{
    "version" : "https://jsonfeed.org/version/1",
    "content" : "news",
    "type" : "single",
    "title" : "Decoding public sentiment: Harnessing open data to gain insights into service delivery |Digital.gov",
    "description": "Decoding public sentiment: Harnessing open data to gain insights into service delivery",
    "home_page_url" : "/preview/gsa/digitalgov.gov/cm-topics-button-component/","feed_url" : "/preview/gsa/digitalgov.gov/cm-topics-button-component/2023/10/30/decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery/index.json","item" : [
    {"title" :"Decoding public sentiment: Harnessing open data to gain insights into service delivery","deck" : "An analysis of High-Impact Service Provider (HISP) data","summary" : "Exploring High-Impact Service Provider performance through open source data analysis to enhance customer experience with federal services.","date" : "2023-10-30T10:00:00-05:00","date_modified" : "2024-04-02T09:45:13-04:00","authors" : {"michael-kelty" : "Michael Kelty"},"topics" : {
        
            "customer-experience" : "Customer experience",
            "user-experience" : "User Experience"
            },"primary_image" : { "uid" : "dashboard-templates-dmitry-kostrov-istock-getty-images-1145665438-comp", "alt" :
  "Vector illustration of a colorful dashboard with charts on a dark blue background.", "width" :
  "1200", "height" :
  "630", "credit" :
  "", "caption" :
  "Dmitry Kostrov/iStock via Getty Images", "format" :
  "png" },"branch" : "cm-topics-button-component",
      "filename" :"2023-10-30-decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery.md",
      
      "filepath" :"news/2023/10/2023-10-30-decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery.md",
      "filepathURL" :"https://github.com/GSA/digitalgov.gov/blob/cm-topics-button-component/content/news/2023/10/2023-10-30-decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery.md",
      "editpathURL" :"https://github.com/GSA/digitalgov.gov/edit/cm-topics-button-component/content/news/2023/10/2023-10-30-decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery.md","slug" : "decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery","url" : "/preview/gsa/digitalgov.gov/cm-topics-button-component/2023/10/30/decoding-public-sentiment-harnessing-open-data-to-gain-insights-into-service-delivery/","weight" : "1","content" :"\u003ch2 id=\"the-value-of-decoding-open-data-for-understanding-customer-experience\"\u003eThe value of decoding open data for understanding customer experience \u003c/h2\u003e\n\u003cp\u003eIn the era of digital transformation, open source data has become a critical tool in shaping customer experiences and informing strategic decisions. One way to use open data for insights into customers\u0026rsquo; experiences is to hack it. Hacking open data refers to the process of using creative and innovative techniques to extract insights from data that is not traditionally considered a source of business intelligence.\u003c/p\u003e\n\u003cp\u003eThe GSA Analytics and Decision Support Division within the Office of the Chief Financial Officer recently \u0026ldquo;hacked\u0026rdquo; open data from \u003ca href=\"https://www.performance.gov/cx/hisps/\"\u003ePerformance.gov\u003c/a\u003e to see what we could learn about public sentiment around federal service delivery. The analysis focused on survey data collected by High-Impact Service Providers (HISPs), and a dashboard created to visualize the data. The \u003ca href=\"https://d2d.gsa.gov/report/hisp-cx-dashboard\"\u003eHigh-Impact Service Providers Dashboard (HISP-D)\u003c/a\u003e is available on the \u003ca href=\"https://d2d.gsa.gov/\"\u003eData2Decision (D2D) platform\u003c/a\u003e to all federal employees who are curious about customer experience (CX).\u003c/p\u003e\n\u003cdiv\n  class=\"image\"\n\u003e\u003cimg\n      src=\"https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_w800.png\"\n      \n        alt=\"Screenshot of HISP CX Dashboard landing page top section where user can select parent organization, HISP, and service. Above the filters are definition of survey rating scale and links to data sources.\"\n        srcset=\"https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_bu.jpg 48w,https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_w800.png 800w,https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_w600.png 600w,https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_w400.png 400w,https://s3.amazonaws.com/digitalgov/high-impact-service-provider-hisp-dashboard-filter_w200.png 200w\"\n\n      sizes=\"(max-width: 800px) 100vw, 800px\"\n    /\u003e\u003c/div\u003e\n\n\n\u003ch2 id=\"collecting-data-on-high-impact-service-providers\"\u003eCollecting data on High-Impact Service Providers\u003c/h2\u003e\n\u003cp\u003eAs defined in \u003ca href=\"https://www.whitehouse.gov/briefing-room/presidential-actions/2021/12/13/executive-order-on-transforming-federal-customer-experience-and-service-delivery-to-rebuild-trust-in-government/\"\u003eExecutive Order 14058\u003c/a\u003e, HISPs are \u003ca href=\"https://www.performance.gov/cx/hisps/\"\u003ethose federal entities designated by the Office of Management and Budget (OMB)\u003c/a\u003e that provide or fund high-impact customer-facing services, usually due to a large customer base or a significant impact on those served by the program. These services are essential to ensuring a great experience with federal government services for the public.