Geocoding

Identifying customer locations for advertisement campaign in Wake County

Problem Statement

A local business in Raleigh North Carolina is planning to expand its customer base. It intends to understand its current customer base which will help with highlighting areas which are underrepresented, using geospatial lens. This will in turn inform its targeted and tailor-made campaign plans to effectively reach out to the potential customers, in areas where it is underrepresented. The objective of this assignment is therefore to use the addresses obtained from the company in identifying customer location for advertisement campaigns

Analysis procedure

To address the problem, I used virtual computing lab’s ERIS ArcGISPro 2.9.1 which is hosted on Remote Desktop Web Client. Main tools used were Create Address Locator, Geocode Addresses, and U.S. Postal services ZIP Code Lookup. Data used on this assignment came from the instructor, and included the excel file for customer data, shape files for Wake Streets Address and Wake Zip Codes which I used to geocode the customer data. The instructor also provided Wake County Line which I used to showcase the visual boundary of the county.

The whole essence of geocoding was to convert customer address into physical locations which can be pinpointed on a map. To do this, I created address locators which were used to match the customer database into physical points on earth’s surface. I started with developing ZIP locator using the ZIP Code layer which was provided as a reference layer. I then used Excel to Table tool to convert the customer data to a table to the map. I created the ZIP locator using the Create Locator Tool which was found in the Geoprocessing group tool icons in the Analysis ribbon tab. The geoprocessing pane report window had indicated a number of errors and warnings which were then resolved using Repair Geometry Tool.

After creating the Wake ZIP Locator, I performed Batch Geocoding using Geocode Addresses Tool. In the Geocode Addresses dialog box, I set the input table to customers data layer, input address locator to Wake ZIP Locator, and Input Address Fields to Single field. This produced geocoded customer addresses layer which had only 49% of its features matched, lest were either unmatched or tied. I had to rematch some of the unmatched features. This is where I used the U.S. Postal services ZIP Code Lookup to identify the missing ZIP codes which were entered into the unmatched dialogue box, to rematch some of the selected features.

Part 2 of the assignment involved geocoding by street addresses layer. I used the Create Locator Tool box to create a Wake Street Locator using the Street Shapefiles. I edited the locator on the geocoding options to set the minimum match score to 85 and match with no zones to yes. The aim was to display only those features with a match score of 85 and above in the matched customer location by street. Using Geocode Addresses tool, I performed a batch geocode of the customers data, using the Wake Streets Locator. The diagram at the end of this section summarizes the process used.

Process Diagram

Figure 1: Process Diagram for Data Geocoding

RESULTS

Figure 2 and 3 shows results obtained from the geocoded data.

Application and Reflection

Problem description: Identify and map USAID/Malawi activity locations to understand where USAID is implementing activities across Malawi.

Data needed: USAID Malawi activity lists, and where its working (addresses). Malawi District boundary shapefiles and geocodes. The data will be obtained USAID/Malawi GIS Database.

Analysis procedure: I will use the Geocode by Zip address sort of assessment to conduct similar exercise which will focus on USAID activity location data. For Malawi, this will be ideal since we don’t have a more established street naming for all our streets. Only major streets have proper naming and data is readily available, whilst data for minor streets and other new streets is scarce. I will create a locator using the Malawi districts shape files. I will then run a batch geocode of the activity location addresses using this locator, which will be tied either on district name or object ID, which will be my point of single matching. I will rematch the data sets that will not match using this, and then produce maps to showcase where we are working, using a quick and simple methodology of batch geocoding.