Adobe Marketing Cloud Review: 2015 Digital Marketing Summit Announcements

March 23, 2015

Last week Adobe held their annual Digital Marketing Summit in Salt Lake City. Adobe is known for putting on a solid show each year and with Keynotes featuring engaging brands and celebrities, including two climbers that recently made news in our home state of California. Last week’s event was no exception. Adobe made a large number product announcements including some that expand their “Marketing Cloud”.

The Adobe “Marketing Cloud” is not a product - it is a brand


It is important to understand that Adobe’s “Marketing Cloud” is not a product - but a brand that includes eight (!) different products: Adobe Analytics, Adobe Audience Manager, Adobe Campaign, Adobe Experience Manager, Adobe Media Optimizer, Adobe Primetime, Adobe Social and Adobe Target. Many of these products come in various versions and editions - including old and new. We will come back to that when we discuss Adobe Campaign.

The Adobe “Marketing Cloud” products come from acquisitions - making integration a challengemarketing clouds_pic3

Just like the offerings of Salesforce, Oracle and IBM, most of Adobe’s products were created through company acquisitions. Each acquired product comes with its own data store. Adobe is better than most and will help you - where possible - integrate their eight products with services they call “core services”.  Much of the mapping is still custom, and has to be done on a customer-by-customer basis. As a customer, you may not see “core services” as a line-item on your quote from Adobe, but the (many) people hours required to make the different Adobe products talk to each other are accounted for in the product subscription fee you are paying.

We believe that marketers would benefit from a true purpose-built product, built from the ground up to integrate customer data profiles, predictive analytics and personalized omni-channel execution. We think no marketer can live without customer profiles or predictive analytics, to power relevant omni-channel campaigns.

The rest of this review will focus primarily on updates to the Profiles and Audiences component of Adobe’s “Core Services”, as well as on their Adobe Analytics (website analytics) and Adobe Campaign (multi-channel campaign management, originally acquired from Neolane), since these are the most relevant to our readers.

Does Adobe Do Customer Data Integration and Data Quality?

On stage, Brad Rencher announced the ability for users to upload customer data into Adobe Analytics via an one-time upload into the “Profiles and Audiences” Core Service. The launch of this feature shows that Adobe recognizes the importance of customer data that typically resides outside of the Marketing Cloud. It appears that Adobe does not yet provide true data quality in the cloud - and we are also puzzled by the statement that this data upload is “one-time”.

How Does Adobe's data integration work?

We know that very few clients have their data located in one place, and it’s also quite obvious that this data will change almost daily. Therefore it puzzles us that Brad talked about an one-time upload. Shouldn’t customer data be pulled from different sources automatically, on a daily basis to be useful for analysis and campaigns?

Does Adobe provide data quality?

Uploading data feeds from customers’ various cloud and on-premise solutions isn’t very difficult. The most challenging part is making sure that the data has been appropriately cleansed and de-duplicated so that accurate and meaningful actions can be taken from the insights it offers. And by de-duplication we don’t just talk about a simple “exact match” integration. The average American has between 3 and 8 email addresses, several variations of their name (married, abbreviated, misspelled etc.) and moves 12 times in their lifetime! The real challenge is to recognize all these variations as the same person. To give you an idea: if you don’t, you will mis-classify 70% of your best (top decile) customers - and needless to say, all your analytics will be off!

While we know that Neolane, the product Adobe acquired (that became Adobe Campaign), has some ability to de-duplicate data - it appears that these capabilities have not made it into the cloud-based version of Adobe Campaign yet, and are not yet available in the Core Services (and hence useless for any of the other Adobe “Marketing Cloud” products).

Marketers will get the most out of a marketing cloud solution that automatically integrates data from multiple sources such as CRM and Order Management Systems, while de-duplicating, cleansing and updating all customer data on at least a daily basis.

Adobe Campaign Standard or Adobe Campaign 6.1?

The latest version of Adobe’s multi-channel campaign automation product is Adobe Campaign v6.1. This is a  Windows based application (which can have an on-prem, cloud-based, or hybrid back end). Adobe announced a new, cloud-based version at the show: Campaign Standard. It has a flashy tablet focused UI, but very limited capabilities. Right now the cloud-version of Adobe Campaign can execute only email campaigns and is missing data cleansing capabilities. We are not aware of any true data cleansing as part of the “Core Services” either.  If you are interested in multi-channel or data quality, which frankly should be just about any modern marketer, you need to stay on Campaign v6.1. However, know that there is no upgrade path from v6.1 to the flashy, cloud-based Campaign Standard.

