The Relativity Behind Measuring Marketing Performance
Data is everywhere in marketing. A successful marketing program requires finding the right audience, which is then measured with the right data. Everyone understands that good reporting requires good data. Yet, few people understand what good data actually looks like and where it can be found. Every technology platform has its own analytics features. Every channel has its own KPIs. When measuring marketing performance, deciding how to calculate a KPI is more difficult than identifying which KPIs to measure.
In any technology stack, there will be inconsistent analytics. Event platforms and CRM platforms will contain contradictory contact profiles, while web analytics and marketing automation may contain different metrics about page views and form fills. This is normal, and ultimately comes back to the different ways data is collected in each application. Wherever there are multiple ways of measuring a particular reporting metric, identifying the most accurate source for that information is essential.
Measuring Data Accuracy
In most cases, the app used to execute a particular channel will provide the most accurate reporting metrics. Tracking information often isn’t passed from channel to channel as contacts progress through a campaign. This is a particular problem for digital advertising, where accidental clicks are common and cookie consent requirements can prevent sites from collecting conversion information. Bot clicks are also a widespread problem, and not all channels can successfully remove them from reporting. Then there is the impact of internal visitors or test clicks which can’t always be suppressed, and is particularly problematic in a world where remote work is widespread.
When reporting on a specific metric, decide which dataset will be used as the single source of truth for that particular information. This won’t necessarily be the application which provides the best numbers. Consider instead the underlying mechanisms that each platform uses to collect data. Choose the data source which most accurately reflects the KPI being measured.
When reviewing data sources, do remember that no dataset is 100% correct, whatever its origin. Every report has a built-in margin of error due to inconsistencies in the collection methodology or the inaccuracies of the underlying data. Any data engineering exercise should aim to build a database that is good enough to properly reflect campaign performance or to mirror business trends.
Marketing leaders are often less than willing to accept inaccuracies in reporting. Yet, the risks of poor-quality analytics data are often overstated. Absolute accuracy is only really needed for financial reporting when actual revenue is being measured or when budgets are being reconciled. For marketing performance reporting, the accuracy of the interpretation matters more than the accuracy of the measurement. So long as a consistent reporting methodology is used across quarters and years, a small margin of error will be sufficient even for the most closely watched metrics. This is because the trends being observed are generally more important than the absolute numbers.
Ultimately, the reports produced by marketing ops only need to be accurate enough to guide investment decisions and campaign priorities. This applies to individual campaigns and entire marketing programs but also to specific technology or data investments.
The value of good data to sales and marketing teams is enormous but often appears in many intangible ways. Better data leads to better lead conversion rates, higher opportunity win rates or higher click rates. It gives marketers the ability to tune their programs for maximum output. In many ways, data is the hardest investment to measure in any business, but it can also be the biggest differentiator. In a business environment overwhelmed with big data, finding a marketing ops team that can select the right data sources is the quickest shortcut to improved marketing performance.