As a profession, marketing has generally struggled to translate their activities in to numbers that demonstrate the effectiveness of what they have done. Perhaps this has improved somewhat in a digital age where everything is trackable and easily calculated (in fact GA may do most of the work for you). Whilst these are important and valuable, they only tell part of the story and should not be treated in silos.
I’m not going to lie, talking numbers, particularly in a way that matters to senior management, is hard work and it’s the source of many headaches and corrupted spreadsheets. But the upshot is, without this skill you leave your budget and career exposed.
Let me explain.
Assuming you are a digital marketer working in an established organisation, you are functioning as part of a wider marketing team, who in turn is working with other departments in the organisation, such as sales, customer services, etc. All of which influence the generation of actual sales. In order to be profitable the cost to sell needs to be less than the revenue of each sale (margin). Simultaneously, for the organisation to survive, sales (volume) need to reach a certain level, not just to cover fixed cost but to allow for future investment, return a dividend to investors, etc.
The CFO, CEO and other senior leaders, are concerned about profit and in order to improve profitability, they make decisions on what is most efficient and effective; where to invest to get the best return… Return on Investment. At top level, marketing is in competition with every other department in the organisation to secure investment / budget. Marketers therefore need to be able to provide convincing arguments in the format they understand, which is primarily number based.
They are primarily interested in three sets of indicators:
- Business performance metrics and KPIs
This is a historic assessment of what you have done such as ROI, sales funnel contribution, market share, average selling price, lead generation versus targets.
- Diagnostic metrics
This is a snapshot of what is working now and these tend to be the metrics we as marketers are more comfortable with. Conversion rates, response rates, cost to serve, acquisition costs.
- Forecasting indicators
This forecast is built on factors such as brand perception, net promoter scores, size of prospect database, retention rates. These are indicators of future success.
I would also suggest that by developing this skill, you will also help not only promote the professional competence of your department, but it would also benefit your career tremendously. Talking about ‘likes’, ‘click through rates’, ‘conversion rates’ is one thing, but monetizing your activities is going to get the attention of senior leaders.
Starting from the top – Business performance indicators
The mother of all metrics is the much celebrated, but rarely articulated, Return on Investment. Whilst it’s easier to measure for products or services with short buying cycles (FMCGs for example), it proves somewhat difficult to measure for more complex environments such as B2B where a buying cycle could be anything from a day to a few years.
Even then, there seems to be much confusion about how to actually calculate Return on Investment. I’ve seen many examples where marketers have provided a screenshot from Google Analytics highlighting the PPC campaign and Ecommerce Revenue from each source. They then detract the cost from the revenue and divide this by cost. This is wrong.
The correct method is shown below:
Incremental Gross Profit – Marketing Investment
In order to do this successfully, you will need to be able to track where each and every sale has come from. This is why it can be difficult on a campaign by campaign basis, or in B2B environments; although a robust lead tracker within your CRM may go some way to help this.
My recommendation in these instances would be to report on ROI on a yearly, or quarterly basis for the marketing function as a whole.
Aligning metrics with your sales funnel
We have a tendency to report metrics in a haphazard way that can be disparate to the organisations’ sales funnel. The danger with this is that we (marketers) make decisions in silos.
Take a look at the below example of a sales funnel.
(Adapted from Marketo’s Definitive Guide to Marketing Metrics & Analytics)
In B2B industries your sales funnel may look similar to something like the above. It may be simpler, it could be more sophisticated. The point is, that understanding the complete sales process helps you identify the significance of the activity on business performance.
For example, say you recently exhibited at an event and that generated 10 names, whilst also running a LinkedIn campaign that also generated 10 names. On the face of it, a cost analysis may lead you to recommend no longer exhibiting at the event but instead investing more in LinkedIn. However, 12 months down the line you discover the exhibition generated more customers than the original LinkedIn campaign, that would completely change your thinking.
Now I know that’s a very simplistic example, lots of activities would and should have happened in-between that each present a variable for success. What it demonstrates is the need to report regularly on progress of a ‘name’ through the sales funnel.
In this situation you may decide to create a dashboard that looks similar to the below. This could be at source or campaign level – depending on how granular you want to be.
