Business intelligence (BI) is the practice of using technology to centralize, analyze and present business data. This helps business owners make smarter decisions about internal processes, like budgeting and R&D.
Guiding principles (aka B.I. on the fly)
A little too busy to scour a business guide? No offense taken. We know it’s common to get caught in the martech whirlwind.
In lieu of a full read, feel free to skim our “abridged” version of how to succeed with business analytics:
Know the language
The data section of the martech landscape is probably the most bewildering — mostly due to the fact that we use its terms interchangeably.
Before anything else, you need to know exactly what BI implies, and that means knowing the language through and through. Understand, at the very least, how BI, marketing analytics and predictive analytics relate to one another.
Take it step-by-step
BI requires a systematic approach — beginning with data cleaning and ending with interpretation of your BI insights.
And like a row of dominoes, imbalance at any one step has an immediate negative effect on the steps that follow. Because of this, you can't afford to rush through BI prep, so take it slow.
Give yourself context
Many BI guides out there only talk about BI in theory. And there's nothing wrong with having that foundational knowledge — It's preferred, in fact. But what will really help you succeed with BI is to have context.
This means painting a clear picture of what questions you want to answer with BI before you get started.
What is business intelligence?
Upon first read, “business intelligence” (BI) usually denotes one of two things:
- It’s a complex concept reserved for highfalutin data scientists, or…
- It’s yet another buzzword touted by clickbaiters and con artists.
Truth be told, both ring true. Because like many other trendy martech terms, “business intelligence” is often misused and confused.
As TechTarget explains, “Business intelligence is sometimes used interchangeably with business analytics; in other cases, business analytics is used either more narrowly to refer to advanced data analytics or more broadly to include both BI and advanced analytics."
Just to keep things clear, this is how we’ll be defining business intelligence from here on out: Business intelligence is the practice of using technology to centralize, analyze and present historic (i.e. “past”) business data.
BI aims to not only summarize what has happened within the business but also why it happened and how the end result was impacted by other factors. BI connects the dots between cause and effect (as shown below) to ultimately paint a more complete picture of the state of the business.
"Business intelligence has been the most powerful advancement in modern marketing; the ability to easily access and interpret data has allowed for a marketing revolution." — Karisa Booth, Director of Service Delivery, Marketing Cloud at RelationEdge
”The use of [BI] tools is multi-faceted: they provide data storage, data integrity and most importantly, reduce the number of manual hours it takes to pull data and compile reports.” — Marc Cerniglio, Manager of Insights & Automation at Chacka Marketing
Tactics of business intelligence
Business Intelligence is an inside-out onion — Instead of peeling layers back to reveal deeper insights, you instead add them on. And it’s these layers that cause so much confusion over what BI definitively is.
In reality, BI isn’t a single tactic or tool. It’s a collection of many different tactics and tools that vary in complexity but, in the end, all trace back to our aforementioned objective: To centralize, analyze and present business data.
Examples of BI applications
For the best understanding of BI in context, let’s examine how it can be used within the different sections of the martech landscape.
Each of these areas brings with it a few common conundrums, like:
- Advertising & promotion: How many publications do we successfully pitch compared to our top competitors?
- Content & experience: What was the ROI of last quarter’s email marketing campaigns?
- Commerce & sales: Which of our retail locations sells the largest quantity of Product A?
- Data: What is the annual cost of our manual data entry efforts?
- Management: What frequency of staff training sessions results in the fewest customer complaints?
- Social & Relationships: What is the average conversion rate of an event attendee?
Balancing the benefits & challenges of BI
The general process for BI looks something like this:
- Data is clean → Data is analyzed → Resulting in BI
- BI results → Results are interpreted → Interpretation guides next steps
Simple enough, right?
Not quite. While the premise of BI is pretty straightforward, things can get messy when you start to execute. Marketers are forced to balance on the BI teeter-totter, with big benefits on one side… and big challenges on the other.
