you trying to get visitors or customer to do? What are trying to better understand?
This goal will set the direction for everything so ensure you are setting a goal which is not only achievable, but is also relevant to how your users behave, the customer experience.
TIP: A good example is trying to improve conversion rate. When in fact the root cause might be a high bounce rate. So improving your content on your website might be one route to improving bounce rate as an outcome. So analysing the bounce rate and engaged users could be the initial task.
STEP 2: Collect the right data (owned by analysts)
Ensure you have ALL the right data you need to be able to perform your analysis based on your goal. You don’t need to capture everything.
Quite often this information is not available or being recorded on websites. The core actions from users are captured, such as adding to cart or checkout completion, but it is sometimes the softer or non-transactional engagements which are important.
Ensure you can capture all engagements your visitors could have with your website or App not just the outcomes, as these will only tell you what NOT the why.
STEP 3: Clean and prepare the data (owned by analysts)
Prepare the data ready to be analysed. Often overlooked but this can take up to 80% of the analysis time.
Make sure you have a data set which is clean and can be easily analysed. Things like different currencies, different date formats can all make life tricky. This might not be the most exciting part of the analysis, but poor quality data will mean no analysis or analysis which is meaningless.
STEP 4: Analyse the data (owned by analysts)
Analyse the data to determine what has happened. The output from here will vary a lot depending on how you can implement the findings.
This might mean you create a segmentation of user behaviour to group together the best performing visitors and their behaviours.
You might use this to inform your machine learning product recommendations tool on your website.
You might build the predictive model to deploy for targeting in your email campaigns.
Whatever the outcome the key thing to remember is the outcomes need to be presented to the business for interpretation. So before jumping to an answer it can sometime be best to understand user behaviour first.
STEP 5: Interpret the findings (owned by the business)
Interpretation is all about understanding what the data means and what we have to do, to get it actioned within the business.
Start with reviewing the findings from the analysts and understand what this means and why it is important. Identify the actions required for the business and prioritise the actions so you know what things you need to look at first.
STEP 6: Translate into actions (owned by the business)
This step is quite often overlooked.
A lot of people talk about actionable insight but in fact what is provided is factual observations.
To make something truly actionable we need to be able to take the interpretation and turn this into an action for the owner who is going to implement it.
Outline what is required for each action for the owners in the business, in a way which is easily understood by them.
STEP 7: Execute your Actions (owned by marketing, IT or operations)
In our final step, it is all about action. We need to be executing the outlined actions in priority order. Once the action has been deployed it is best to ensure this is measured and the results analysed to understand if it had a positive impact on the business. This is often a technique used in AB testing.
In summary: To ensure we have a successful analytics in marketing programme, run through the 7 step process, having clear owners for each step.
40% of the effort should go towards analysis, but as much effort should go to understanding what the analytical results tell us and what actions owners in the business need to do.
From here it is all about getting the actions into market, by each of the owners within the business.
Want a Real World example?
Bringing analytics in marketing to life with a real world example
Putting the 40/40/20 rule into action
To highlight how analytics in marketing can work we are going to look at an Ecommerce example.
You can see here the middle 40% (the business) sets the overall goal for the analysts.
In this example the challenge is to increase revenue, as this business has seen revenues decline month on month.
Interpreting the Ecommerce findings
Here the analysts have looked into the data and highlighted on the left their findings.
The key task here is to interpret the findings, otherwise this will be just be analysis. We want to be able to look at the data and see what this means for the Ecommerce business and what actions we would need to be taking.
Translating the Ecommerce actions for the organisation
Based on the interpreted findings, the team now look at translating these into marketing actions.
These actions are the split by Ecommerce role. In this example the team has the following roles:
Traders: Responsible for revenue
Merchandisers: Responsible for getting customers to the right products
Media buyers: Responsible for delivering quality and profitable traffic
Content Creators: Responsible for creating engaging content
Deploy the Ecommerce actions to grow revenue
Each of the actions are deployed by the different Ecommerce team members.
Every action which is put into market is then measured to understand the overall impact it had. Those which were ineffective will be identified by the analytical team and re-interpreted, re-translated and new actions put in place.
The key thing here is timing, you may not want all actions to take place at the same time, especially if you’re testing something to better understand it.
Below is some recommended content to help you with your analysis in marketing. This content is mixture of content, tools or our book to help you get more out of your digital marketing and data in your business.