Most organizations are replete with data and data sources. What to do with that data for better business outcomes is not simply a question of database management, but one of strategy, enrichment, visualization, analysis, and taking action.
Even with all the data businesses collect, there are still unsatisfactory answers to the important questions being asked. Extracting insights to answer those questions with Location Intelligence (LI) is the valuable bond that allows for effective data-driven decision making.
This guide can help the executive, data analyst, or enterprising associate, with a cache of location data, curiosity to explore it, and a need to communicate data-driven business solutions.
Just like pressing “0” in an European elevator to go to the ground floor, your business strategy is the ground floor of the Location Intelligence methodology.
Building a strategy before actually handling data will assist in a focused and effective implementation of LI for your overall outcomes.
Start with a clear understanding of goals, objectives, roles, responsibilities, executive expectations, and the channels of communication available.
First, assemble a LI team within your organization. This cohort is responsible for implementation and execution of LI across the company. It should be cross-functional and include strategists, data and business analysts, a marketing representative or customer relations manager (to deliver results externally, if necessary), and be overseen by an executive-level employee.
If your company already has a GIS department you may want to consult it. However, LI does not require any specific GIS expertise.
Next, assign a project manager or point-person. The project manager will be responsible for documenting strategy, providing regular communication to external stakeholders, and ensuring the entire team is on target to reach the outlined LI goals and objectives.
A goal is a broad outcome that encompasses the desired business result or achievement, while an objective is a measurable action taken towards reaching that goal. Now is a good time to document the goals and objectives for this project.
We recommend using the often employed SMART framework to determine goals. The SMART framework guarantees that your goals are:
The objectives will function as indicators that measure the progress toward the overall LI goal(s). Objectives can be comprised of quantitative or qualitative criteria, such as the amount of increased market spending for a specific customer segment or the efforts and activities performed to improve customer satisfaction.
Try to revisit your goals and objectives once a week to ensure the team is moving in the right direction. This may mean, overtime, original goals and objectives are modified. Be sure to document any emendations in a single place that all team members have access to.
Location Intelligence really starts with a question: What opportunity or challenge is my organization trying to solve?
When beginning a comprehensive Location Intelligence plan, it’s common to focus on all the collected data. Try not to do this! While the gigabytes of information might be enormous, excellent data analysis never starts with the dataset – it starts with questions. Why was the data collected, what’s interesting about it, and what stories can it tell?
It’s also typical to propose questions like, “I want to know what’s in the data,” or “I want to know what the data means.” Sure, but what is meaningful?
If you’re lucky, your company has effectively communicated company-wide initiatives and challenges. The LI team can easily align efforts and potential questions with the overall company strategy. However, if your situation is less advantageous, conduct a sprint or scrum-like brainstorming session with the executive team to inform and jumpstart your business questions.
A good and appropriate question might be something like: Where is the best place to target my marketing efforts to reach my most engaged customers? The more specific your question, the more precise and clear the visual result will be. When questions are too general and broad, (i.e. “exploratory data analysis”), the results and outcomes will be generic and often only understood by those who are versed in the data.
Ultimately, the proposed questions should be revisited, refined, and thought over (and under) based on insights learned from each of the different stages of LI implementation.
The successful implementation of Location Intelligence often depends on an organization’s support infrastructure. To reinforce the effectiveness and prosperity of LI, leverage multi-departmental expertise. Identify existing knowledge bases, portals, or intranets for communicating findings, as well as any visualizations and models. The idea is to create a regular and recurring feedback loop to guide efforts and decrease time-to-insight. Make sure there are clear and open channels of communication to reinforce location data insights.
LI should not be seen as a separate business undertaking, but as an instrument to inform and influence decisions company-wide.
Applying Location Intelligence can only provide a return on investment if the proper questions are asked while using the most precise sets of data. Organizations often work with incomplete, inconsistent, or even (gasp!) incorrect data, from which few, if any, actionable and accurate insights are discovered.
Through data enrichment, internal datasets become more useful, adaptable, and “spatially ready”. Data enrichment is a process that enhances, refines, and augments existing data from imported datasets.
We recommend enriching your data with the following steps.
Before identifying potential external sources of new data, complete an audit of your existing business data. The audit should provide a visible scope of the type(s) of data readily accessible to you and the Location Intelligence team.
It can also be helpful to explore current datasets and repositories while asking these questions:
While many businesses track data based on postal codes (i.e. zipcodes), a true LI-whiz wouldn’t consider using these geographically bounded numbers and letters as the ONLY basis for their business insights. Using postal codes as a determining data tracking measure limits the ability to go further with and perform the best analysis.
Datasets cannot be properly visualized and analyzed if they are not normalized, structured, and uniform. Unstructured or “dirty” data can mean that an address was recorded manually, resulting in differentiations in string data, such as “St.” vs. “Street” or even blank cells.
To normalize your data, we recommend tools like Alteryx, Trifacta, Dataiku, Openrefine, FME, or Excel. Best practices suggest using a data normalization method compatible with your Location Intelligence platform. Additionally, some LI providers offer data cleaning services.
