Our Blog

where we write about the things we love

14

Feb

Three common marketing data analysis pitfalls and how to avoid them

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half” was a quote famously attributed to the American merchant John Wanamaker, considered by some to be a ‘pioneer in marketing’.

Three common marketing data analysis pitfalls and how to avoid them

With the increasing availability of data and a bit of help from data scientists to gather, model and analyse it for us, marketers have no excuse to waste half of an advertising budget anymore. We can now have the right insights to know exactly where to invest, which segments to target and what messages to use, ensuring campaigns are a success.

The thing is, while it sounds easy in theory, we still don’t always get it right. Not all campaigns and product launches are a success and even some of the leading, well- resourced brands continue being challenged in defending their positions in the market. The pace of change has reached a new level. In the 60’s a Fortune 500 company enjoyed an average 33-year ‘tenure’ in this highly contested ranking, while that number has dropped below 20 years and at the current pace is going to be soon reduced to about ten.

Many of the new challengers in the market who managed to build their way up to the Fortune 500 have done so because they have learned to derive better value from their data, establish rapid learning cycles and build the agility that simply makes it hard for competitors to catch up.

There is a lot we can all learn from them. To adapt to this new way of working, we need to develop numerical and analytical skills and gain confidence in working with data. What are some of the most common traps we face along the way and what strategies can we use to navigate the marketing data analysis territory confidently?

Perfect is the enemy of good

One of the common traps marketers fall into is the constant quest for perfection. Anyone working in a marketing team has seen this – the 100th iteration of the same visual, the perfect packaging, fully optimised website etc.

I once worked in a company that took over a year to design new campaign visuals. Many design iterations were developed, then thoroughly tested and critiqued with all of our key stakeholders. Before agreeing on the final version, one of the key stakeholders changed and requested that we start again – the design wasn’t good enough. One year and $50,000 later we still didn’t have an agreed design to use.

It’s the same with data. In today’s fast-paced world, we don’t have the luxury to be perfect. Data becomes obsolete very quickly. We either use it to our advantage now, or our rival will use it to enter the market or develop a new product that is a better market fit. It’s important to remember that not all of our data needs to be perfect; it just needs to be good enough to help us make a more accurate or better decision.  

One of the successful strategies I’ve seen working is grading data by categories based on their quality – and labelling them ‘gold’, ‘silver’ and ‘bronze’. Just like the names suggests, gold-standard data has been adequately checked to be 100% reliable and always trusted. It is of high enough quality to support critical business decisions. Bronze-standard data could be raw data from a third party, which might be perfectly fine to model market trends or build hypotheses, which then need to validated by further testing.

This approach ensures that everyone in the business is on the same page and reports based on this data can indeed be trusted.

If the grading of data seems like a daunting task, it’s important not to overcomplicate it. You most likely don’t need a three-year strategy or a budget to grade everything. Just take a small subset of your data and begin (or ask us along to help do this for you). It’s often easier to convince the business that something is worth investing in when they see it adding value.  In the same way, even the most complex data warehouse could start its life as a proof of concept (POC) and be gradually enhanced as the marketing team build their confidence in working with data.

Not asking questions

Critical thinking is one of the most sought-after skills in marketing management, and it will continue to grow its importance with the increasing availability of data further.

What are some of the questions we should be asking when designing a data strategy for marketing? First of all, it’s essential to remember why we are doing it. Are we looking for ways to identify opportunities for new product development, gain insights into our key segments, reduce churn or improve engagement with our messaging or conversions? How is our data currently collected? Can we trust it? How many systems hold this data? Do we have a single view of our customer? Is there anything more we can do with our existing data to get more value out of it? If we get more value out of it, does it help us meet our KPIs and does it ultimately help the organisation progress against its goals?

The thing is, as data becomes more accurate and more readily available, and online marketing more complex to understand for those who are not subject matter experts, it becomes easier for marketers to use data to force it into the story they want to tell.

All of us have seen this – a marketer seeking budget for a campaign that is backed up by highly accurate data on forecasted reach and conversions. Who wouldn’t trust it if it comes from a media agency or Google Analytics? Unfortunately, not all data, whether it comes from internal or external sources, show the full story.

Too often, the high accuracy of data creates a false expression of a more reliable truth and could lead us to reach wrong conclusions. Asking questions allows us to nurture critical thinking and test hypotheses before reaching conclusions.

The business, brand and shareholders will all be better for it.

Failing to make actionable conclusions

One of the best pieces of advice I’ve received as a marketer was: “Looking at reports isn’t the same as finding insights and taking action”.

It was in reaction to a sales report one of my colleagues continued bringing to sales-marketing meetings. The report showed a summary of historical sales data for one of our overseas markets. Every time we looked at that report and discussed the trends, but given the marketing budget was approved a year ahead and all of the campaigns locked in, there was no appetite to take any data-based action that would change the path we were already on.

This colleague was adept at challenging the marketing team, but he was right. It’s essential that before we look at reports or start analysing data, we have a crystal-clear view of what questions we need answering. If we knew those answers, would we want to change anything, and would we have the authority to do so? If not, would anyone we report to have the power and the appetite to change? If the answer is still no, then we are most likely wasting our time.

Just like the previous point on critical thinking, this one has a lot to do with company culture. In results-oriented cultures, where work environments are outcome-oriented, and people have incentives to win and achieve their goals, we have no choice but to seek insights that will help us take action.

We have to collaborate with IT to find ways to connect our disparate data sources from across the business to get those insights. In such cultures, marketers’ KPIs are not limited to ‘delivery on-time and on-budget’ – what counts is the ability to help the business progress against its goals.

This blog is part of the #datareimagine series. For more experts' insights, clients' experiences and to download our datasheets, click the banner.

For more experts' insights, clients' experience and to download our datasheets, click the banner #datareimagine

Posted by: Katerina Thomas, Client Manager | 14 February 2020

Tags: Marketing, Power BI, Data Intelligence, Data Analytics, #DataReimagine


Top Rated Posts

Blog archive

Stay up to date with all insights from the Intergen blog