Win today by developing the skills to deliver growth during inflation, with up to 50% off memberships in August!

Explore our Digital Marketing Strategy and Planning Toolkit

Remixing your marketing mix online

Author's avatar By Expert commentator 18 Oct, 2011
Essential Essential topic

How to apply the marketing mix to develop online marketing strategies

We often hear that the concept of the marketing mix isn’t so useful any longer in this era of customer-first. But I believe it is still highly relevant today as a framework to develop digital marketing strategies. In this post I’ll take a look at situations where I have found it useful to develop and refine customer propositions.

Why the online marketing mix still matters

The marketing mix is a conceptual framework, and as such it is useful since it enables a common language to be used in the planning, execution and measurement of a number of coordinated activities that deliver the desired marketing outcomes.

Customer centricity demands that organisations becomes a lot better at collecting and reacting to customer insight and adapt their offering to best suit an ever growing number of narrowing customer segments – ever approaching the ideal of the completely personalised product. As marketplace complexity increases, it is crucial to be able to rely on tried and tested concepts like the Marketing Mix.

I use the basic components of the marketing mix (such as the Four Ps) at work daily, and as such for me the Marketing Mix remains a practical tool. In a digital environment routinely identifying and profiling individual consumers over time and adapting to their needs and wants, the 4 Ps become elastic. Is my website a Place, or is it an integral part of the Product? When more efficient digital channels directly influence my ability to Price, do Place and Price become synonymous?

This elasticity needs to be managed effectively to avoid the pitfalls of too rigidly applying the model. One must always challenge the validity of every tool in the digital Marketing armoury; adding, changing and discarding as business and customers evolve. Proposed evolutions of the marketing mix concept (such as the Four Cs) may ensure it remains relevant today. My personal view is we’re still a long way away from seeing the end of its useful life.

Examples of applying the marketing mix online

In my view, the key insight is that digital media can contribute to every element of the mix. Therefore we must avoid a narrow categorisation of digital as contributing solely (or even primarily) to a single component of the mix.

While I haven’t come across an organisation that fundamentally disagrees with that view, some organisations find it easier than others to put it into practice. I have worked in organisations where the mix is embedded in the corporate structure with separate Pricing, Product, Distribution (Place) and Marketing (Promotion) departments. Embedding a digital mindset across those silos can be a daunting task.

I’ve experimented with channel pricing, as pricing is a critical driver of conversion and business value in a services organisation. Here are some of the ways I’ve tested the price/product component of the mix and lessons learned:

  • Straight Online Discounts have proven difficult to justify. Online discounts aren’t valued by customers (in the age of price comparison, they focus on the total price, rather than its components) and often do not reflect a real lower cost to the business (lower costs to sell and serve are offset by lower transaction value and lower retention rates).
  • Using channel data as a pricing factor has proven more successful. As historical data is accumulated, it is possible to really offer competitive prices to those customers identified as high value at the point of application. An accurate value/propensity model can use the wealth of information available from digital visitors (geography, visit trigger/campaign, past visits, customer history, etc) to drive truly personalised pricing. In this example, price follows place and both price and promotion reflect the individual customer.
  • Personalising the product offer. I’ve successfully extended the data-driven approach to other elements of the mix: dynamic packaging (the creation of a personalised offering from the basis of a modular product) has proven successful many times: at easyJet we built a product that included car hire recommendations based on a predictive model that took destination, seasonality and party size as inputs – increasing car hire uptake very significantly. More recently I’ve applied the same insight-driven dynamic packaging approach to RSA’s Central and Eastern European businesses, increasing sales of optional covers and add-ons very significantly.
  • Varying the base product. Further opportunities exist around selecting which default base products to present: are you more likely to want a cheap energy tariff that tracks price rises or a fixed price deal that ensures protection against future rises? What we know about you from your digital “footprint” may provide the clues we need.

Testing value propositions online

I have also used online channels to test propositions in a couple of ways. First you can assign propositions randomly to visitors on first arrival to test interest/sales. I would typically run this against a large control group (being offered the current main proposition) to both protect commercial results and detect the effect of any external influences that may be otherwise wrongly influence the experiment. This approach can be extended beyond the site, via randomised allocation of marketing messages on Display, Search etc to measure a proposition’s attractiveness. It’s important to test all aspects of the proposition: a very successful proposition at attracting interest may convert badly if it can’t be priced at a level that matches customer expectations.

Second, you can provide a modular proposition platform and allow customers to combine proposition elements. We can easily analyse popular combinations as well as secondary correlations such as the propensity to add a certain ancillary product to a particular proposition configuration, and understand the compound impact on profitability, retention and advocacy from what we know about each of the modular components in isolation

I’ve found that proposition testing rarely fits neatly an A/B scenario, with tests quickly developing into complex multivariate experiments with a significant number of variables. It is important to ensure the tests are solidly planned, rigorously executed and statistically significant. Free tools like Google Website Optimiser provide information as tests as run that help assign a confidence to a temporary result. However, it will do nothing to prevent badly designed experiments from yielding wrong data. In my experience, the only way to ensure valid tests and improved business results is to bring in the best analytical brain you can find. Analytics is the first role I fill when I build a digital team, it’s that important...

Author's avatar

By Expert commentator

This is a post we've invited from a digital marketing specialist who has agreed to share their expertise, opinions and case studies. Their details are given at the end of the article.

Recommended Blog Posts