Author: Michael Andrews

Categories
Content Marketing

Finding the hidden ROI in content marketing

Content marketing is generating lots of buzz, and not a few questions.  Content marketing’s leading pitchman, Joe Pulizzi, recently published a book boldly entitled “Epic.”  Money is pouring into content marketing efforts, while executives ask if content marketing is a good long term strategy, and a sustainably profitable one.  There’s a lot of confusion about the ROI of content marketing because there’s still a lot of confusion about exactly what content marketing is, and what it can accomplish.  I believe content marketing has an important role, but I also believe many people expect the wrong things from it.  They are wasting money on things content marketing can’t deliver, and ignoring things content marketing can deliver.

Content marketing is an umbrella term that covers many different activities that in reality share little with one another.  The moment someone says “everyone’s doing it” we should ask ourselves whether definitions are overly loose.  Examples of content marketing range from email newsletters produced by small businesses, sales testimonials in whitepapers from widget manufacturers, to expensive social media gaming platforms created by billion dollar beverage brands.  The purpose of these activities can be vastly different, as will be the ROI they deliver.

Much of the buzz about content marketing highlights compelling examples of new genres of content that seem far more exciting than conventional marketing collateral:

  • Native ads: articles that look like normal journalism, and appear along side articles in leading publications
  • Branded content: videos, games and storytelling that are as fun as stuff you normally have to pay for
  • Communities: places where audiences can relate to each other about the topics that they are most passionate

Once the awe has quieted, people may reasonably wonder what the actual payoff is for producing such engaging content.  It sounds expensive, and it is.  Do brands spend big money giving away free content out of altruism, or perhaps out of vanity?  Not likely.  Does splashy content generate so much word of mouth that it becomes contagious and sales just happen?  Again, no.  Making a splash is harder than ever, and content initiatives increasingly succeed or fail based on who they reach, rather than how big an audience they reach.

Common definitions can mislead

While definitions of content marketing vary, they generally involve three aspects:

  1. that the content is owned and created by the brand
  2. that the content is meant to focus on audience or customer interests
  3. that the content delivers a result for the brand

Of those three items, only the first is clear cut.  The second, about being audience centric, is open to much interpretation.  The third, about ROI delivered, is even less clear, partly because of the squishy definition about what actually constitutes audience centric content.

As for defining audience-centric content, I prefer Google’s engagement guidelines, which ask: “What value do you offer consumers who visit your owned channels? Are you offering entertainment, education, peer recommendations and feedback?”  Note that no mention is made of imparting information.  Simply telling audiences what you are doing is not audience centric, no matter how well written.  They would never pay for that anyway, so offering it to them for free is hardly generous.  So content marketing definitions that describe audience-centric content being “relevant and valuable” are insufficient, because too often they assume the audience is in the mode of buying.  Even if someone is a potential customer who may eventually purchase an item, they may not think of themselves that way if they haven’t yet decided they want or need it.

In terms of the business payoff of audience-centric content, discussion is often aspirational rather than concrete.  There is much attention given to views and likes and shares, to big numbers or trending momentum.  As I will argue shortly, these metrics are a distraction from the actual business value of audience engagement.  The Content Management Institute furnishes unspecific guidance, setting an “objective of driving profitable customer action.”   Exactly how profits result is not detailed.  There is discussion about messaging to customers for “the 99 per cent of the time they aren’t ready to buy,” of “marketing less” or “selling less,” while at the same time trying to “change.. customer behavior” to drive “conversion” and boost sales.  Such statements, even if each sounds appealing on its own, don’t add up to a coherent story of how customers can at the same time not be shopping but persuaded to buy.  Stories are meant to help us buy into a rationale for taking action, but the plot here feels a bit sketchy.

Setting sales goals for content marketing is a mistake

When the discussion of content marketing turns to ROI specifically, the expected role of content gets more explicit, but at odds with the narrative that content marketing is about less selling.

What some marketers and CRM vendors call content marketing focuses on demand generation (creating awareness of products you sell) and lead generation (getting names of prospective buyers.)  Brands may believe they are being more customer-centric in how they talk, but the focus here is about promoting a product.

