By Justin Norris June 18, 2018

Reporting and Attribution in Marketo: Comparing Your Options

Attribution tools are to a marketer what a compass is to a hiker—both provide direction in your journey and guide your next steps. We’re familiar with current Marketo compasses but now Marketo has embedded a new GPS: Bizible. How does Bizible compare to the Marketo compass you’re currently using, and more importantly, how will it impact your daily operations?

Attribution Review

If you’re accustomed to tracking attribution using Marketo and Revenue Explorer, Bizible represents a significant change. In my last post, I covered the primary functions of attribution tools: capturing data, modeling data and reporting on that data. In this post, I’ll focus on each function, to illustrate the variances of the tools as well as provide common applications to emphasize the impact on your operations.

Data Capture

The scenario: You want to track what marketing initiatives are bringing people into your database (ie, “Lead Source”).

The process with Marketo: Most marketers can track an offer (ex: a content asset) in Marketo fairly easily using a combination of the form and/or landing page. However, tracking a channel is much harder, especially when it involves digital channels.

To obtain channel information in Marketo you’ll need to use UTM parameters on your landing page along with web referrer data to deduce organic channels. For example:

utm_medium = paid-social, channel = paid social

Web referrer contains “google.com” and no UTM parameters = organic search

Of course, you also need to ensure all links are tagged correctly.

Next, you need to get that data into Marketo. You might be asking, what about using hidden form fields to capture these parameters? Certainly possible, but what if people don’t fill out a form on the first page they visit?

To solve, you’ll have to implement your own tracking script to capture the data, convert it to cookies that persist as the visitor jumps from page to page, ensure all forms have hidden fields to capture UTM and referrer values and finally, pass these values to fields on the person object, which can then trigger adding the person to an appropriate tracking program. Moreover, scripts must be flawlessly written, to avoid failure in certain browsers or Marketo logic fails. Many marketing organizations use these methods very successfully, but there is, never-the-less, complexity and potential for error with this approach. 

The process with Bizible: There is little configuration to start capturing data with Bizible. Since Bizible has its own tracking script, tracking is simplified, especially for digital activity.

What’s happening behind the scenes: when someone fills out a form, Bizible detects and logs the URL of the form completion page from which you would deduce the offer. Where Bizible excels is in tracking channels. Bizible automatically detects and stores the UTMs and web referrer data associated with the session, without having to set custom cookies or modify your forms. Finally, because Bizible has direct API connections with ad platforms like Adwords, Bing, and Facebook ads, it automatically pulls ad and cost detail from these platforms, without any manual tagging required on your part.

A Functional Comparison:

 

MARKETO

BIZIBLE

OVERALL

Munchkin tracking script captures web activity. Marketo also logs form fills and email interactions. Referrer/UTM data stored but only in activity log. Custom script required to make this data accessible at the field level.

Bizible tracking script tracks web activity and form fills as well as referrer and UTM data, which is associated with touchpoints for easy reporting. Tracks full clickstream data via Bizible Data Warehouse product, giving it the detail of a full-featured web analytics solution.  

OFFERS

Easy to track; typically uses a combination of a form plus landing page to identify offer, which can trigger the addition of a person to the corresponding program.

Detects and logs the URL of the form completed; offer is deduced based on URL.

CHANNELS

Harder to track, especially digital; requires custom script to convert UTM/referrer data to cookies and form management to ensure data is mapped to fields on the person record in Marketo.

Automatically detects and stores UTM parameters and web referrer data, without creating custom cookies or modifying forms. Direct API connections with ad platforms capture ad and cost data without manual tagging. Platforms currently supported: Adwords, Bing, Facebook Ads.

 

Data Modelling

The scenario: You’re launching a multi-faceted marketing campaign consisting of many offers (blog posts, webinars, ebooks, etc.) across numerous channels and your boss requires a report showing which channels are driving the most engagement with particular offers. In your preparations, you must also plan for the “human factor”— people who interact multiple times with a single channel.

The process with Marketo: Your set-up revolves around Marketo’s unit of attribution, the program, which represents a single marketing initiative you want to track. You can create as many programs as you like, and stages indicate the level of interaction with each program and if that interaction was successful. Programs correspond to a Salesforce campaign (if you track this in Salesforce). How do you capture the relationship between offers and channels? 

Your options are less than ideal:

  1. Create separate programs for offers and channels: This allows you to capture each marketing asset and traffic source the person engages with, but there is no connection between them. Also, if someone engages with the same channel (e.g., organic Twitter) twice, you can’t track the second interaction, because a person can only be a member of a program once.
  2. Create a program for every offer + channel combo. Ex: ABC Ebook + social; ABC Ebook + paid search…and the list goes on. Prepare yourself for a mountain of work and huge task list to maintain.
    programs set up to track offer + channel data
  3. Copy UTM values to the campaign member, which also moves reporting to Salesforce (as you cannot store this additional metadata related to a specific program status in Marketo). This option gives you a more flexible model but requires extra configuration and custom code in Salesforce. Additionally, there are some challenges to tracking costs for ROI and reporting by channels with this method, as you no longer have a distinct campaign to represent the channel.

campaign member example

The process with Bizible: Bizible uses “touchpoints” (or the marketing interaction between a person, offer, and channel) as the unit of attribution and captures every web visit, form fill, and offline touchpoint, grouping them into channels or subchannels that you create. Theoretically, unlimited touchpoints are possible.

The process is straightforward as Bizible creates touchpoints automatically with little to set-up or maintain. Also nice, all interaction data, both offer and channel, are stored on the same record. As a result, there is greater flexibility in reporting options, including changing channel groupings midstream and reprocessing of all data without information loss.

