Attribution 101

The C3 Metrics Advantage

• Superior Data Accuracy:  C3 provides the most complete and comprehensive data accuracy through a proprietary tagging infrastructure collecting all data including non-converting users.
• Machine Learning:  C3 is the ONLY platform with a continuously updating model, which is improving and learning with every conversion as opposed to taking a point of time view.
  
• Real-Time Viewability:  C3 is the ONLY platform providing a real-time mechanism to suppress conversion credit for non-viewable ads.
  
• Programmatic Feedback: C3 is the ONLY platform delivering performance data in real-time to each of your partners, completing the required feedback loop to optimize your digital ad delivery.
  
• Vendor Management:  C3 is the ONLY platform to provide your vendors a siloed login, enabling them to see inside their buys and optimize towards the campaign goals.
  
• Cross-Device Built-In:  C3 automatically connects users across their different devices via a probabilistic match.
• Speed of Implementation:  Due to our extensive experience and scalable SaaS architecture, most clients are live within 7 – 10 days.
  
• CMO / Executive Dashboard:  C3 provides an additional hierarchy of all data for Executives not requiring nor wanting granular detail.

How The C3 Platform Works

The machine-learning Bayesian model is 100% data driven and is the same technology behind self-driving cars, email spam filters and military code decryption.

At a high level, there are 6 components to the C3 Metrics platform:

1. Proprietary tags capture every ad impression, click & conversions, including all converting and non-converting data.
   
   
2. All spend and Offline media is imported into the platform.
   
   
3. The machine-learning algorithm connects users across different devices, removes touch points which have not been seen, ads which have zero contribution and each conversion is fractionalized with each causal touchpoint receiving a percentage of credit.
   
   
4. Real-time feedback loop is completed for your programmatic vendors.
   
   
5. Vendors managed via a siloed login, enabling them to see inside their buys and optimize towards the campaign goals.
6. Granular ROAS or CPA is determined for Optimization.

Analytics & Modeling

C3 provides the most complete and comprehensive data accuracy by utilizing a proprietary tagging infrastructure that collects customer journey data on all converting and non-converting users.  The journey takes the form of the traditional A-I-D-A "purchase funnel" of Awareness, Interest, Desire and Action; represented within the C3 Metrics algorithm as O-R-A-C, Originator, Roster, Assist and Converter.  

Digital data is added via the C3 Metrics proprietary tags; Television and Radio via post-log data with the algorithm correlating spots times to jumps in navigational search traffic (direct navigation, brand search or SEO) when compared to a rolling same day/time 4-week average; and Direct Mail and Print via proxy URL’s.

An example path may appear as follows:

In addition to collecting all converting and non-converting data from a single source in real-time via the C3 Metrics proprietary tagging infrastructure, the C3 Metrics algorithm incorporates the following additional data validation routines in real-time:  a) Different Devices; b) Viewability; c) Navigational Search at the bottom of the conversion (‘Converter’); d) Touchpoints appearing in the funnel which based on time cannot have any causal impact.

Different Devices
According to research, 90% of consumers move between devices to accomplish a conversion goal, with 67% of people using multiple devices sequentially to shop online.   Without connecting users to their different devices, the other devices would be considered ‘non-converting’ and adversely impact the model for those touchpoints leading to non-detectable errors as high as 90%.  For example, if the converting device contained the following path:

Both the 2 ^nd and 3^rd device would be part of the non-converting or ‘negative’ data set for purposes of modeling.

The C3 Metrics algorithm incorporates real-time cross device connection as part of data validation.  The user’s converting device path would be automatically incorporated with the 2 ^nd and 3^rd path for a final path showing the following:

Viewability
According to research, greater than 50% of digital impressions are never seen with 69% not in view for ad networks and programmatic vendors.  Upwards of 90% of the data utilized for digital modeling is impression data. Without determining viewability, the model results will have a large margin of error.  For example, if the final conversion path was the following:

The ‘Originator’ on Day 1 is shown to be responsible for originating or building awareness.  In Digital advertising it is common practice for a vendor to purposely place an advertisement at the bottom of pages, which results in a higher profit for purchasing ads, yet results in ads which will never be seen.  In this instance, there is a large degree of likelihood those unseen impressions could ‘cookie’ a large number of converting paths and receive undue credit.

The C3 Metrics algorithm incorporates real-time viewability as part of data validation.  With C3 Metrics, the user’s final path would automatically remove ads, which have not been seen.  The final path would show the following:

Non-Causal Touchpoints
With greater than 90% of advertisers currently utilizing a Last Touch Attribution method, it is common to see many touchpoints appearing in the funnel within the last couple of minutes or even seconds prior to conversion.  

In the above example, if the conversion event takes an average of 5 minutes to complete and advertiser data shows that the minimum time to completion is 3 minutes, the Converter (‘Facebook’) which appeared in the path only 1 minute prior to the conversion could not have had any influence and was likely the result of the user clicking over to Facebook upon receiving a Facebook notification.

The C3 Metrics algorithm incorporates a time-decay in the algorithm to remove Touchpoints as part of data validation.  The user’s final path would automatically remove touchpoints where it would be impossible to have any causal impact based on the time difference between the touchpoint and conversion.  In this example, incorporating time decay, the final path would show the following:

Navigational Search
According to research, nearly half of all digital conversions end in either Brand Search or Direct Navigation. Without correcting for this, Brand Search may receive an undue percentage of credit.

