Five things I know about digital marketing (and five things I don't)

I know that… digital requires greater creativity in research design

There is no one-size-fits-all methodology – different types of digital marketing activity require different approaches. We recently interviewed over 23,000 consumers to understand their responses to different formats of digital activity and develop a toolkit for evaluation. It was clear that understanding the objectives of each piece of activity was crucial, as this informs not only the appropriate methodology but also the questions to ask. Consider the bigger picture when thinking about the best approach. Is the communication designed to simply drive clicks, or is it brand-building? Is it a piece of stand-alone communication or part of a wider campaign?

I don’t know… why research has been so slow in exploring the scope for digital evaluation

As a result, analytics have stolen the march and become the default evaluation tool. We took tools and methodologies designed around offline measurement and transferring them online has meant that we, as an industry, have missed opportunities to place research at the heart of the evolution of digital communications and have been too backward-looking in our approach. It is something I think we are in the process of fixing now, but clients have been left with a lack of insight in this area and developed an over-reliance on behavioural measures that tell only one small part of the story. It should be the research industry’s mission to start taking the analytics data and integrating it with our attitudinal metrics to start giving clients the complete picture.

I know… digital communication needs to be judged on more than behavioural metrics

Not all digital activity is designed to drive clicks, and consumers can often do a lot with a piece of communication before the ultimate purchase. Using metrics solely tied to clicking and purchase without exploring softer behaviours and attitudes can be misleading and unfair. We have built a database of benchmarks and case studies that allow us to understand what constitutes a good piece of creative across different formats and objectives. The success of digital activity should be judged by marrying together the behavioural metrics with the attitudinal information collected through sensitive research.

I don’t know… why so much digital communication still focuses on driving short-term purchasing

We have some great case studies from brands such as Adidas, Canon, HBOS and Mercedes that demonstrate the power of digital in building much deeper relationships with brands. The future of digital as a marketing tool lies not just in exploiting its power to reach target audiences more efficiently, but also in building two-way brand conversations with consumers that are more interactive than ever before.

I know… that recruiting people who have been exposed to digital activity for research is a challenge

The digital realm is vast and much of the activity out there is quite niche, so finding the right respondents efficiently can seem almost impossible at times. Again, creativity in design is vital. Traditional tracking will rarely find enough of the right people to respond in detail. It is worth seeing if you can pre-test the activity, pre-recruit respondents or perhaps force exposure to build robust base sizes. We are also placing cookies on panel members, tracking their online behaviour and tagging those who we know have been exposed to activity to re-contact at a later point to interview. This can be tied with their exposure to offline brand touchpoints more accurately using mobile methodologies.

I don’t know… why research isn’t better integrated into the planning of digital activity

Often it is treated as an afterthought or not included in the budget at all. Worse still, the activity is evaluated using behavioural analytics metrics (e.g. impression, interactions, click- through ) alone. This means that the true potential and impact of this sort of innovative activity are not understood or reported on, and we don’t learn from our successes and failures. It can also mean that over time clients are less likely to include this sort of activity as a part of their media plans. Research needs to be considered early on in the planning process.

I know… research should not detract from the consumer experience online

Successful research should involve strong collaboration between the client and research and digital agencies to integrate research in the best possible way. This can help in targeting the right consumers and increase response rates dramatically, particularly if using intercept methodologies or where we want to talk to the same consumers a number of times. Well designed questionnaires incorporating Flash can often become brand-building communication in their own right. And we have found recruiting consumers to blogs, forums or diaries online can be a really rewarding and interactive experience for them.

I don’t know… whether all brands should track social media coverage

For it to be worthwhile, brands or campaigns have to be discussed online in the first place – which is by no means a certainty. There are a lot of companies offering software that tracks social media, and the whole area of online communities and word-of-mouth is incredibly complex. Simple keyword and buzz measurement doesn’t provide the depth of information needed to truly understand its impact. There is a lot of variety in the way data is collected and how online influence is defined. It is essential to work with the best software for the job at hand and to manage clients’ expectations about the fact that consumers may not choose to talk about them online. When consumers do sense brands online, this can happen in a huge variety of ways and we are embracing a number of techniques in our interviewing to capture these.

I know… digital activity can evoke responses that other media can’t

Being inherently interactive, digital moves beyond the simple ‘thinking’ that is the more traditional response of offline media, to involve ‘doing’ as well. This has major implications for the questions you need to ask. Again, it is simply not enough to add a couple of questions to an already packed tracking questionnaire or evaluate digital according to the traditional advertising response model. Digital requires new questions to cover the level of interaction, the depth of exploration, the sharing, the creation, the personalisation and the sense of community that certain pieces of activity are designed to build.

I don’t know… why there is still nervousness about using digital in marketing and research

It is an incredibly flexible medium that lends itself to exploration and trying new things, with efficiency and accountability. I believe that clients, agencies and researchers will begin to collaborate more closely throughout the digital planning and evaluation cycle. Success will depend on this integration and collaboration to bring all data sources together and evaluate the success of a campaign. No one has the perfect solution quite yet, but being flexible and open to reviewing all the information around you will help us get there.