Analytics: 5 Key Steps to Generate Value

I recently read an interesting article from the MIT Sloan Management Review about how their research revealed a significant “information transfer gap” or chasm between capturing the data and getting people to act on the insights from the data. Their survey showed 65% of organizations felt they were effective at capturing data, but just 46% were effective at disseminating information and insights. While the research primarily focused on this issue from a big data perspective, this has been a recurring challenge for all forms of analytics—big data or small data, online data or offline data, structured data or unstructured data, etc.

As you consider how you can more consistently translate raw data into meaningful insights, remember your mission won’t be complete if you are only effective at sharing information and insights. In order to get value from analytics—and all the data that’s being collected—your company has to be effective at acting on the insights taken from your data. As I like to emphasize—no action, no value. In fact, there are a number of steps that need to happen after you’ve collected your data before you can create any value from it. Companies can sit on mountains of data that may be viewed as “valuable” but never actually generate an ounce of value from their data if they’re unable to use it effectively.

Analytics Path to Value: Set ‘Em Up and Knock ‘Em Down

In my eight years of web analytics consulting experience, I’ve seen many large companies struggle to generate tangible value from their online data. I like to compare the whole analytics process to a line of dominoes, where achieving value—the last green domino—is dependent upon completing or knocking down a series of key steps. Along this path to value, a missing or misaligned domino in the line can impede an organization’s ability to generate value from its data. Based on the nuggets buried within their data, an organization could streamline processes, lower costs, increase revenues, develop new innovations, or establish a competitive advantage, but not if it can’t effectively knock down every domino on a consistent basis.

If you aspire to be an analytics action hero or want to cultivate a data-driven environment at your company, you’ll want to understand each step or domino in the Analytics Path to Value.

1. DATA

Regardless of what type of analytics you’re focused on, the first step is to collect useful and trustworthy data. Without data that is relevant, complete, and reliable you will not get very far with analytics. Data is the basic blocking block for all your subsequent insights, optimizations, and innovations.

2. REPORTING

After your organization has collected its data, this raw material usually needs to be cleaned, converted, combined, or organized before it is truly useful and accessible to business users (might involve an ETL process outside of digital analytics). Effective reporting provides the broader base of business users with an important lens into the performance of the business. Through different vehicles such as scheduled reports, scorecards, and dashboards, business reporting transforms data into information that can be used to monitor how different parts of your organization are performing.

3. ANALYSIS

While reporting might be used to spot potential problems or opportunities, it will rarely initiate any action on its own. A deeper exploration of the reports and underlying data is needed to convert information into meaningful insights, which can be used to better understand and improve business performance. Analysis is not just about mining the data for insights but also sharing key findings with stakeholders and recommending options to address a problem or seize an opportunity. The potential impact should be monetized so decision-makers can weigh the various opportunity costs. One of the biggest problems in analytics is too many organizations confuse reporting with analysis and fail to move beyond just providing information.

4. DECISION

Before insights can be translated into action, a decision maker (or group of decision makers) typically needs to decide whether or not to act on a particular insight or recommendation. No resources will be assigned and no changes will occur until a decision is made. A decision may require additional data or further analysis before an executive feels comfortable with moving forward. The final decision may not align with the analysis recommendations, or the execution plan may be altered (e.g., project scope increased or decreased).

5. ACTION

After a decision has been made, the next step is to plan an appropriate course of action and execute on it. If a company is unable to successfully execute on the valuable insights coming from its data, the execution team needs to be held accountable. Taking action can include testing different approaches before settling on the best one. Not every action will achieve its expected return. However, a data-driven organization will learn from both its failures and successes so that it is constantly refining and improving its approach over time.

When all of these stages are aligned and knocked down, you can generate significant value from your data. However, organizations frequently don’t make it all the way down the path and may end up questioning the value of their analytics efforts. If that’s the case at your company, you need to determine if you have a gap in your firm’s path to value. For example, many companies place a heavy focus on reporting but very little emphasis on actual, deep-dive analysis. Their understaffed analytics teams have no bandwidth to focus on anything besides implementation and reporting. Running into the “information transfer gap” mentioned in the MIT article, these companies get stuck in Setupland and don’t set foot in Actionland, where the real value is created.

In other cases, you may appear to have all of the dominoes in place but a misaligned domino may be the source of your woes. You need to investigate the misaligned domino to see what factors might be throwing it off course. For example, you might run into one of the following scenarios at your organization:

  • Incomplete or insufficient data is the real reason why senior managers are unable to move forward with a proposed optimization.
  • Inadequate training of end users leads to a low adoption of the current analytics reports.
  • Turf wars and internal politics prevent key stakeholders from agreeing on a decision even though it would be in your company’s best interests.
  • A business team feels threatened by a proposed change and sabotages its success during the implementation phase.
  • An IT team misunderstands the prioritization of a tagging or optimization project and postpones its deployment.

The key is to isolate where problems are originating and understand what other less obvious factors may be contributing to them. With the path-to-value concept in mind, you can identify areas where better processes, added personnel, or technology can streamline and improve performance. New developments in predictive analytics can significantly augment analyst capabilities, and automation can streamline inefficient, manual processes. For example, decision automation capabilities (algorithms similar to paid search bidding rules) can handle micro-optimization opportunities that would overwhelm traditional decision-making processes.

In addition, it’s important to note the path to value is not a once-and-done scenario, but a process you’ll repeat over and over as you seek to optimize different aspects of your business. Hopefully, with experience and discipline it will get easier over time for your organization to turn valuable data into actual value. As you’re able to successfully drive more value with your analytics, don’t forget to evangelize the heck out of your data-driven successes to build momentum across your organization.

This entry was posted in Optimization and tagged , , , , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *