Integrating Dynamic Data

headshotimage1 Robert Novick is the CEO of enVision Business Consulting. He has over 20 years of experience working with clients to achieve quantitative results through operational improvements and well-conceived and executed strategies. He can be reached at rnovick@envsion-bc.com.

orangearrows


Key performance indicators (KPIs) have become woven into the fabric of most businesses and pushed into the psyches of our organizations. In an effort to better understand and manage our businesses, we’ve created more of them and in the process have diffused their impact and value. And although a few organizations have evolved and cleverly devised leading KPIs, the rest of us have lost sight of their purpose and construction.

The answer then is not to create more KPIs but to create better KPIs.

History

The origin of KPIs and their relationship to Critical Success Factors (CSFs) is most often attributed to D. Ronald Daniel of McKinsey and Company in the sixties. Jack F. Rockart of the Sloan School of Management further refined the concept in the eighties. The idea is simple enough: focus organizations on that which is important to the business and measure it. There are many ways to categorize KPIs, and experts disagree on which is best. One way is to group them into five buckets:

A Housing KPI Revealed Direct Mail for many companies is still the most cost-effective marketing campaign with an acceptable response rate for the cost.

Direct Mail is based on a few KPI assumptions that have held true for years: by using relatively cheap mail rates, an advertiser could reach a zip code or address range with a known demographic and predict a response based on an accepted industry response rate of 1-2% per mailing. This KPI leads to other KPIs such as cost per lead, cost per sale, and customer acquisition cost. These collective KPIs drive a marketing budget and a sales target associated with the campaign. Inventory was purchased and fulfillment centers were staged and the whole scheme worked as planned, except when…

Somewhere in the middle of this story the bottom drops out of the housing market and populations shift. No longer does an address alone matter. What matters is the number of addresses—homes or businesses—with people in them. The “addressable market” has shifted and the “KPIs” of marketing, sales, inventory, and fulfillment based on a direct marketing campaign have effectively been changed. Response per number of homes or businesses is no longer a good “KPI” from which to make policy or business decisions.

FIGURE 1. The data outside the walls of internal corporate data can often provide a more robust and comprehensive view of overall business performance.
FIGURE 1. The data outside the walls of internal corporate data can often provide a more robust and comprehensive view of overall business performance.

Another conventional advertising/house-based KPI has been TV ratings. These ratings systems made sense when a household had one TV and the viewing options were but a few “channels.” Just as an empty house contributes to bad direct mail metrics, today a household may contain dozens of devices that enable “video/advertising” eyeball opportunities. How can one be sure that an ad reached the target person in the household? Clearly “time shifting” and multi-video consumer electronics platforms have made the conventional TV rating KPI less relevant to advertisers. What is the value of the ad? How can I estimate a response rate? How do I manage inventory?

How does one construct meaningful KPIs? How do you know when assumptions must change? Is the data collected really measuring whether the business is successful? Perhaps creating a “better” KPI is found in this case by evaluating housing information that is associated with location.

In fact, by incorporating economic, construction, and existing home sale trends by geography, KPIs become a lens into the future. We have seen the integration of location intelligence and geospatial data enhance existing KPIs and unlock a dimension of analysis and enlightenment that excites corporate strategists and executives like never before.

Financial
Productivity
Quality
Growth
Innovation

In addition, the best KPIs have a target associated with them.

KPIs in Today’s Environment

As described above, today’s business environment can change quickly and dramatically. And leaders often experience the change long before their measurement system recognizes it. This is fundamentally due to business leaders being bombarded with environmental stimuli (like the rest of us) and the way KPIs are constructed.

The majority of today’s KPIs are constructed from operational data like sales, inventory, churn, network performance, etc. Some corporate KPIs broaden the lens by integrating customer demographic (firmographic) and psychographic information. But with increasing frequency, organizations are looking to integrate environmental data like economic trends, weather, and health statistics into their KPIs.

The result is that the further one moves outside the walls of the enterprise, the more robust the data becomes and the more predictable the results. See Figure 1 for a data enrichment view of operational KPIs. Interestingly, most corporate strategy groups integrate market research data into their plans. This is logical, as market research includes customer and environmental information, but the data seldom make their way into corporate KPIs.

