Only a decade ago, if you told the CMO of an SME they would need a full-time data scientist to guide their marketing team, you would likely have been laughed out the building. But today, data scientists are venerated as a pivotal cog in company operations. According to Indeed.com’s chief economist Dr Tara Sinclair, the number of job postings for data scientist grew 57% for the first quarter this year compared to the same quarter last year, while searches for data scientist grew 73.5% for the same period.
Despite the profession’s relative infancy, I’ve worked with companies who genuinely believed a data scientist could alleviate all their commercial woes, and it’s easy to see why. It stands to reason that a company capable of understanding audience behaviour will thrive. But would every business benefit from a data scientist? It sounds like a cliché, but each business is unique, and it’s imperative C-suite professionals consider their company’s needs before investing in a full-time data scientist. Through working with a range of companies in a variety of industries, I’ve drawn up some questions to help guide your data science policy.
Each business is unique, and it’s imperative C-suite professionals consider the company’s needs before investing in a full-time data scientist.
Could the role be divided between your current team?
Writing in the Birst company blog, Chairman Brad Peters explained the data science conundrum by asking which would you rather have at your company: “The world’s greatest data scientist working alone in a corner lab… or data that will make all of the employees of your company one percent more productive?” The answer, of course, is the latter and points to a recurring issue found in businesses across industries. Rather than turning to the latest trend to identify areas of improvement in their content, businesses should strive to integrate data into their overall marketing strategy through regular team sessions and measurable trial cycles.
By allocating individual metrics to relevant members of your team, you can give them quantified goals without the need to invest in a full-time data scientist. I’ve seen companies hire data scientists without once considering how much business awareness they have or how much they really understand of the company, and it rarely ends with either party optimising their talents.
How much data does your company need?
It’s up to you to study your current business output and consider your future content marketing strategy to decipher exactly what your marketing is missing. To many B2C’s, marketing attribution is essential to informing future marketing strategy, but not all businesses require this level of tracking. For many B2B’s today, their client-base for an entire year can be made up of just a handful of key industry leaders.
Most mainstream analytics packages offer a huge array of metrics, when in truth, specific industries will often need just two or three at the most. Studying the metrics behind the people visiting your site every month can guide your overall marketing strategy, but there’s no way of knowing exactly what impact it’s having on the people who will most affect your business. Remember, it’s not how many people you engage with your content, but who.
Most mainstream analytics packages offer a huge array of metrics, when in truth, specific industries need just two or three at the most.
Even for those who want to study the browsing habits of large groups visiting their sites, a full-time data scientist might not be necessary. Consider hiring a data science consultant and have them lay out the groundwork for your upcoming campaigns. If the wealth of data is too much to handle, then you can begin to consider employing a data scientist in a more permanent role.
Does your website fulfil every requirement for your marketing strategy?
For years now, online marketers have based their business around their official website. Of course, that’s hardly surprising when you consider how much technology has altered business over the last decade. Your website is undoubtedly more versatile, unique and interactive than a TV or print ad, but that doesn’t mean it’s going to meet all your business needs. A 2015 B2B Web Usability Report by Komarketing found that once on a company’s homepage, 86% of visitors want to see information about that company’s products/services relevant to them.
In the age of personalisation targeted marketing is quickly replacing the one size fits all approach. Visitors to your webpage don’t want to trawl through reams of information to find the content relevant to them, and the information gleaned from their interaction with your website isn’t always going to be relevant to your marketing strategy. Clients and consumers alike now expect a tailored experience, and those unable or unwilling to provide it are at a real disadvantage. This goes some way to explaining the growth of microsites as a targeted marketing tool.
Visitors to your webpage don’t want to trawl through reams of information to find the content relevant to them
Microsites can be populated with content from your official website, but have been tailored to provide only the information relevant to the client in question. As a technology, they’re quickly gaining traction among B2B organisations as they look to optimise content and limit data fields to the metrics relevant to their particular needs.
Can your data analysis be automated?
The rise of automation is already placing doubts on the future viability of a full-time data scientist. Data science is still undeniably valuable to many companies marketing campaigns, but parallel to this we’ve seen a huge rise in the power of automation and AI. While the few models currently in place using AI to turn data into actionable observations are woefully simplistic, it is more than likely that AI will be capable of handling complex regression models and providing real insights to inform your marketing in the future.
Why should this matter now, you ask? Data automation is becoming ever more entangled with AI, and in turn, deep learning is evolving to become more adept at providing answers to the big questions data throws up. Before hiring a full-time data scientist, consider just how much data you need to process and explore all your options for automation. You could find the analytics that matters most to you can be captured and analysed without employing a data scientist in a permanent role.
Before hiring a full-time data scientist, consider just how much data you need to process and explore all your options for automation.
Conclusion
There can be little doubt the growth in data-harvesting and the subsequent explosion in data professionals has benefitted businesses across industries. But rather than rushing out to hire your own in-house data scientist, see this new data dependence as an opportunity to consider exactly what your company needs according to its core functions. The data scientist may be “the sexiest job in the world” right now, but the hype won’t last forever. As the profession loses some of its novelty, companies will come to recognise that data doesn’t provide a magic bullet to any and all marketing woes. Let’s just hope it’s a realisation that comes sooner rather than later.
An amended version of this post originally appeared in Clemmie.
Source: B2C
No comments:
Post a Comment