The main reason why it is hard to capture the whole essence of “the right thing to do” in the data business is because there multiple viewpoints that must be considered at the same time. Allow me to share a few essential dimensions here.
- Industry: Often consulting companies and service providers build their practices around industry verticals. I don’t fully agree with such an approach, as there are many more critical dimensions that I am about to list here. Nonetheless, the industry that a company belongs to — finance, banking, hospitality, entertainment, online and offline retail, publishing, telecommunication, utility, non-profit, etc. — is indeed a very important factor, not just because the business models could be vastly different, but also because the shape of collected data and their success metrics inevitably calls for different types of analytics. On top of the industry breakdown, we must also examine B-to-C and B-to-B cases separately.
- Marketing and Sales Cycle: Even within similar industries, we often observe immensely dissimilar marketing practices. Sales-oriented marketing organizations, for example, require different solutions in comparison to more mass marketing-oriented companies.
- Marketing Channels Employed: I have been emphasizing the importance of the buyer-centric view in marketing, but the reality is that most companies even break down their marketing departments and all related activities solely based on channels. For that reason, channel usage certainly is one of the most important factors, as each channel produces a different type of data and calls for different messaging strategies.
- Target Buyers Lifecycle: Buyers go through different cycles of becoming a customer of a company, and each stage yields a different type of data. For example, available data don’t even look similar for prospecting and CRM stages from the marketer’s point of view. For win-back programs, marketers would have to deal with aged data and third-party data.
- Data Availability and Shape: As I mentioned earlier, some data are messier than others. Marketers and analysts should never give up on any data source so easily, and unorganized datasets certainly call for separate treatments. Plus, consolidating disparate data sources around customers to create a 360-degree view is one of the most important, yet often neglected steps.
- Existing Teams, Divisions and Vendors: Political barriers among divisions are often the main reasons why data initiatives get derailed. Navigating through political mines, unfortunately, is part of the data player’s job, as guardians of real of fictitious domains often become bottlenecks.
- Level of Sophistication of Users: Expert, intermediate and novice users require different types of solutions and toolsets; not because they must go through set courses of analytics in a certain order, but because marketers ultimately get to make decisions with resultant analytics within the confinement of their skillsets. Some may want to put their hands on the data with sophisticated tools, while others may not even want to look at more than a few pages of reports.
- Immediate Pain Points: All companies have pain points, even advanced ones. The key is to fix immediate problems without losing sight of the long-term goals. Too many analytical solutions are nearsighted, and too many projects are designed just for quick results, leading to “Oh, no” moments later. Even for quick fixes, data projects must begin with a clear roadmap.
Yes, analytics consultants certainly have many factors to consider when prescribing solutions. However, from the user’s point of view, data must be something they can access easily, regardless of hardships and headaches that professional data players must have gone through. Much like daily weather reports that we take for granted, we can all imagine weather forecasting is anything but simple and easy.
Data scientists must be always mindful that, from the users’ point of view, data must be:
- Easy to understand and intuitive to all, not just experts
- Small, bite-size answers, not mounds of unrefined information
- Consistently reliable, accurate and effective
- Available most times, not just sometimes
- Easily accessible via users’ favorite devices and channels
This is what I mean by “Smart Data” (refer to “Smart Data, Not Big Data”). There are all kinds of users, business models, challenges, and clean or messy data out there. And navigating through them is what data players signed up to do. Good data scientists should never complain about the users, though some need to be told rather bluntly that they are on a wrong path. As, in the end, even bad patients deserve the best possible treatments, too.