\u003c/p\u003e\n\u003cp\u003ePer OMB, HISPs are required to collect feedback from customers after they have completed a transaction. OMB provides guidance for this collection and survey design in \u003ca href=\"https://www.whitehouse.gov/wp-content/uploads/2018/06/s280.pdf\"\u003eOMB Circular A-11 Section 280 (PDF, 385 KB, 14 pages)\u003c/a\u003e, which was most recently updated on August 11, 2023. Most questions are rated on a 1-5 Likert scale (view an \u003ca href=\"https://digital.gov/resources/customer-experience-toolkit/#measurement\"\u003eexample on the CX Toolkit\u003c/a\u003e), where 1 represents a very negative experience and 5 represents a very positive experience.\u003c/p\u003e\n\u003cp\u003eIn its guidance, OMB has established seven drivers of customer experience:\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eEffectiveness\u003c/li\u003e\n\u003cli\u003eEase\u003c/li\u003e\n\u003cli\u003eEfficiency\u003c/li\u003e\n\u003cli\u003eTransparency\u003c/li\u003e\n\u003cli\u003eEquity and Humanity\u003c/li\u003e\n\u003cli\u003eEmployee interaction\u003c/li\u003e\n\u003cli\u003eOther\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eGSA is responsible for consolidating governmentwide HISP feedback data on a quarterly basis and reporting it publicly on the \u003ca href=\"https://www.performance.gov/cx/data/\"\u003ePerformance.gov customer experience data page\u003c/a\u003e. Visit the \u003ca href=\"https://github.com/GSA/touchpoints/wiki/Data-Collections\"\u003eTouchpoints GitHub Wiki data collections page\u003c/a\u003e to learn more about the data submission and collection process.\u003c/p\u003e\n\u003cp\u003eThe raw data gathered by GSA includes ten quarters of submissions for 29 out of 35 HISPs reporting customer feedback. To make this data set more transparent and actionable, the analytics team created a dashboard with visualizations of response trends, question-level score distributions, and channel-level score distributions at both the provider and service levels. \u003c/p\u003e\n\u003ch2 id=\"data-driven-insights-into-customer-experience\"\u003eData-driven insights into customer experience\u003c/h2\u003e\n\u003cp\u003eTo build the dashboard, we began by consolidating various data fields from the \u003ca href=\"https://github.com/GSA/touchpoints/wiki/Data-Model\"\u003eTouchpoints data model schema\u003c/a\u003e into a single machine-readable file. During this process, we identified significant fluctuations in response rates, differences in how agencies adapted the seven survey prompts, missing data, and changes in the definitions of services over time. These irregularities dictate caution in comparing results among different services and HISPs, as well as interpreting trends over time.\u003c/p\u003e\n\u003cp\u003eAlthough these data quality issues make comparisons challenging, the dashboard does offer insights into a range of business questions concerning the performance of individual HISPs and the services they provide:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat are the overall satisfaction levels for a specific HISP? \u003c/li\u003e\n\u003cli\u003eWhat drivers have the most significant impact on a typical customer’s experience?\u003c/li\u003e\n\u003cli\u003eHow does customer satisfaction compare between different HISPs offering similar services?\u003c/li\u003e\n\u003cli\u003eAre there any notable trends in customer experience or response rates over time?\u003c/li\u003e\n\u003cli\u003eWhich services have the highest and lowest response rates?\u003c/li\u003e\n\u003cli\u003eAre there services that are underutilized, despite high customer satisfaction rates?\u003c/li\u003e\n\u003cli\u003eWhat services might need more resources based on low satisfaction and high demand?\u003c/li\u003e\n\u003cli\u003eHow have different HISPs customized the survey prompts, and what impact has this had on the data collected?\u003c/li\u003e\n\u003cli\u003eAre certain survey channels (email, website, in-person, mobile, etc.) more effective for particular types of services?\u003c/li\u003e\n\u003cli\u003eIs there a noticeable difference in feedback quality based on the survey channel?\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eTo highlight how analysts can use the dashboard to generate insights on both service delivery and survey design, we offer two examples.\u003c/p\u003e\n\u003ch3 id=\"example-1-customers-service-perceptions\"\u003eExample 1: Customers’ service perceptions\u003c/h3\u003e\n\u003cp\u003eFirst, an analyst can use the dashboard to see customers’ perceptions of the service, including how respondents are rating an agency or a specific HISP across the seven driver areas, as well as trends over time. For one HISP, depicted in the screenshot below, we observed that although the percentage of positive responses had been on the rise, there were still clear areas where the service was underperforming. Specifically, a significant number of customers reported negative responses across the \u003ccode\u003eQuality\u003c/code\u003e, \u003ccode\u003eSimplicity\u003c/code\u003e, and \u003ccode\u003eEfficiency\u003c/code\u003e questions. This suggests that even if a service seems to be improving overall, there could be certain aspects that need targeted improvement.