How to use data to plan your digital marketing for 2021

Learning from the past can help you plan for the future. Rest assured I’ve got a bit of experience with this approach as my marketing forecasts have hit close to target on more than one occasion by applying it. However, the last 12 months and the upcoming year has and will continue to throw curve balls and make predicting performance harder, and whilst the past isn't as reliable for forecasting future gains, it can still be pretty helpful to get your started.

Whatever you do, do it fairly swiftly

The defining word for the year ahead should be ‘agility’ - being willing, able and committed to making quick decisions, pivoting your approach or testing and learning as much as possible in order to keep up or hopefully stay ahead of consumer behaviour changes, demand and expectation.

To achieve this you definitely need quick data collection methods and ways of processing the information to make smart decisions fairly swiftly. Some methods I’ve used are definitely time consuming when stitching data sources together, so perhaps a first port of call is to review the platforms and processes to identify if any may slow down progress, and could be improved?

So in this time of uncertainty how can you plan your digital marketing activity?

Data by it's very nature is typically in the past, it's rooted in what's already happened and even predictive analytics needs to look back in order to think forward. So, even in times of flux, past performance can certainly provide some benchmarks from which you can create marketing plans. Whilst many industries have been shocked to the core with a lack of demand or ability to deliver their service, and others have succeeded beyond recognition, there are still patterns in seasonality, business cycles, and demand that will remain similar no matter what is happening in the world.

The patterns may be skewed in strength or timing but they’ll still appear, so it’s wise to know what these may be from looking back at your data. Fortunately we now have a year’s worth of COVID impact data as well to overlay, to gauge just how much these patterns have shifted.

Business performance

At this level I recommend plotting changes over time in revenue, profitability, number of new customers/clients, sales, AOVs etc (whatever you can get your hands on) to determine peaks and troughs in performance. This can help with planning bursts of activity, or determining whether your marketing strategy should focus on filling the gaps or maximising the peak opportunities.

If you’re in travel, generating sales for the shoulder months is a constant headache, and this exercise helps to establish when the business really needs a marketing boost to generate demand.

This analysis can also be used to look ahead if sales in the future are yet to be fulfilled. For example if you have projects in the pipeline, or guests yet to go on holiday with you (but have booked in advance). Knowing your future capacity, and the gap between actual and target can really inform when campaigns need to be launched, when budgets need to be assigned, and possibly if promotional messages need to be activated.

Here is an example of how I’ve looked at business performance data to help with marketing decision making for a brand in the travel and tourism industry.

This graph is a little bit of a brain paralyser with all the colours and data points, but if approached methodically it offers some useful insight into the patterns of booking and holidaying behaviour vs. target capacity for the year ahead.

What this graph revealed to me was:

• how far in advance people booked a holiday and whether this behaviour had altered between current year and past year
• how well capacity was looking against target, and if I forecast based on the typical behaviour of subsequent months, I may be above or below target by the deadline

It was also adapted to filter by location, customer type and accommodation type to identify variances and inform marketing decision making on targeting and messaging.

Customer performance

Customer data can be analysed on a case by case basis, or more preferably in cohorts to understand propensity for repeat purchase behaviour, the profile of highest value customers etc. but also to help identify new segments and targeting opportunities.

Especially in the last 12 months perhaps you have a new customer profile that has organically grown but could be elevated to a new level if you actively communicated with these consumers. I’d definitely advise reviewing your customer profile, even if it’s to confirm that it hasn’t changed!

Direct to consumer brands (DTC) will likely have seen their customer data skyrocket as ecommerce has exploded, and there is a lot more information at their fingertips. Think about analysing your customer data based on lifetime value, one-off purchasers, repeat purchasers, and perhaps lapsed customers, and potentially creating lookalikes from the most lucrative segments, or honing in on the core groups with better targeted campaigns, products and/or services.

Marketing (but more specifically digital marketing) performance

Since I work for a digital agency I’m more qualified to offer advice on the aspects of tracking and measuring digital marketing activity, however I’m also a strong supporter of closing the loop between online and offline marketing wherever possible, as well as understanding the impact of offline marketing activity on online performance. But for planning your marketing activity, I’m staying firmly in the digital marketing camp when recommending how to plan for the year ahead in this article.

