Internet marketing metrics
Imagine achieving so much more for so much less! Click here for free tools and know-how materials from the Mud Valley™ strategy & brand marketing community.
Question - are all marketing activities measurable nowadays?
Answer - no, although there has been massive progress in relation to Internet-based activities. This improvement has, in turn, impacted overall marketing behaviour and expenditure. ROI return on investment is an ever more central issue for marketers, driving communications expenditure increasingly online where its impacts are more readily measurable
Key points
Many companies nowadays are classified as ‘pure Internet plays’ – they never produce a physical product or service; everything is providing via the Internet. For these organizations there is a growing mass of techniques which facilitate analysis and optimised response,and are increasingly ‘lights off’ (automatic) in specific functionalities. If you are one of these companies, we recommend that you visit Larry Chase’s Web Digest For Marketers to keep up with the latest techniques.
For those organisations who are still primarily ‘real world’ or ‘mixed plays’, the situation is improving because you can at least measure certain elements of your marketing and business activities successfully, which is why the balance of communication budgets is tipping radically towards Internet-based expenditure. However, the real-world aspects remain as difficult as ever.
In more detail ....
What should you measure?
The first point to note is that Internet marketing should only operate in the context of the brand it is serving, for which it is probably playing one or more of three roles:
- increasing sales
- reducing costs
- enhancing the long term customer relationship
It should also be recognised that it is only doing this in the context of the full brand communications mix.
Pure Internet plays
New systems for automated marketing are being developed for the Internet every day. These systems increasingly seek to optimise marketing budgets by structurally analysing the possibilities and technologies of the Internet itself and your response to those possibilities and technologies, and then executing consequential solutions.
For instance, the whole field of behavioural marketing is shifting from sign-ups and permission-based promotions to cookie-collection. Large aggregators are collecting cookie profiles and selling them as people and systems who evince specifically sought behaviours, e.g. visiting identified sites and interacting with them in a particular way.
The legalities of these aggregations are dubious (the EU is trying to stamp out non-consensual collection) but hard to eradicate among maverick organisations, so they will almost certainly increase on the 'black market' although reputable data collectors will have to comply with the law. People leave their cookies on because they are useful to them in visiting sites and being recognised by those sites, thus obtaining relevant data without having to sign in – therefore their data can be collected.
Within Internet systems themselves, Internet marketers can analyse a wide range of information including:
- Whether their site content is optimised for gaining traffic
- Whether their sites are ‘sticky’ – numbers of page views
- Whether their sites are receiving repeat visits
- Whether site visits are converting to sales
- The most popular pages on their sites
- Links to their sites from other sites
- Mentions of their sites / brands anywhere on the Internet
- Cookie profiles of site visitors
- Profiles of visitors in terms of lead generation sources, geographic location etc.
- Optimised search terms relating to their sites (cost vs. performance)
- Number of visitors to specific content hosted elsewhere on the Internet
- Network growth
- Affiliate growth
- Impacts of Internet and real world campaigns as they impact landing pages (including ‘split testing’ – testing different communication treatments and targeting)
- Competitive site performance
- Customers and channels / intermediaries, including attitudes
Pure Internet plays now have a wealth of data and tools to work with, and that knowledge can be rapidly transformed into consequential action. The more intransigent problem is where an organisation lives in both the Internet and the real world.
Real world and mixed strategies
The fundamental questions of each activity are:
- How much does it contribute to achieving the branding objectives set for it?
- Could the money be better invested in another element of the branding mix?
Specifically for each of the three brand communications mix objectives:
Increased sales
- Does it increase sales?
- If so, by how much?
- Would another element of the branding mix deliver more sales per $ invested?
- Is it a necessary complement to another part of the branding mix?
Reduced costs
- Does it reduce costs?
- If so, by how much?
Enhanced customer relationship
- How much influence does it have on the customer's perception of the brand?
- Does it build, support or undermine the value of the brand?
- Which attributes of the brand does it build, support or undermine
Measuring sales impact
This comes in two parts:
- Measuring the historical impact on sales in order to predict future behavior
- Identifying the best opportunities for the future
There are now statistical and neural network modelling techniques that can link branding communications mix activities to sales outcomes. All you need is the data on which the analytical model can be based. The type of data that is needed is:
- "Sales out" data to the customer, either from your own records, if you deal direct with the customer, or from distributors'/retailers' sales data
- the amount spent on each activity, and when
By linking output (results) to input (costs) over time, the effectiveness of each activity can be assessed, along with the multiplicative effects of running two or more activities together.