\u003c/p\u003e\n\u003cdiv\n  class=\"image\"\n\u003e\u003cimg\n      src=\"https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_w800.png\"\n      \n        alt=\"Screenshot of Question Level Analysis for an unnamed HISP, showing the distribution of scores across the seven driver questions, and the trend of positive vs. negative scores over time.\"\n        srcset=\"https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_bu.jpg 48w,https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_w800.png 800w,https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_w600.png 600w,https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_w400.png 400w,https://s3.amazonaws.com/digitalgov/question-level-analysis-hisp-dashboard_w200.png 200w\"\n\n      sizes=\"(max-width: 800px) 100vw, 800px\"\n    /\u003e\u003c/div\u003e\n\n\n\u003ch3 id=\"example-2-survey-design\"\u003eExample 2: Survey design\u003c/h3\u003e\n\u003cp\u003eA second use for the dashboard is to analyze the design of the survey itself. The dashboard displays the count of responses for each question (respondents are not required to answer every question). At the aggregate level, we saw that the \u003ccode\u003eSatisfaction\u003c/code\u003e question had the most responses, and the \u003ccode\u003eEquity\u003c/code\u003e question had the fewest responses. At the individual HISP level, we saw that some services have relatively equal response numbers across prompts, whereas others reveal a drop-off in responses after certain questions. The screenshot below shows that respondents for this particular HISP are answering the \u003ccode\u003eSatisfaction\u003c/code\u003e and \u003ccode\u003eTrust\u003c/code\u003e questions at the highest rate. This data may have implications for the length of the survey and how the questions are sequenced and worded. \u003c/p\u003e\n\u003cdiv\n  class=\"image\"\n\u003e\u003cimg\n      src=\"https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_w800.png\"\n      \n        alt=\"Screenshot of Question Level Analysis for an unnamed HISP, showing the counts of responses for each question across 2023 Q1 and 2023 Q2.\"\n        srcset=\"https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_bu.jpg 48w,https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_w800.png 800w,https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_w600.png 600w,https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_w400.png 400w,https://s3.amazonaws.com/digitalgov/question-response-counts-comparison-hisp-dashboard_w200.png 200w\"\n\n      sizes=\"(max-width: 800px) 100vw, 800px\"\n    /\u003e\u003c/div\u003e\n\n\n\u003cp\u003eThe dashboard also displays the distribution of scores by survey channel, as well as the trends in the numbers of responses collected through each channel over time. We found that for certain services, customer satisfaction ratings differed significantly depending on the survey channel. In the screenshot below we see that the HISP receives a high volume of survey responses by computer but customer satisfaction levels are lowest among those responses relative to other channels. This data may point to a number of implications that would require further investigation, for example, the channel may be underperforming or the timing of email surveys may affect responses.\u003c/p\u003e\n\u003cdiv\n  class=\"image\"\n\u003e\u003cimg\n      src=\"https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_w800.png\"\n      \n        alt=\"Screenshot of Service Level Analysis for an unnamed HISP showing the distribution of scores by survey channel, and the survey channel response counts over time for automated phone and computer (other channels not shown).\"\n        srcset=\"https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_bu.jpg 48w,https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_w800.png 800w,https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_w600.png 600w,https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_w400.png 400w,https://s3.amazonaws.com/digitalgov/survey-channel-response-values-hisp-dashboard_w200.png 200w\"\n\n      sizes=\"(max-width: 800px) 100vw, 800px\"\n    /\u003e\u003c/div\u003e\n\n\n\u003ch2 id=\"final-thoughts-using-open-data-for-service-improvement-and-collaboration\"\u003eFinal thoughts: Using open data for service improvement and collaboration\u003c/h2\u003e\n\u003cp\u003eThis analysis serves as a compelling reminder of the power of “hacking” open source data to drive service improvements across federal agencies. We invite you to explore the \u003ca href=\"https://d2d.gsa.gov/report/hisp-cx-dashboard\"\u003eHigh-Impact Service Providers Dashboard\u003c/a\u003e to find data related to your agency, HISP, or service.\u003c/p\u003e\n\u003cp\u003ePlease keep an eye out for future Digital.gov blog posts from the GSA Analytics and Decision Support Division as we continue to dig deeper into the HISP survey data and explore both the structured and unstructured responses.\u003c/p\u003e\n"}
  ]
}