A dashboard similar to this will help you identify the performance of a campaign or source over time and provide the a more intelligent allocation of marketing budgets going forward.
Remember it’s about quality not just quantity.
How are we doing now?
The top-level metrics will take time to filter through and show their return or value to your business. Time is not always something we have, the competition is always watching, our customers’ habits change frequently and trends come and go. So, we have to react quickly.
This is where your analytics and campaign monitoring come into play and this set of data is probably one we are the most comfortable with. This includes
- Cost per acquisition, lead, name
- Website visits, conversions, sales
- Click through rates
- Keyword performance
- Shares, likes, views
- Direct Mail response rates
- Media coverage
- …the list goes on.
The danger, again, is that these are treated in silos. Analysing the traffic generated from a campaign or activity can lead you down the garden path. As mentioned previously, knowledge of the contribution to the sales funnel and eventually the bottom line, is important. It’s pretty normal now for marketers to move past the initial indicators (likes, visits, time spent on page etc) and analyse behaviour or conversions from an activity.
What isn’t normal is the appreciation of cross-channel contributions. Google Analytics goes some way in demonstrating the cross-channel interdependency with their top conversion paths and assisted conversions; but this is primarily digital only.
I’m going to make an educated guess, for those that are working with established organisations. If you were to check the performance (CTR, behaviour and ultimately, conversion) of your organic search traffic, your branded keywords (company name, product names, etc) outperform the nonbranded / generic keywords.
Why is this? I’m going to generalise, but branded searches indicate awareness and interest in the business or product, that has been generated in the past through marketing or other activities. Nonbranded searches suggest the user is at a much earlier stage in the buying cycle, are information gathering or are trying to solve a problem.
With this in mind, why as digital marketers, do we rarely recommend activities to enhance the brand or product recognition? By doing this we remove the competition from the equation before they get to the search engine.
That being said, I appreciate for many me-too or start-ups, this is a future goal. But it doesn’t mean we should shy away from encouraging and improving recognition.
Metrics that help you take action now!
I’m just going to crowbar this in here now as I guess so far, what we’ve discussed requires a bit of data history and reporting procedures. Assuming you are new to the world of marketing, you would have quickly been exposed to the initial indicators of success; ‘likes’, ‘retweets’, ‘comments’, ‘visits’, etc. I’ve included a few others that would be of interest and why.
Organic Search – Click Through Rates
Many businesses / marketers have connected their website to analytics but do not integrate with Google’s free Search Console. Once up and running you can get a better understanding of your website’s search performance on Google. This includes the position, number of times your website has been listed and clicked on by keyword, page, device and more.
Google’s Search Console will indicate the Click Through Rate (CTR) for each. By understanding the search terms people are using, you can improve your meta descriptions and page content for those searches where you have strong presence but little click throughs generated.
In order to be able to calculate the conversion rate for traffic sources, campaigns, pages etc, you will need to configure your website to track conversions. I covered how to do this in my ebook – How to turn visitors into customers.
Once you have done this, you will be able to evaluate quickly the efficiency of a campaign or source in generating a lead or sale. This will help you better allocate marketing budgets moving forward.
Please also note, that many social advertising platforms have their own conversion tracking pixels which either need to be added to the website or via Google’s Tag Manager.
Cost per acquisition
I had a manager who would ask, “if I have $500 left what would I spend it on?”. To help you make that decision, you would need to know what activities had good conversion rates, but also the cost per acquisition. This is simply the cost for the campaign divided by the leads/sales generated.
Video – average watch time
We tend to get obsessed with reach and the number of views. What we should pay more attention to is the average watch time for a video. This fluctuates depending on demographics but also on quality of content.
Say, for example, you had a video that had a campaign message 1 minute into it, but viewers were switching off after 30 seconds, then you need to build this knowledge into future video content to ensure the important messages are conveyed within the attention span of your audience.
Social Media – insights (scheduling), interests
Facebook and Twitter both provide analytics that can be really useful in dialling up the effectiveness of your marketing efforts.
