How Real are Automated Multi-Channel Marketing Operations?

November 19, 2012 | John Berndt


Sophisticated marketing organizations have long since realized that it is a multi-channel world. Brands and messages are likely to be through multiple digital channels by customers, including at minimum web, email, social media, and Internet advertising. Customers may also pass those bit of information themselves, through various kinds of (also multi-channel) forwarding.

As much of print design fades in importance and the landscape of these channels clarifies, it’s daunting for organizations to find their way into multi-channel strategy, not to mention the operational challenges that come with trying to align writing, design and production and then follow-up in 4 – 8 discrete digital channels.

It’s been a long time that various pundits and vendors have been talking about the “holy grail” of centralized marketing automation that would create major efficiencies in the production, publication and tracking of multi-channel campaigns. The management of creative and content strategy for multi-channel marketing is a complex subject, and we can only touch on a few aspects here. I want to pause for a moment to try to assess the reality of these claims, and to pop the hood on what people are really doing today. Is the “holy grail” real or just so much… swamp gas?

One-Size Fits All Creative?

When does a new channel mean new content? And how can systems help make multiple versions of content for different channels efficient?

Marketing organizations need to be realistic about the level of effort involved in adapting their “creative” to multiple media—where an effective image, call to action, or block of text that worked somewhere else may be entirely inappropriate. Especially in the context of the new A/B testing capabilities in some systems (which often go under-utilized due to time constraints on production), there is great value in taking the time up front to make multi-channel production processes appropriate and sensible.

Campaigns can be said to fall into two categories “Parallel Channel” or “Channel Specific.” A channel specific campaign might be a “liking” contest that only works on Facebook. A parallel channel campaign is essentially an adaptation of the same creative campaign to different digital channels. It is typical for such adaptation to require a review of the size, tone, and visual design of the content—as well as the formulation of calls to action. Some aspects of this can be formalized in standards—such as the size (but not crop) of images, amount of content, and in some cases, tone—but often a creative review pass needs to take place to get a truly high quality result that works in the particular medium. Savvy organizations add something like 15% design review overhead for these differences.

How Automation Platforms Help

Automated platforms have significant potential to help with different aspects of the efficiency of campaign construction here, but because of the up-front investment that is required, many of these features are only used moderately. Those that do manage to get them in place can expect production value and faster time-to-market.

Some of the more promising areas include:

  • Using the better CMS products to manage both email and social content, via integrations with those systems. Basically, this amounts to streamlining and avoiding “cut and paste” to enter and categorize content, making your CMS the final source of your postings and emails.However, the need for editing between different multi-channel versions is standard, so don’t be surprised if you need to maintain separate fields in your CMS for “Twitter version” and “email lead story version”—reducing the total efficiencies just a bit.

  • Some of the better CMS products allow for mass creation of personalized emails from CMS content via metadata taxonomy, which can be hugely powerful for targeted marketing, though difficult to track. A secondary problem here is that many of the CMS platforms that have this capability have embedded email delivery programs that are not up to current best practices or quality, with an “out of box” approach committing you to an inferior sending product. It can take a significant integration to make CMS personalized content with an email marketing program you would actually want to use.

  • Content federation, driven by CMS automation, is the ability to reference and present content in multiple places. Sometimes, content simply plays a “supporting role” (as in related links) and doesn’t need to be adapted for the specific media. There are cases where it really makes sense to leverage this approach, without multi-channel editing, and use the same CMS to make similar content blocks show up in multiple campaign sites, websites, and web-based applications. The result, where appropriate, is generally faster delivery times and, of course, consistency. This consideration is most important for organizations that have large e-marketing activities, and/or are under a lot of regulatory content scrutiny— for instance, in medical or financial industries.

The Trouble with Tracking—Data Silos

One of the major efficiencies that is a part of the “holy grail” of multi-channel marketing is the idea of a single, top-down reporting view into the effectiveness of the various channels, campaigns, and cross-channel conversions. How real is this?