To use another advertising example, accepted response rates are associated with radio and television ads. For example, it’s easy to imagine Campbell’s soup running television ads at specified times in specified locations based on market research of a target audience. Purchasing 30-second spots is predicated on the demographic expected to be watching or listening at a certain time, which may revolve around type of programming or time of day. Imagine now the integration of weather data. Where is it cold, rainy, and blustery? What about where it’s hot and clear? The former sounds like a perfect day for soup. The latter, not so much. Dynamic data such as weather significantly increases the predictive power of response when combined with programming and time of day. See Figure 2 for weather-related performance.

The challenge then becomes identifying, acquiring, and integrating dynamic data.

FIGURE 2. A colder January lifted demand for soup in the Northeast and Great Lakes regions while warmer temperatures dampened demand in the West. Campbell’s soup advertising can be customized according to weather, by region.
FIGURE 2. A colder January lifted demand for soup in the Northeast and Great Lakes regions while warmer temperatures dampened demand in the West. Campbell’s soup advertising can be customized according to weather, by region.

Identifying Dynamic Data

Identification is easier than it might seem. Take a static concept like a map and add the element of time to it. When did the oil pass a certain marker in the pipeline? How long before a vaccine expires while in transit? When does a customer most often visit a store or buy a product or service? “Where” and “when” become interconnected, and the combination of the two provides greater insight, opportunity, and revenue potential.


“Most of business data hasn’t been entered into their systems as such—address data, asset data, customer data . . . The information is there, but it needs to be geocoded to reveal the added insight.”


The location (“where”) data is the common denominator. Everything revolves around an asset’s location, a customer’s location, a producer’s location. With location as the baseline of a KPI, any piece of information tied to that location can be associated with it and create a new picture of the information. When information is tied to location, what previously seemed irrelevant suddenly becomes relevant—time, weather, behavior of customers.

For example, location can tie increased sales figures for a neighborhood to the demographics of the neighborhood and to the events that promoted the product (unbeknownst to the retailer and the product company). Absent location as the common denominator, what other data points would be correlated to a sales KPI? Usually, they are: did sales rise or fall, in which locations, and by what percent? In the world of dynamic data we can still see these numbers, but we can drill down to understand many of the drivers of changes to the numbers (e.g., timelines showing before a campaign, during a campaign, and after a campaign), then compare that to a control group or geography. And we can depict it all geographically so leaders can focus on that which is important to the business.

Now that you know what to do with your data warehouse, you can answer “why.”


“Our advice is to start slow. Identify a couple of KPIs that could be improved with a location data  be improved with location data and involve them. The risk of not getting on this moving train is great.”


Acquiring Dynamic Data

Believe it or not, most organizations already have access to geospatial data. Of business data, 80% are location-based but haven’t been entered into their systems as such—address data, asset data, customer data, sales data, inventory data, human resources data, supply chain data, etc. The information is there, but it needs to be geocoded to reveal the added insight. Databases and data warehouses are now all spatially enabled, which solves one problem. Then there’s all the unstructured location-oriented data that organizations collect from market research reports, Excel spreadsheets that track business performance, weather, customer behavior, and news events.

This information was not easily saved in the data models created for BI (business intelligence), CRM (customer relations management), and other enterprise systems. Location intelligence technology has an embedded correlation engine that allows for the association of unstructured and dynamic data. The ability to search geo-referenced text such as place names like cities, states, countries, and landmarks found in corporate documents is making the acquisition, integration, and distribution of geodata simpler every day.

Additional datasets such as imagery and basemaps are needed to visualize the data on a map. These are now reasonably priced, widely available, and integrated into visualization platforms. For example, aWhere, Tableau Software, Spatial Key, FortiusOne, and Space-Time Insight are just a few companies offering solutions.

Integrating the Data

Most organizations develop their KPIs after identifying strategic objectives. Changing the way the organizations develop KPIs and integrate location intelligence starts with the questions executives ask themselves when attempting to measure the success of those objectives.