The C3 Metrics algorithm incorporates a navigational search component in the algorithm to determine if brand search or direct navigation is purely navigation and not causal.  In the case above, the Brand Search is the result of the user viewing the TV spot and Brand Search acts as navigation only.  In this example, incorporating navigational search, the final path would show the following:

Without C3 Metrics
Without proper data validation provided by C3 Metrics, the following data would be presented to a model:

In these instances, the Display ad on Day 1 and the Facebook on Day 22 would be seen as valid Touchpoints within models provided by other vendors and the TV on Day 22 and the Brand Search on Day 12 would be treated as negative or non-converting.

With C3 Metrics
With proper data validation from the C3 Metrics algorithm, the final converting path is the combination of all the user’s devices, inclusion of real-time viewability, non-causal Touchpoints and navigational search as follows:

The final converting path appears as follows:

Both the Display ad on Day 1 and Facebook on Day 22 are not included, as they are not causal to the conversion and the TV on Day 22 and Brand Search on Day 12 are included.

After completing data validation, the machine learning Bayesian algorithm attributes the fractional credit each touchpoint should receive for each conversion event. 

The model then breaks down every touch point into individual elements, pivoting the results to report across all of the elements desired:  placement, campaign, channel, keyword, tactic, messaging, audience and more.  The results for each touchpoint are updated in real-time for every conversion event, with the model constantly learning and updating in real-time.  The algorithm is so sophisticated that it will bifurcate for different conversion types or audience segments. 

Validation is built into the platform with Confidence scores calculated for each touchpoint and rolled up to channel, tactic and strategy levels to ensure the best possible forecasts and optimizations results.

By employing proper data validation with the most advanced machine-learning algorithm, C3 Metrics provides the most accurate picture of what media investments are working and which ones are not.  In addition, C3 provides a comprehensive and highly granular set of recommendations via our predictive optimization model. 

Why Attribution Now

Our understanding of the customer journey has evolved.  Consumers convert over multiple days via multiple channels utilizing multiple devices.

This leads most marketers to measure their media in silos

Measuring your media in silos leads to over counting each KPI.  As you add more vendors, the claims of credit for each conversion increase exponentially.

If that wasn’t bad enough, there are 5 additional issues facing marketers working in this multi-channel world:

1)  Digital Ad Viewability

Greater than 50% of all digital ads are never seen, which should give every CMO reason to pause.

All of the credit for your conversion is given to the vendor who is able to ‘game the system’ by purchasing low-priced inventory that is never seen which means up to 69% of your media budget is wasted.  For a typical campaign, 95% of all the data we collect is what?  Display views.  So when 69% of all programmatic display is actually not viewable, what's the impact?  The impact is mathematically shocking:

95% of your data x 69% error = 66% of your entire dataset is in error

Imagine if your CFO reported earnings to Wall Street or the IRS, and 66% of the data used to report earnings were in error.

2)  Broken Feedback Loop

Your programmatic ad partners operate on a real-time feedback loop of pixel fires to drive their algorithms.  If your ad campaign fires pixels without viewability, it reinforces the programmatic buying of unviewable ads, and ads only at the bottom of the funnel.

3)  Multiple Tabs Open at the Same Time

Multi-Tab browsers have increased the number of 'last second ads' jumping in and winning conversions as consumers tends to shift back and forth between popular social networking and news sites as part of their regular browsing.

4)  Conversions Ending in Brand Search

Measuring your media with Brand Search at the bottom of the funnel attributes up to 90% of your media credit to ‘navigation’. 

Without controlling for this, you will give up to 90% credit to bottom of the funnel brand search versus the stimuli that drove people to type that brand search.

5)  Current TV, Radio, Direct Mail & Print Metrics are limited

Offline metrics provide data on 'opportunity to view' but are unable to connect your buy to your KPI's.

Measuring your media by GRP’s limits your ability to determine unprofitable tactics from the profitable ones.  TV measurement hasn't changed in 30 years, but the way we consume TV with tablets and Smartphones in hand has changed everything about TV.  

Adding TV & Radio

C3 Metrics measures both the immediate (Direct) and prolonged effect (AdStock) of Television and Radio.

Direct is the immediate effect of the spot and Adstock is the lagged or decay component, typically expressed in terms of the 'half-life' of the ad copy (where new exposures increase awareness to a new level and awareness is higher if there have been recent exposures and lower if there have not been).  

C3 ingests post-log data (date, time, DMA, creative, cost, C3 ratings, etc.) and correlates spots times to jumps in navigational search traffic (direct navigation, brand search or SEO) when compared to a rolling same day/time 4-week average.  The time frame analyzed (‘TV Decay Window’) is dynamic based upon true spot activity vs. a 5 or 15-minute window.  Overlapping spots are handed via proprietary Station Weighting.

For Direct, unique visitors during the decay window are analyzed and spots are dynamically added to their digital trail.  For AdStock, C3 utilizes a reverse-decay curve based on the GRP delivery levels.

Not hampered like current panel methodologies, which only take into account subsets of data, C3 Metrics is easily able to rule out channel flipping and consumers not paying attention because the methodology is based upon user response within the response curve visualized for each specific spot.

Modeling Example

Final Path (in which Facebook would typically win)

Step #1 - Remove Invalid / Bad Data

Step #2 - Connect Users Across Different Devices