To justify the ROI, proponents may talk about content marketing as using content for in-bound marketing or out-bound marketing.  By directly tying content to sales, they continue old formulas that have been used since the days of direct mail or call center up-selling.  The scripts are longer, but messaging remains paramount. Proponents will suggest the key difference is that the marketing is no longer about campaigns, and no longer creates an abrupt interruption, because they are always doing it.  They are communicating marketing messages to their customers all the time, so now they feel they have a relationship.

But as the legendary marketing scholar Philip Kotler notes: “Marketing is too often confused with selling. ”

What many promoters consider to be content marketing, I would describe as long form advertising that happens to be written by the brand.  It seems like the perfect fix: use longer content to explain complex issues over time, when short ads don’t get noticed.  Like other advertising, its purpose is to increase sales, and it is failing if it doesn’t deliver sales growth.  So if you produce content that doesn’t ignite buying impulses, but you are expecting it to increase your sales, you will have an ROI problem.  Those nice infographics and video clips you produce look like a cost, rather than as an investment.

When the success of your content is measured by the sales it generates, you will make your content more and more sales oriented, to get it to deliver those goals.  It becomes a death spiral, as streams of sales oriented content tax the interest and attention of audiences, who tune out, so the marketers in turn amp up the sales message.  Rather than developing a sustained long-term relationship with audiences who might be inclined to identify with your brand’s vision, you instead are left with casual and fickle audiences who are checking out information as part of their prepurchase research.

Understanding the core value of content marketing

If sales revenue is not the right ROI metric for content marketing, which after all is supposed to be different from sales-oriented content, what other meaningful business outcomes can content deliver?  The purpose of investing in high quality content needs to be more specific than aiming to be “epic,” and needs to offer more justification than promising profits without explaining how it actually happens.

So let’s return to the apparent enigma of why big brands are spending big money on content that on its surface doesn’t seem to be about selling units of products.  What is the ROI of such spending?

Surprisingly, many marketers think about the ROI of content marketing by applying pre-digital concepts of product-focused marketing and making a sale, and ignore profitability factors.  They make little reference to data-driven digital marketing practices, or to the ad tech infrastructure that enables it.  They don’t talk about how data analytics of audience behavior can show the revenue value of customers.

Marketing today is less about driving a sale, and more about determining what specifically you sell to whom, with the goal of using available data to maximize profit yield across different segments.   Understanding the customer is more important than showcasing the product.

Large brands are interested in content marketing for two reasons:

  • it is an effective way to reach certain audience segments that offer attractive margins (the engaging media aspect)
  • it provides more immediacy to learn from and interact with these segments than traditional bought media (the owned content aspect)

Branded content exists to understand customers and establish a relationship with them, and is not about selling to them. A branded game, for example, can be effective content marketing, but won’t directly result in sales changes.  The popularity of branded content is best measured in the intensity that it engages audience members, that is, how much they use it and how much it means to them, rather than the numbers of audience members who see it.  While big audiences are a plus, branded content can be successful with small audiences, provided it attracts the right audience.

Until recently, brands had to rely on traditional media publishers to reach segments of individuals.  These segments didn’t define their identities in terms of the products sold by the brand, but instead define themselves by their attitudes, interests, and lifestyles.  Effective content marketing connects the brand with the concerns that matter most in people’s lives.

Refining the scope of content marketing

There are already many, sometimes contradictory definitions for content marketing, but none adequately addresses precisely how content marketing makes firms more profitable.  I want to offer my own definition, in the interest of clarifying what works, and doesn’t work, with content marketing. Branded content is probably a better term than content marketing, given the widespread misunderstanding about the difference between marketing and selling, but since content marketing is widely used already, I will stick with it.

My definition adheres to many of the intents of other definitions (focus on attracting interest, building relationships and serving the financial interests of the organization), but drops some of the more problematic aspects (trying to change people’s behavior, focusing on conversions and other transactional processes.)  If content strategy is about the functional side (don’t waste people’s time, give them exactly what they are seeking), then content marketing is about the emotional side (make them like being with you.)

Content marketing is content:

  1. owned and created by a brand
  2. that is enjoyable to use because it is meaningful or fun
  3. that builds long-term relationships
  4. with specific communities of people who have similar interests, life goals and other motivations
  5. who are attracted to the content (though do not ”need” it)
  6. because it persistently addresses their interests over time
  7. which are not related to any specific transaction needed from the brand
  8. and who would likely want to be customers with a brand sharing their values and understanding their needs
  9. which creates profit potential over the long term.