Discovering trends

Like business performance data it’s all about understanding trends and gathering as much historic marketing performance data as possible. This reduces the impact of anomalies and can make you appreciate how things may have changed (or not) overtime, perhaps without you even realising it.

When planning my clients’ digital marketing campaigns I jump straight to their website analytics platform. Usually this is Google Analytics. I take a look at patterns in traffic acquisition, device usage splits, location variances and landing pages, noticing % increases MoM and YoY as well as diving into why these changes may have occurred.

It takes a confident planner to appreciate whether stark changes in performance over time is as a result of a proactive decision or activity, or as a result of a macro trend (yes, something like COVID!). The confidence to determine whether the trend will continue if you do nothing (or continue doing what you’re doing) vs. steering an alternative course with a new strategy, approach or channel is what can set you apart and improve your chances of success.

Here are some examples of how I show my plans, taking the key variables and plotting through based on their inter-relationship with one another. These are all based on collecting data from multiple sources (usually other calculations behind the scenes power these overviews), and considering the feasibility of hitting these numbers given what I know about past performance, industry standards, and what the future could look like:

Example 1 - channel expectations and one key scenario switch (conv. rate)

Example 2 - emphasis on scenarios and impact on channel split

Example 3 - marketing planning for different budgets

Plan out scenarios

Right now, I’m looking at clients’ ecommerce performances as we’ve seen phenomenal conversion rates since lockdown 1.0 in March 2020. Rates which I thought would drop dramatically in January, and yet, even before lockdown 3.0 this month, remained fairly buoyant and far higher than ‘normal levels’. So I’ve got to advise on whether these will continue or perhaps fall (and if so by how much) when high streets start to open up again.

To plan for 2021 you need to understand the past, consider the current (and how that differs from past performances), and hypothesise what the future may hold for your business.

The way to do this is to plan for scenarios rather than one single truth, consider what would happen if:

• Demand increases (or decreases)
• Intent improves
• Average order values increase (or decrease)
• The cost to compete (e.g. CPCs) becomes more expensive or cheaper

All of the above conditions happened in one way or another last year, and it’s safe to say that more fluctuation will occur in the next 12 months. ‘Normally’ these metrics shift slightly up / down in any given year, but COVID (and maybe Brexit) has certainly created more sizable differences for a variety of reasons. More so than ever before, planning for different outcomes or uncontrollable variables will help give your marketing efforts some boundaries, if a specific goal can’t be defined.

Data to influence your decision making

To give you an idea of what’s happening in the search world you can head to Google Trends to see if any search terms are making a breakout appearance, or have increased/decreased in popularity over time.

Google’s keyword tool also caters for market dynamics to help plan what ad impressions, clicks or CPCs could be in the future.

Your own business intelligence, and customer data could inform how radical you may be shifting your marketing budgets and strategy and thus the impact that this could cause on performance.

Competitor intelligence

I’ve offered quite a lot of inwardly looking advice, but as well as considering macro factors which may impact performance and forecasts, it’s worth having half an eye on the competition. I recently wrote a piece demonstrating how the tools we use everyday for managing our ad campaigns and SEO strategies can also be used to spy on the competition.

This intelligence could help you plan your attack by offering up new niches, identifying unloved customer segments, or highlighting opportunities to maximise presence more cost effectively because others have left. Whilst you can’t control what others are up to, you can try and make the most of favourable conditions to you when they arise!

Five steps to using data to plan your marketing for 2021

• Be specific about purpose and intention for data use - avoid data paralysis by having a clear goal for your analysis, and ensuring you’re only including information which will help you address your questions or objectives

• Collect the right information - consider everything could be a valuable resource, and look at both internal and external data sources (if you’re not tracking things correctly at present, now is as good a time as any to start)

• Connect the dots - try and build the complete picture, and don’t just rely on the easiest to obtain information. Consider how cause/effect may occur between your previous decisions and outcomes, as well as finding ways to add layers of information to your analysis to make it more usable in a marketing context.

• Plan scenarios - when plotting performance through to likely impact and building media plans, consider a range of outcomes. Risks should be calculated so that both ends of the spectrum can be considered from financial, operational and survival points of view.

• Be willing to adapt and evolve, and act quickly - we’re all human so things don’t necessarily go to plan. We also can’t actually predict the future, so carry your plans with you as sounding boards, but don’t be afraid to adapt if the path you’re on really isn’t going to work for your business.