The downside? You’ll have less rich metadata about offers, because the form URL isn’t as precise as the metadata associated with Marketo programs. For this reason, I recommend that you maintain offer programs in tandem in Marketo, which you would likely need anyway to send fulfillment emails, etc.

Bizible Touchpoint Example

A Functional Comparison:

 

MARKETO

BIZIBLE

UNIT OF ATTRIBUTION

Unit of attribution is the program, a single marketing initiative which corresponds conceptually to a Salesforce Campaign.

Unit of attribution is the Touchpoint, a single marketing interaction between a person, an offer, and channel, stored in a single record.

DATA MODEL

One person can be a member of many programs, but only once per program or campaign. Stages indicate the level of interaction with the program and whether that interaction was successful. Program structure can be as granular as you want, but this becomes hard to manage when tracking offer + channel combinations.

Touchpoints are unlimited; touchpoints are captured for every web visit and logged in CRM for a person’s first anonymous visit, every subsequent form fill, and for offline touchpoints when synced from a campaign, even when repeat interactions with the same offer and channel occur. Channels are assigned to touchpoints dynamically based on business rules, eliminating the need to maintain a set of pre-existing campaigns for tracking purposes.

METADATA

A program can have various types of metadata to add additional dimensions, ex: region, product line, type of marketing initiative, etc. but program tags are limited to a fixed set of values, limiting their usefulness.

Touchpoint data is set based on UTM parameters, providing more flexibility to describe channels according to your prefered taxonomy. Channel and subchannels group touchpoints according to your configuration, which can be altered and data reprocessed. Metadata about offers is limited, as only the form URL is captured.

 

Data Reporting and Visualization

The scenario: You’ve completed your marketing campaign, and now it’s time for the fun part— reporting on the results. Your CMO is excited to see how each effort performed and eager to know which was the most profitable.

The process in Marketo: To run the requested reports and provide revenue attribution metrics, you’ll need Revenue Explorer or the newer Marketo Performance Insights (MPI) product. Both apply a multi-touch, even-split model to attribution with slight variances between the tools. With RE, reporting visualization is pivot table style with some ability to create calculated fields and add custom dimensions; however, flexibility is limited and the calculated data is in a silo that can’t be connected to other sources. The MPI reporting dashboard is more visually appealing, but customization is very limited and not designed for ad-hoc data exploration. Overall, if multi-touch, even-split attribution is not the model you want, then Marketo-based attribution does not meet your needs.

The process in Bizible: Bizible offers an array of choices for users accustomed to RE or MPI, including multiple attribution models ranging from traditional to complex plus a newer machine learning model. Determining which of your marketing efforts are most profitable is easier in Bizible, as it allows for more flexibility in manipulating and combining data with other objects and systems. Moreover, because attribution calculations are stored in fields on CRM objects, you have full access to the native reporting capabilities of your CRM. The data is also accessible via API, can be added to other systems or exported to a data warehouse. Marketers will also appreciate Bizible’s Discover UI, which is similar to MPI, for a familiar and aesthetically appealing view.

A Functional Comparison:

 

REVENUE EXPLORER

PERFORMANCE INSIGHTS

BIZIBLE

REPORTING TYPES

Pivot table style ad-hoc analysis as well as charts.

Visually appealing dashboard but limited customization.

Inherits all the benefits of your CRM reporting engine; fully flexible to manipulate data and combine with data from other objects.

MULTI-TOUCH ATTRIBUTION MODELS SUPPORTED

Even split model is the only multi-touch option.

Even split model is the only multi-touch option.

Multiple attribution models: u-shape, w-shape, full-path, custom, machine-learning, to name a few.

CUSTOMIZABILITY

Limited flexibility: analysis areas are fixed. Some ability to create calculated measures and add custom dimensions.

Analysis areas are fixed, cannot create custom measures. Not designed for ad-hoc data exploration.

Very flexible: attribution calculations are stored in fields on CRM objects and can easily be combined with custom fields and objects or custom calculations in code, etc.

DATA PORTABILITY

Attribution data is locked and cannot be manipulated in other systems.

Attribution data is locked and cannot be manipulated in other systems.

Data is accessible via API, can be added to other systems or exported to a data warehouse.

PERFORMANCE

Reporting engine runs very slow.

Too soon to tell.

CRM reporting engine is typically more performant than RE.

 

Which Tool Is Right for You?

Marketing isn’t getting any simpler. Instead, we’re working with more channels, more offers, and a non-linear customer journey. Marketo and Revenue Explorer track the offer and do simple attribution reporting reasonably well, but quickly become labor-intensive and extremely limited when attempting more complex analysis. In contrast, Bizible offers easier tracking out of the box, less maintenance and work for marketing operations teams and more powerful (and flexible) reporting for CMOs. In short, Bizible does the heavy lifting that formerly required an extreme amount of work.

It’s not perfect, but I believe most Marketo users will quickly grow accustomed to the new GPS. The increased flexibility Bizible offers and reduced infrastructure it requires far outweigh any of its drawbacks and the improved directional insights will be appreciated throughout your organization. And for those who want it all, Bizible is an easy stepping stone in your journey towards a full data warehouse.

I don’t know about you, but I’m all for better tools, simpler processes, and added conveniences.  

 

Bizible is now readily available to Marketo customers. Get up and running faster and more efficiently with Perkuto’s Bizible Kick Start package—complete implementation and training, to help you and your team become Bizible savvy and revenue-contributing rock stars. Contact us to learn more!

Justin Norris

Director of Solutions Architecture

Justin is a Marketo Certified Solutions Architect and 4x Marketo Champion. Expert builder of scalable and automated sales and marketing funnels. When not in front of a computer, you can find him enjoying nature and spending time with his wife and daughter.

Leave a Reply

Your email address will not be published. Required fields are marked *