A not-for-profit whose mission is to improve the future of low-income parents and their children by sending medical practitioners to patients’ homes needs to demonstrate thoughtful growth to its board. One key measure is the year-over-year (YOY) increase in the number of families serviced by geography. The KPI exists in part because the information is easy enough to gather. However, a more appropriate measure is the geographic penetration rate. As saturation due to penetration or demographic shifts occurs over time, reaching an ever-increasing number of families becomes impossible—and the YOY growth rate will decline.

Core to knowing what percent of the population the organization could service is the number of families that fit the organization’s criteria for support and the number of local practitioners available to participate in the program. Although the data exists from several government agencies, such as the Commerce Department, Department of Health and Human Services, and The National Institute of Health, they have not been integrated into the nonprofit’s planning assumptions. So while YOY growth provides a dimension of program growth, it doesn’t provide a strategic rationale for expanding to a new area or for deciding when the organization should shift their energy from “sales” in that area, for example, to solely program delivery.

Therefore, only when growth slows, as communicated through this YOY growth KPI, would the organization investigate the reasons. Had they asked the right questions when developing KPIs and been thoughtful about answers relevant to the business, they might have noticed an opportunity to develop a more forward-looking measure, not the rearward one we’ve come to expect.


“With location as the baseline of a KPI… what previously seemed irrelevant suddenly becomes relevant-time,weather, behavior of customers.”


Conclusion

What was once an ethereal idea is now reality. Next-gen KPIs can be developed to provide the forward-looking, competitive edge that managers and corporate executives need to survive in a hyper-competitive, ever shrinking world. The ability to develop meaningful KPIs and understand changes to those metrics as they occur requires us to think differently about how we create KPIs.

Being aware of dynamic data is the first step in updating your measurement system. Identifying, acquiring, and integrating dynamic data is the second, but it requires executives to partner with their marketing and IT departments to bring to the surface the needed data. Our advice is to start slow. Identify a couple of KPIs that could be improved with location data and evolve them. The risk of not getting on this moving train is great. Someone once told me that the best way to fail is not to try. That axiom is more true today than ever.


A Housing KPI Revealed

Direct Mail for many companies is still the most cost-effective marketing campaign with an acceptable response rate for the cost.

Direct Mail is based on a few KPI assumptions that have held true for years: by using relatively cheap mail rates, an advertiser could reach a zip code or address range with a known demographic and predict a response based on an accepted industry response rate of 1-2% per mailing. This KPI leads to other KPIs such as cost per lead, cost per sale, and customer acquisition cost. These collective KPIs drive a marketing budget and a sales target associated with the campaign. Inventory was purchased and fulfillment centers were staged and the whole scheme worked as planned, except when…

Somewhere in the middle of this story the bottom drops out of the housing market and populations shift. No longer does an address alone matter. What matters is the number of addresses—homes or businesses—with people in them. The “addressable market” has shifted and the “KPIs” of marketing, sales, inventory, and fulfillment based on a direct marketing campaign have effectively been changed. Response per number of homes or businesses is no longer a good “KPI” from which to make policy or business decisions.

Another conventional advertising/house-based KPI has been TV ratings. These ratings systems made sense when a household had one TV and the viewing options were but a few “channels.” Just as an empty house contributes to bad direct mail metrics, today a household may contain dozens of devices that enable “video/advertising” eyeball opportunities. How can one be sure that an ad reached the target person in the household? Clearly “time shifting” and multi-video consumer electronics platforms have made the conventional TV rating KPI less relevant to advertisers. What is the value of the ad? How can I estimate a response rate? How do I manage inventory?

How does one construct meaningful KPIs? How do you know when assumptions must change? Is the data collected really measuring whether the business is successful? Perhaps creating a “better” KPI is found in this case by evaluating housing information that is associated with location.

In fact, by incorporating economic, construction, and existing home sale trends by geography, KPIs become a lens into the future. We have seen the integration of location intelligence and geospatial data enhance existing KPIs and unlock a dimension of analysis and enlightenment that excites corporate strategists and executives like never before.