Benefit: Predictive insights offer a competitive edge
History repeats itself, as they say. And it’s a truth that data-driven marketers rely on.
Because not only do they use BI to understand the past, but they also — more importantly — use it to forecast the future.
I covered predictive analytics in detail in the past, and I can shamelessly say what I learned left me in awe. This area of the martech landscape is poised to launch BI into its next evolutionary phase.
"In the next five years, business intelligence for marketers will continue to become more predictive and forward-looking rather than the traditional rear-view mirror into past performance." — Gil Allouche, Founder & Chief Executive Officer at Metadata
For instance, up-and-coming BI technology will allow us to predict changes in our market size (or the emergence of smaller, niche markets) based on demographic and economic trends.
Predictive BI will also help us more efficiently get products into our customers’ hands — in part thanks to enhanced inventory visibility, which only 16 percent of North American marketers have today.
"Our ability to forecast what and how much inventory to order and when will streamline operations and maximize profitability. We will even use business intelligence at a decision-making level, basing strategic decisions and optimizing current marketing efforts on this type of data to maximize returns." — Justine Beauregard, Founder at Mirelle Marketing
Challenge: Bad data can result in out-of-whack trends
Unless you’ve been reading under a rock until this point, it should be blatantly obvious that data is the foundation of business intelligence. More specifically, quality BI is entirely dependent on accurate, comprehensive and clean data. This means:
- The data is as recent as possible, ideally real-time
- The data includes all the information you need to answer a given question
- The data is consistent in its formatting and organization
But research from BARC shows that poor-quality data — also known by the endearing term “dirty data” — has ranked among the top three BI issues every year… since 2002.
Without clean data as a guiding light, there’s no way BI-based predictions can be touted as true.
"When planning marketing campaigns, there has always been a trade-off required between immediacy and accuracy. Traditionally, any reactive marketing campaigns require assumptions based either on past data or the knowledge of experienced marketers." — Jordan Harling, Chief Digital Strategist at Wooden Blinds Direct
"Marketing is built on business intelligence. A marketing campaign that isn't supported by real consumer insight will fail every single time. If you don't have the right data supporting your marketing activity, it's going to get lost in the noise." — Tim Jernigan, Head of Product Marketing at Badger Maps
"Marketing isn’t much without data, and if you trace back anything that’s essential to the practice of influencing buyers, you’ll unveil a fundamental reliance on data — accurate data." — Justin Gray, Founder & Chief Executive Officer at LeadMD
Benefit: Deliberate analysis forces you to rethink spending
More than anything, the goal of business intelligence is to inform decision-making.
How can we reduce costs? How can we increase profit? And how can we minimize issues along the way? These are the governing principles of business, and they’re at the root of all the other questions we ask of BI.
For example, marketers can use regional and federal data to track changes in capex (capital expenses) and opex (operating expenses) — from commercial rent prices and average salaries to IP costs and payroll tax.
Challenge: Faulty data interpretation leads to misguided advice
Yes — You need good data to get BI off the ground. But that’s a small hurdle compared to what comes next. Because in order to do better for your business in the future, you first need to have an accurate, unbiased conclusion as to what went wrong.
But the harm comes when you BI insights at face value and fail to account for other explanations.
Remember earlier when I gave the BI example question, “Which of our retail locations sells the largest quantity of Product A?” Let’s say your BI system tells you “Store #1 sells twice as much of product A as does Store #2.”
Theoretically, there could be multiple explanations for this, including:
- Store #1 may have more square footage than store #2, enabling more inventory to be displayed and thus, more gross sales
- Store #2 may be situated directly next to or near a competing store, resulting in fewer sales for all products at that location
- Economic ups and downs in both store locations could affect customer spending patterns — either in general or for certain product categories
See what I mean? Drawing correlations without all the necessary data is a dangerous game. It’s like pulling a piece of paper out of a hat and calling it your fortune.