Our experience shows that today’s advanced marketers demand a fully cloud based solution that is truly multi-channel and is powered by complete, accurate and clean customer data profiles.

Adobe Campaign's Triggered Campaigns

There was a lot of talk about abandoned carts, predictive recommendations and machine learning at the Summit. Many of today’s multi-channel campaign management solutions allow marketers to send re-target emails to customers or potential customers that have abandoned carts. We agree with the focus on triggered campaigns: the most engaging campaigns are triggered at the right time by customer behavior (such as abandonment campaigns, but also post-purchase or replenishment campaigns and others which they didn’t talk about).

However, if you want to send advanced triggered campaigns with Adobe, you will need to buy at least two products: Adobe Analytics and Adobe Campaign. You can then import event information from Adobe Analytics into Adobe Campaign. Campaign customers that have not also purchased Analytics are not able to run website activity triggered campaigns.

Does Adobe Have Predictive Analytics?


Users of Adobe Campaign can send recommendations to their customers today using Campaign’s native rules-based recommendation engine. These are man-powered, not machine-powered recommendations. If you want to use machine-learning based (collaborative filtering - “people who liked x, also liked y and z) recommendations you will need to buy Adobe Target and incorporate Target’s more advanced recommendations into multi-channel messages executed with Adobe Campaign.

Clustering and propensity models

Adobe customers that want to take advantage of even more advanced predictive algorithms need to use Adobe Analytics Workbench for k-means clustering algorithms and to build out propensity scores, which can then be used in Adobe Campaign for targeting.

So while it’s true that Adobe Campaign customers can use predictive analytics to segment and target customers, fully executing these capabilities requires the purchase of not only Adobe Campaign, but also Adobe Analytics and Adobe Target. Customers of these solutions will then have to learn the intricacies of three different solutions and possibly involve consulting manpower or a full-time analyst to ensure that everything gets set up correctly so that they work together.

We believe that propensity (like “likelihood to buy”) and clustering are table stakes in marketing today - that can delight customers and deliver sometimes double or triple the revenues. Without likelihood to buy for example, you are not able to optimize discounts, which could result in 20% less profit. In our opinion, advanced recommendations, propensity models and clustering should all be standard features of modern multi-channel campaign management.

Customer-First or Product-First?

A final word on Adobe’s DNA: their success as a company has largely been dependent on the success of Adobe Creative Cloud and their Adobe Experience Manager, their content management system. So, it is no surprise then that many of their workflows are optimized around assets (content/products) rather than customers. It makes sense because many marketers are still thinking about their campaigns in terms of “finding people to buy your products” (or “finding people who might be interested in your content”). However, we feel that the crux of the customer-centric revolution underway is actually to organize around the customer.

The following slide summarizes the differences between a product-centric approach and a customer-centric approach:

customer centricity_pic2


We strongly believe that the customer-first approach is the right way to go. We even encourage companies to organize their team around the customer: some companies we know now have an acquisition team and a retention team for example and also allocate their budget that way. Existing customers really deserve your attention: reengaging an existing customer is ten times cheaper than acquiring a new one!

Adobe Marketing Cloud Review Summary

Adobe’s announcements represent significant advancements and thousands upon thousands of hours of development work. However, we believe that marketers are looking for a simpler, easy to use, more automated solution that will meet all of their multi-channel campaign management solution needs and provide predictive intelligence that achieves breakthrough results in topline returns from day one. Today’s marketers don’t want to buy a long list of different products, spend time negotiating for additional IT resources, or professional services budgets. Also, some major features like continuous data integration and cleansing in the cloud are missing from the Adobe brand - even if you buy all products. Other features, like truly predictive predictive recommendations, clusters and propensity models don’t come standard and have to be built using data scientists. As VentureBeat pointed out in their recent Marketing Cloud buying guide: “New, purpose-built solutions, like AgilOne, are giving incumbents a run for their money - with a broad scope of features and ease us use”.