Before exploring big data, there are far more fundamental foundational steps to take.
"Big data" is one of the most used buzzwords in the modern marketing vernacular. No matter if I meet blue-chip multinationals, startups, or am going to conferences across the globe, everybody seems to be talking about big data. But let's face it - while complete access to big data could help most companies; it is simply not feasible for the majority. Prior to exploring big data there are far more fundamental foundational steps to take. For many, big data seems to have become a synonym for a utopia of full information, which of course can put any marketer at ease.
From my experience, most companies pursuing a big data strategy, or seeking advice on how to incorporate such, should ask for advice first on how to think about data. I have had the privilege of working with and building some of the largest Internet companies on the planet. In doing so, I realized that data is a mindset, and having the right mindset is more important than having fancy software.
I commonly see the following six mistakes in the way people think of data:
1. Isolating traffic (cost) generation and revenue generation
Many companies, even agencies claiming their pride in digital, are stuck in a world where an understanding of integrated analytics is an unknown thought. Isolating the traffic generating segment of your marketing (search, display, and other efforts) from the revenue generating (on-site optimization) is a common mistake. It is a common (but unjustified) excuse that the marketing team might not be in control of on-site conversion as they do not influence design and pricing (if e-commerce). However, landing page selection, source of traffic, etc. can still help get a better performance. For me, understanding the conversion to the site, without understanding the conversion on the site has little value. Understanding relative efficiency of keywords or banners on specific placements is the cornerstone of efficiency, and is a feasible concept without fancy software and "big data."
2. Allocating budget based on gut feel
Bad digital marketers base their budget allocation on gut feel without a true understanding of relative efficiency. When companies show me their budget allocation, I am often surprised to see how even the numbers look. Is it really so that search ads are exactly 50 percent better than display ads? If not, then why allocate the budget so? Such may provide round, even numbers, but it is rarely the most efficient approach.
3. Separating media buy from execution
A separation of media buy and execution/optimization rarely yields the best return. As efficiency is linked to price and as price of performance marketing ultimately is linked to quality, the relative efficiency of channels or sub-channels (banners, keywords, campaigns, etc.) is defined by the optimization. For this reason, it is impossible to make the best decision of where to buy your media from without having a full understanding of the performance and scalability of the channel relative to the rest of the marketing mix.
4. The fear of having too many keywords
While there is definitely a diminishing return to having a large keyword portfolio, most companies are not even close to seeing a significant drop. For 99 percent of all companies with an ambitious digital plan, counting the keywords in hundreds or a few thousand tells me that they think about keywords in the wrong way.
If you believe that the way to optimize a keyword portfolio is by going through every single keyword one by one daily or weekly to look at the CPC, revenue, bounce rate, etc., then you probably think about it the wrong way. You should not be looking at individual performance, rather, you should be looking at the bigger lines and isolate the negative and positive outliers.
5. Believing that digital brand building is about number of impressions and CPM only
Digital often has a very performance-based feel to it. Many marketers mistakenly take this as those channels are an ill use of their marketing budgets, if they are looking to build a strong brand rather than to sell products online. Nothing could be further from the truth. I once advised a large car manufacturer on how they should build their brand online. While they didn't foresee a chance to sell their cars online any time soon, they did realize that a visitor to their (offline) showroom, who had already spend some time researching on their website, was way more likely to buy the car in the store. We therefore developed a KPI based on time spent on site.
6. Not understanding assisted conversions and multi-touch
In the offline world, brand building has been a major topic for many years. When the world went digital, and tracking of orders back to the source became an option, many companies forgot what they did until the day the Internet was made commercial. The easiness of tracking the last click before a conversion was so convenient that marketers stuck to this attribution when selecting their channels.
Attributing value across channels can be very difficult - especially without the right tools. I hear many marketers complain that they cannot do this, because they do not have very sophisticated software. However, there are many ways to get an idea about if banners or other channels are playing a major role in your click chain. Using this knowledge to attribute some value is a very cost effective way to think about cost attribution for most companies. You may not get it 100 percent right, but getting it 70 percent right is still far better than completely disregarding the effect (which is how most people look at it).
I am not saying that big data is not good. It definitely is for some, but for most companies that do not measure their revenue in billions, there are easier and more cost efficient ways to think about data. For transparency of efficiency, there are already plenty of free tools that allow you to get a better overview. I have seen many examples of simple plug-ins to excel or other basic software that has revolutionized the transparency of data and in so doing provided the opportunity to increase marketing efficiency dramatically.
This is a guest column by Stefan Bruun, managing director at Lion&Lion, a SEA-based digital agency leveraging world class best practice knowledge and technology, to help clients jump light years ahead on the digital marketing knowledge and execution curve. Their portfolio of clients include global blue-chip companies as well as startups.The data-driven team builds on the lessons from countless hours of marketing execution and learnings from spending more than US$300 million online over the last years. He was formerly the Asia Pacific chief marketing officer and the leading expert on online marketing for the largest Internet venture company in the world, behind e.g., Zalora, Lazada, Groupon, called Rocket Internet. While at Rocket he was the biggest client of Facebook and Google worldwide.
Image on home page via Shutterstock.
Learn Digital Marketing Insights From Leading Brands!
ClickZ Live Chicago (Nov 3-6) will deliver over 50 sessions across 4 days and 10 individual tracks, including Data-Driven Marketing, Social, Mobile, Display, Search and Email. Check out the full agenda, or register and attend one of the best ClickZ events yet!
Hong Kong, October 21-22
London, November 13-14
San Francisco, November 13-14
London, November 18-19
IBM Social Analytics: The Science Behind Social Media Marketing
80% of internet users say they prefer to connect with brands via Facebook. 65% of social media users say they use it to learn more about brands, products and services. Learn about how to find more about customers' attitudes, preferences and buying habits from what they say on social media channels.
An Introduction to Marketing Attribution: Selecting the Right Model for Search, Display & Social Advertising
If you're considering implementing a marketing attribution model to measure and optimize your programs, this paper is a great introduction. It also includes real-life tips from marketers who have successfully implemented attribution in their organizations.
October 23, 2014
1:00pm ET/10:00am PT