Facebook insights enable you to assess the engagement of each of the posts you make. Through monitoring this you will gain knowledge on what post types, content topics, time / day of posts perform better. Helping you optimise your future social media content planning and scheduling. To gain an even better understanding you can export Page Level Data into MS Excel for a more detailed analysis.
Twitter’s analytics provides the top interests of your followers. This will help guide you when developing future content for your existing audience. Like Facebook, it also presents the engagement rate for a post, helping again with content creation.
Email engagement by device
Did you know that two thirds of emails are opened on a smartphone or tablet? Understanding the differences in open and click-through rates by device will help you quickly identify any issues with device compatibility.
Predicting the future – forecasting
When putting together your marketing plans and budget, it is always important to be able to forecast what you are going to achieve in the future but also to identify any future threats and opportunities that marketing can overcome. Assuming few of us are clairvoyants, how do you put together a well-informed, directional forecast?
Simply providing a year-on-year (or any other period) analysis of marketing performance, may illustrate overall positive growth, the problem here is that this requires you to wait until you have two complete periods to compare. This turnaround may be too slow in order to remedy an issue or take advantage of an opportunity. Annualisation of data can help provide a sensible estimate as to how you are performing against the previous period.
Annualisation takes the result of an action for a time period, less than a year and projects the result over the year. To calculate this, you will need the following formula:
- Determine the time period covered by the result to be annualised
If your website generates 10 leads in 12 weeks. The number to be annualised in 10.
- Divide one year, or 52 weeks, by the time period monitored to determine the number of periods per year.
In this example it would be 52 / 12 weeks = 4.33
- Multiply the result by the number of periods per year.
10 leads x 4.33 = 43.3. So, we would expect 43 leads for the ongoing year.
If you track this at regular intervals you should see this projection increase, decrease or remain steady. You can then compare this against the previous years’ performance to see if you are projected to meet, exceed or miss previous results.
Naturally this requires everything else to remain equal, which it rarely does.
Measurement of the brand is sometimes seen as a vanity exercise. If you take into consideration the importance of branded keyword traffic and the fact that many companies add brand valuations as an intangible asset to their balance sheet; then brand awareness and perception play a very important role in the organisation’s marketing success.
Brand measurements such as consumer awareness, preference against competitors, customer satisfaction (NPS), loyalty and market share provide an indicator as to how well your marketing and organisation will perform in the future. By getting closer to the customer’s needs and wants, digital marketing will also identify new ways to satisfy the customer. Brand research highlights what they consider to be important in the buying decision making process. In B2B this isn’t always cost, technical support, demonstrations, self-serve functions, clear and easy to use websites, play a vital role. By gathering this data, you can then plan future digital developments that better meet their needs.
Where metrics go wrong
In a digital age, there are literally thousands of marketing metrics to choose from.
What is important is to distinguish between those that hold value within the marketing department and which are key indicators for the CFO, CEO and senior management.
Marketing may be encouraged to learn about engagement of a Facebook post, but it would be unusual to present that to the Board.
Try to avoid simply measuring what is easy. In the absence of data to calculate return or revenue, sometimes we default to what can be easily identified. The problem here is that marketing is not speaking the financial language of the rest of the business and risks losing credibility or becoming seen as a cost-centre not a revenue generator.
Finally, know the difference between efficiency and effectiveness. So many digital marketers are focused on efficiency (CTR, CRO, etc.) that they forget about demonstrating its effectiveness (sales qualified leads, customers, retention). At the end of the day you can generate 100’s of likes, names or leads, but if they are the wrong type of people, they are relatively useless.
What you can do now…
The great thing about digital marketing is that the majority of the tracking is already in place. If not, it’s fairly easy to set up or identify where to gather that information. So even if nothing is in place, it’s probably more about organisation and identifying sources than anything else.
If you sell online the process is even easier as you should be able to identify revenue by source and campaign too. The only other consideration is margin, but that knowledge must be held in the business on a product by product basis.
The key thing to do, is to identify what is important to monitor and report on and ensure it is in line with your business’ sales funnel and objectives. A balanced scorecard will help you keep track of these over time. I would recommend starting with around 6 metrics but no more than 12 metrics. You will need to confirm what each of these metrics will be used for, how they will be calculated and the source / owner of the data required to calculate them.