In The News
-Staff
Marketing attribution rarely fails in a dramatic or immediately visible way. More often, the problem develops quietly as campaigns expand across platforms, teams adopt different naming conventions, and reporting systems interpret the same customer journey differently.
Nimisha Reddy Kolukuri developed a groundbreaking marketing attribution framework at Groundworks. Facing inconsistent data across platforms, she created a standardized system that corrected $11 million in misattributed revenue. This innovation provides reliable performance insights, boosting ROI and enabling confident budget allocation for over 70 locations.
Over time, those inconsistencies can distort performance metrics, complicate budget decisions, and make it difficult for business leaders to determine which channels are actually generating results.
Nimisha Reddy Kolukuri encountered this challenge while working as a Marketing Analyst at Groundworks. As the company’s digital marketing ecosystem grew across paid search, display advertising, affiliate partnerships, organic search, referral traffic, and direct traffic, its platforms did not always classify leads consistently. A customer journey recorded one way in Google Analytics 4 could appear differently in CallTrackingMetrics, the sales funnel, or downstream reporting databases.
The consequences extended beyond reporting discrepancies. When leads were assigned to the wrong acquisition channel, important metrics such as return on ad spend, cost per lead, cost per qualified lead, conversion rates, and channel contribution became less reliable. Referral traffic presented a particularly difficult problem because some visits originating from organic search, paid campaigns, artificial intelligence search engines, and mapping services were being incorrectly classified as referrals.
Nimisha led the design and development of a Marketing Source and Medium Attribution Classification Framework to address those inconsistencies. Her objective was to establish a standardized system that could classify marketing traffic accurately across the organization and provide teams with a dependable source of performance data.
Rather than relying on a single source field, the system uses hierarchical, rules based logic to evaluate multiple signals before assigning each lead to a channel. These signals include UTM parameters, campaign identifiers, referral hosts, landing page information, call tracking attribution fields, and other available metadata. Paid channels are evaluated before nonpaid sources, reducing the risk that paid campaigns will be incorrectly credited to organic or referral traffic.
Nimisha developed the automated classification logic in SQL within BigQuery and worked with stakeholders across marketing, paid media, SEO, analytics, and web teams to understand how traffic was generated and recorded throughout the customer journey. The resulting classification logic standardized channels including paid search, display, organic search, organic social, direct, referral, and affiliate marketing.
Technology alone, however, could not solve the problem. Consistent attribution also required governance. Nimisha created a Master Attribution Spreadsheet that served as the organization’s single source of truth for marketing sources, mediums, campaigns, UTM structures, naming conventions, and routing rules. She also introduced validation procedures to identify duplicate sources, missing parameters, invalid campaign identifiers, conflicting attribution records, and newly introduced traffic sources requiring classification.
“The key design decision was to evaluate every signal in a consistent hierarchy rather than trust whichever source field appeared first,” Nimisha explains. “Prioritizing paid channel evidence before nonpaid sources helped prevent campaigns from being misclassified as organic or referral traffic.”
The project identified and corrected approximately 2,870 leads that had previously been classified as referral traffic but should have been attributed to organic search or paid marketing. Based on Groundworks’ Average Dollar per Lead methodology, those corrections represented approximately $11.09 million in attributable revenue. The figure did not reflect newly generated revenue. Instead, it revealed the value that had previously been credited to the wrong channels and provided leadership with a more accurate view of marketing performance.
The system now supports measurement across more than 70 locations and approximately 20 brands. By improving the reliability of attribution data, it has given marketing, finance, analytics, and executive teams greater confidence when evaluating campaign performance, allocating budgets, forecasting results, and identifying opportunities for optimization. Its automated classification logic has also reduced the need for recurring manual corrections.
The initiative reflects Nimisha’s broader experience at the intersection of marketing strategy and analytics. At Groundworks, she supports performance across Meta, Google Ads, TikTok, Nextdoor, and Criteo while helping analyze monthly marketing investments ranging from $900,000 to $1.1 million. Her work has contributed to a 14 percent increase in lead generation and a 17 percent annual improvement in return on ad spend across 20 brands.
Nimisha holds a Master of Science in Business Analytics from Arizona State University’s W. P. Carey School of Business. Her background also includes business intelligence, SEO, SEM, reporting automation, machine learning, and natural language processing. Across those disciplines, her focus remains consistent: ensuring that complex data is not merely collected, but organized and translated into information leaders can trust.
For most organizations, marketing attribution stops at building another dashboard. Nimisha’s work went a step further by fixing the data feeding it. That distinction is what allowed a single framework to bring reliable measurement to more than 70 locations and roughly 20 brands, turning a chronic reporting gap into a governance system the entire organization could trust.