Let’s first focus on the mid-sized organization’s marketing infrastructure. With any luck, their website is being served from a good CMS (AKA WCM) product, usually with tracking in the form of advanced Google Analytics. They typically have an email marketing program (Exact Target, Constant Contact, Mail Chimp, or the like) for which the email campaign and mailing list are the unit of meaningful activity; somewhere in the shadows lurks a “social media consultant,” if you are lucky with tracking tools of some kind. Last but not least, behind the scenes there may be a customer database (CRM), which holds records from eCommerce and call centers.

Eight times out of ten in such situations these systems are all standalone silos. The process of creating meaningful reports that cross even two of them tends to be a very manual and limited process that gets done on special occasions, if at all. There are obvious exceptions—for instance, we see a lot of clients adding email tracking codes to their websites to at least allow for round-trip tracking of emails going out through final website “conversions” (purchase, email sign-up, etc.). Often the email program’s tracking tags and/or Google Analytics are the sole source of information about user conversions and cross-channel behavior, and the view is a minimal one.

Larger, better funded organizations have more options, and although they tend to do more, they also struggle with realizing that “bigger picture.” Some of the techniques we’ve seen work in larger organizations include:

  • Various degrees of integration with CRM, allowing for tracking of what channels a converted user has been contacted through. This can take many forms, but the simplest level is tags for various campaign sendings and conversion appended to the customer record.

  • Use of the more complex “marketing automation suites” that are either stand-alone email-centric products (for instance, Eloqua) or additions to larger CMS products. These systems allow for a workflow-like approach to customer engagements, essentially automating the issues of who to send what to when, and can be highly useful for aggregating tracking. However, the setup time and mappings needed to make this happen, as well as testing of the systems, are not small, and tend to happen in pieces.

  • Deeper integrations are often possible and needed, usually organized around a central reporting hub (depending on the product and degree of focus on anonymous or converted users, either a CRM or a marketing automation system). The difficult issue is that different systems have very different data structures (definitions of a user, a page, a goal, engagement value, categorization, etc. vary dramatically between systems).The key issues here are to focus on where the best reporting value will come from, and where the most strategic integration points are located. For instance, it might make sense to bring CRM data into a marketing automation suite in some cases, where in others, it would make sense to go the other direction, and pull marketing automation data back into CRM, where it can be used to flesh out a customer’s profile.

  • An ultimate best practice for such situations, though it comes at a high price and requires substantial expertise, is to move as much of marketing and customer data as possible into an OLAP (data warehousing) “cube,” a special database analyzed by more powerful business intelligence report-generation products. Rarely do we see this effort taking place in organizations with under $50M in annual revenues.The result of this practice, when it does happen, is potentially powerful, enabling an organization to have a more integrated, top-down view of multiple related data sets. There are two obvious downsides, though. First, the effort to reconcile different data models in reporting, though more supported by B.I. software, is a major job, and also somewhat subject to incompatibility problems when systems upgrade.Second, this approach separates the analytics data from any other aspects of the core software, creating a situation where it is possible to slice and dice data but not take immediate action. For instance, one might discern a trend that requires modification of the approach to email list segmentation or web conversion funnels, but not be able to make those changes without popping into another program, finding one’s place, and starting over.

Much more could be said about these issues, but hopefully this overview serves to give a sense of the lay-of-the land. The reality is, reconciliation of these various systems and data silos is possible and when it occurs in a usable way, highly valuable—but such projects almost always are a matter of degree, with many complex decisions taken along the way, multiple phases, and a high learning curve. When giant organizations attempt “big data,” top down integrations, they often take the form of secondary projects, after initial integration, to build complex reports.

The size and degree of sophistication of a marketing organization will indicate the extent to which it can—or should—attempt to build these integrations. Rarely do “out of the box” integrations do much, as they are often superficial and one-sided. The integration question then becomes a strategic one of “what” and “how much”—something best approached in terms of business problems to be solved, and with the help of someone highly experienced in such integrations who can give guidance.

For all those reasons and more, it is better to pursue the more actionable in-this-lifetime goals than the “holy grail” of complete integration. You’ll get something done, and you’ll be glad you did.

About the Author

John Berndt

I'm CEO of TBG and I've been thinking about the Web in creative ways since the year it began.

Leave A Reply

comments powered by Disqus