Next determine what data is needed to produce your data visualizations and to perform the appropriate spatial analysis. For a more practical and malleable dataset, this will help reduce load times and other issues attributed to size and configuration of your datasets, remove any extraneous variables or entities that aren’t relevant to the underlying business question.
For example, to analyze trends, patterns, and outliers related to U.S. regional markets, removing data columns and rows attributed to international regions reduces the size and scope of your dataset.
Your team has made it this far (high five!) and the internal data is starting to take shape. Now, consider what other types of data can assist in a solution to your business problem.
Revisiting your original goals, objectives, and questions can help determine potential third party data enrichment sources. Fortunately, in addition to paid data sources, there are thousands of open data portals that can be accessed to augment business data. Check out this list of 40 Brilliant Open Data Projects for inspiration, many of which feature open data portals.
Up to this point, the previous steps have involved examining data at a granular level. From this view it can be extremely difficult to gain a “big picture” understanding of your location data. A platform that allows for easy data integration, as well as a myriad of visual displays, is the best solution in providing a global scope.
At this stage, questions that arise may include: How should the Location Intelligence team interact with “live” data? How can the representation of data unravel over time?
Using dynamic animation to document the evolution of a dataset or interaction to control the time span, can be a useful function for understanding and interpreting datasets.
Ultimately choosing a data visualization method that best meets objectives is correlated with the location data needs. In LI the most important visual asset is the interactive map.
The process of choosing your data visualization is possibly the most important decision in a data visualization project. Deciding how to represent data can influence the previous steps, like what type of data is acquired, as well as subsequent processes, like the particular information you extract.
Most analysts and executives are accustomed to thinking of data as fixed values, but data isn’t static. It is important to consider how to represent data to adjust to new values. This is a necessity because most data comes from the real world, where absolutes don’t exist: temperatures change; purchasing patterns shift; a product launch causes consumer behavior to drastically change.
A fundamental purpose of your data visualization is to be seen and shared. Sharing the data visualization means sharing insights and aids in the collaboration of meeting higher-level organizational goals and objectives. Taking audience into consideration will make the interpretation of data and maps more understandable.
Making a data visualization clear doesn’t mean assuming people using or viewing the map are idiots and require the “dumbing down” of the interface. It does mean valuing clarity and accuracy over ostentatiousness.
Further clarifying location data representation is an essential step in data analysis. By calling more attention to particular data layers and establishing a hierarchy contributes to the readability of your Location Intelligence findings.
Adding interaction by granting the user control over the exploration of data, the selection of a subset of data, or the viewpoint can be the functionality that transforms insights into action.
Similar to selecting the data visualization, the analysis stage can also affect refinement, as a change in viewpoint might require a redesign of the data distribution.
You’re almost there!
Now that a visualization has been generated, it can be used as a working model to analyze and iterate. Decide on the best initial analysis by revisiting the original questions, business challenges, and desired outcomes. What questions were posed and what is the right way to test those questions? This can get a little tricky. It’s helpful to consider some of the many possible analyses out there.
Database Analysis: Data manipulation and filtering, numeric aggregations, and many other techniques found in traditional BI tools.
Geospatial Analysis: Measure distance and proximity, count points in polygons, spatial manipulations and joins, as well as many other practical spatial functionalities found in GIS platforms.
Location Data Analysis: Detect clusters and outliers, predict market volatility, predict future customer patterns, and incorporate scientific modeling and machine learning.
Now that you have enriched, visualized, and analyzed the data, it’s time to take that beautiful data visualization and make real change. With a clearly represented location data visualization and application constructed, it’s now possible to:
If you’ve made it this far you might be scratching your head. No, the methods for LI implementation are not new, but isolation within individual fields has prevented these methods from being used cooperatively.
Location Intelligence decentralizes individual disciplines and places the emphasis and focus on providing a new and contextual meaning to data, instead of the siloed perspective and tools of a particular field or department. Complex datasets can be accessed, explored, and analyzed by anyone in a way that simply was not possible in the past.
There are dozens of quick tools for developing data-visualizations in a cookie-cutter fashion in office programs, on the Web, and elsewhere, but complex datasets used for specialized applications require unique treatment.
Any data visualization tool used for generic purposes will produce generic displays, which ultimately produce generic AND disappointing results. This guide aims to help in the understanding of the value of location data as a tool for human decision-making - using Location Intelligence.
If you are interested in taking your strategy, plan, and analysis further, download the free template add-ons below and sign up for a free consultation with an LI expert.
Implement Location Intelligence in your businessRequest a free consultation
This post was written by Liveli, our master reseller in the Asia Pacific region. –Use Cases
Site Selection, Relocation, and Consolidation are often referenced as the defacto use cases of Location Intelligence, particularly in verticals such as Retail & Real Es...Use Cases
Given the large number of open source spatial analysis libraries available today it can often be difficult to understand how best to combine the strengths of each tool - es...Use Cases
Please fill out the below form and we'll be in touch real soon.