What this will not tell you is what would happen if you were to employ types of activity you are not currently using. However, here too there are techniques you can use to assess which activities would be most effective, this time using marketing research. The main topics it would cover would be:
- the role of each activity at each stage of the sales process - awareness building, information gathering, detailed consideration, purchasing phase, after-sales
- the part actually played by each of your current activities at each stage of the selling process
- which techniques would have been most effective, had you used them
Measuring cost savings
In order to assess whether each element of the branding mix is optimised to reduce costs there needs to be an activity based costing (ABC) structure, or similar.
Within activity based costing, you need to identify:
- the major processes within each element of the branding mix:
- customer acquisition
- customer welcome
- customer maintenance
- identifying and anticipating customer estrangement
- customer retrieval
- the activities associated with each process
- the costs associated with each activity
- the drivers of those costs
Activity based costing is also the platform for assessing the profitability of each customer, which is a worthwhile exercise in its own right. Research shows that around 60% of customers are profit neutral (they don't make you a profit or cause you a loss), around 20% actually cost you money every time they place an order, and only 20% of customers (usually long term loyalists) provide with all your profits. Knowing which customers are in which category is critical, because you can then combine:
- an understanding of which customers you should be focusing on
- which processes they value
- the cost of providing these processes
- which elements deliver these processes most cost effectively
Enhanced customer relationship
Measuring the impacts of each activity on customer loyalty is fundamentally the same exercise as measuring its impact on sales. You have a final result that can be measured - loyalty - and now the question is to identify which activities drive that loyalty.
The problem is that loyalty measures are not taken as frequently as sales orders, and therefore there are far fewer data points - a fact that is critical when you are trying to discern the differential impacts of several activities on one outcome. So, neural network techniques, which require frequent iterative data, are probably not the answer.
There is, however a marketing research solution to this that will assess:
- which activities are having the greatest impacts on customer perceptions of your brand
- whether the impact of each activity is positive or negative
- the 5-6 aspects of each activity that drive satisfaction with that activity
- the 5-6 aspects of that activity that drive customers' perceptions of the brand's value
- which brand attributes drive the customers' perceptions of the brand's value
To get these results, you need to measure each interviewee's:
- brand loyalty
- perceptions of your delivery of your brand attributes
- awareness and satisfaction with each activity
- satisfaction with 5-6 aspects of each activity
The steps to doing this are as follows, taking a call center as an example:
- what activities within the full brand communications mix have you run over the last year?
- what do you want your customers to feel after they have experienced each activity? Taking your call center as an example of one activity, what do you want your customers to feel after they have experienced your call center, and what 5-6 aspects of your call center are meant to get your customers feeling that way? These are the aspects of the activity you will be measuring
- what are the 5-6 key attributes of your brand? If you do not know, you need to start a process to define your brand
- how are you going to measure loyalty? Our suggestion would be that you ask:
- You then construct a questionnaire based on:
- Awareness of each activity
- Satisfaction with each activity
- Ratings for 5-6 aspects of each activity
- Ratings for your delivery of your 5-6 brand attributes overall
- Loyalty to your brand
When you have the results, you would use statistical techniques to tie the different areas of the questionnaire together.
This is fairly heavy duty and sophisticated as marketing research exercises go, and would entail significant costs. On the other hand, it is critically important to know which activities have the best potential for building loyalty towards the brand, and how effective, efficient and economic they are.
Imagine achieving so much more for so much less.
We can help you in two ways - we have a mass of smart strategic brand marketing tools, processes and workshop techniques for you to use, and a mass of smart brand marketing agencies as members across the world with niche knowledge and experience to support you thereafter.
Click here for free tools and know-how materials from the Mud Valley™ strategy & brand marketing.
Buy our training packs for tools which guarantee your business growth. For more details, click here.
View our video: “Mud Valley – why we exist video”
For further information, please contact us by telephone at +44 208 123 1438 - Skype: mudvalley (Belgium) - or by e-mail at enquiries@mudvalley.co.uk.
© 2000-2009, Mud Valley ™ brand marketing community.
Related answers
Guerrilla marketing
Viral and buzz marketing