There’s a strategy involved in asking questions and interpreting answers with BI. Without that strategy, the insights you’re given could be misleading — or worse yet, completely false.
"In the next few years, it will be increasingly important to differentiate between your standard KPI reports and analysis supported by advanced techniques including AI. These deeper analyses should be held responsible for reporting when your KPIs are off, allowing you to dig into problems and quickly address them as they arise." — Harry Glaser, Co-Founder & Chief Executive Officer at Periscope Data
Current & future trends in business intelligence
What does the future hold for BI and BI solutions — and will it make managing BI any easier? We gathered the data, and a group of in-the-know marketers, to find out.
Waist-deep in data
With over one trillion connected objects generating data as we speak, it’s no wonder we’re burning daylight to bring our rusty tools up to speed. Treasure Data reports that 44 percent of marketers struggle to centralize their data in a single tool as it is.
"Understanding marketing efforts between [online and offline channels has] historically been difficult, but advances in BI such as location-based intelligence is allowing marketers to see the whole picture, even outside of online domains." — Rune Bromer, Chief Executive Officer at Tamoco
Changes to the breadth and depth of BI source data will pave the way to a new interpretation of the buyer persona.
Rather than relying on these potential “stereotypes” as absolute truths, advanced BI will evaluate the presence and influence of dozens of factors — creating a truly data-driven result.
"Traditionally, personas described the typical customer. Modern BI tools allow the marketer to look at a broader range of metrics and score the prospect at an individual level based on actions instead of profile." — Steve de Mamiel, Director at Hostopia
"Business intelligence in the coming years will make marketing much more targeted and personal. Having deep insight into customer data will allow businesses to provide high levels of personalization." — Carmine Mastropierro, President at Mastro Commerce
As the ability for BI to gobble up data improves, so too will the insights it spits out. And this means marketers will be all the closer to that idealized single customer view.
“Once marketers are capable of sending multiple datasets into business intelligence tools, they can evaluate — most critically in a single view — hundreds of thousands of keywords with cost, position, CTR, conversions, etc. This is the ‘30,000 ft view’ of all the things that actually drive the business forward.” — Amanda Orson, Vice President of Growth at Zift
In the past, limitations of BI tools have prevented marketers from utilizing them to their fullest extent.
In fact, BI-Survey.com found that 44 percent of U.S. marketers replaced their BI tool because of its “lack of flexibility,” while 42 percent replaced it because it was “difficult to use.”
But there’s sweet relief on the horizon, marketers. According to Gartner, the BI landscape is well on its way to being more user-friendly, and we should start seeing self-service BI rise to the top.
Combined with AI functionality, these tools will make BI an almost hands-off practice for marketing teams. By 2020, Gartner sees Natural Language Processing (NLP) and artificial intelligence becoming an integral part of most BI tools — 90 percent of them, to be exact.
"BI and dashboard visualization has reached the point not only where real-time feeds populate a dashboard, but AI and ML models can predict customer activity almost as fast as the real-time data is populating those dashboards." — Andrew Wayne Pearson, Founder & President at Intelligencia Limited
"Many commonly used BI tools analyze the past to extrapolate future trends. This is fine, if the outside world remains constant, but that is not reality. New tools, however, [will] be able to better anticipate spikes and declines in demand and changing consumer preferences." — Pawan Murthy, Senior Director of Marketing at Prevedere
"The next step in [the BI] evolution is the incorporation of artificial intelligence into the marketing toolkit, especially a combination of natural language processing and predictive analytics. [These will] replace the pre-defined lead scoring and A|B testing mechanisms of the past and [instead] focus on leads most likely to convert." — Steve Chong, Project Manager & Chief Operating Officer at Projector PSA, Inc.
Business intelligence, truly, is the perfect example of what martech as a whole aims to become. It's a thorough process steeped in scientific methods and presented in a way that marketers can easily understand.
And from what our experts have said, BI is only growing more scientific and more digestible by the day. Contradictory? Maybe.
But that's pretty much martech's middle name.