In short, content marketing is about offering emotionally engaging content that supports the larger, long term marketing objectives of a brand.   Content that is motivated by creating long term relationships can deliver long term profits.  The ROI depends on creating a long term relationship.

Content marketing can enhance long term profitability by furthering two key marketing needs:

  • market research
  • brand building

Focusing on these two needs addresses the core purpose of marketing, as defined by Philip Kotler: “Marketing is the science and art of exploring, creating, and delivering value to satisfy the needs of a target market at a profit.  Marketing identifies unfulfilled needs and desires. It defines, measures and quantifies the size of the identified market and the profit potential. It pinpoints which segments the company is capable of serving best and it designs and promotes the appropriate products and services.”

In a next post, I will cover how content marketing can support market research and brand building, both essential contributors to brand profitability.

— Michael Andrews

Categories
Agility Big Content

Making content updates an intelligent process

In the first part of this two-part post, “Why your content is never up to date,” I discussed how common approaches to managing out-of-date content are focused on first searching for content that’s dated, and then updating it as appropriate.  In this post, I want to explore how to prevent content from being becoming out-of-date.  Making sure content is always current requires more than willpower.  It requires more sophisticated tools than are widely available today.

Unfortunately, for all the bells an whistles in many content management systems, they are generally poorly designed to support real-time enterprise management of content’s “nowness”.  The intelligence of what’s up-to-date resides in the heads of the content creators, and the CMS is largely oblivious to what is involved with that judgment.  The cognitive load of having to keep track of how up-to-date content is, and why, is doubtless one of the frustrations that contributes to user disillusionment with CMSs.

Due to the limitations of existing tools, I will propose some new approaches.  In some cases, organizations will need to build new software tools and business processes themselves to enable proactive management of content.  While this option is not for everyone, it is clear to me that content innovation comes from publishers and not from the CMS industry, and that content leaders are often the ones who build their own solutions.

The solutions I propose fall in three main areas:

  1. understanding the temporal lifecycle of content elements
  2. developing more robust business rules for content
  3. building intelligence into content workflows

Why does content change?

Few organizations at the enterprise level have a good understanding of why their content changes over time, and how often.  Since they tend to devolve responsibility to individuals, they don’t monitor this dimension.  But without insights into what’s happening, they are unable to manage the process more effectively.  They need to understand what elements of content are routinely updated, what business areas those elements relate to, and how often the updates happen.

Organizations need forensic insights into content change.  Content can go through at least three patterns of changes in state:

  1. content that is thrown away because it is no longer useful
  2. content that is temporarily replaced by other content before returning, such as when a limited time offer replaces the standard offer
  3. content that is updated, and evolves from one state to another

The difference between throw-away content and revisable content may not be clear cut. Sometimes content is thrown away because it is too burdensome to revise.  Other times content looks like a revision when in reality is a repurposing of content about one product for use about another (a forking or mutation change).  It’s valuable to know what kinds of content change often (or should change often), and what about the content changes, to anticipate what is a problem area in terms of generating out-of-date content, or generating revision effort.

Gaining an understanding of what content changes is not typically developed during a content inventory, which is one of the few times organizations ever thoroughly examine their content.

Another challenge is to understanding change is knowing what level of detail to examine.  Even interviewing content owners about change will not necessarily reveal all the changes that happen.  Owners will likely focus on changes specific to their content, and then only the most substantive ones.  But changes relating to specific details can happen on a global basis, and can become tedious or worse.  The VP for Customer Relations, for example, one day may decide that henceforth all customers will no longer be described as “members” but instead as “guests.”

Most CMSs are not robust tracking the many content components that can change, such as the terminology used to describe a customer.   Content strategists often advocate structured modularity in content to help manage such issues.  Modularity can be helpful, but is infrequently practiced when it comes to embedded content — content within content. (Notable exception: CMSs optimized for structured online catalog content.) Some CMSs don’t support modular component embedding, and of those that do, they are often cumbersome for endusers.  To avoid having unstructured content embedded with larger content, some strategists recommend avoiding embedded content all together, for example, never having links in-line with the text body.  But scattering content elements in different places can degrade the audience experience.  Content creators reflexively embed content in other content to create a more naturalistic content experience, publishing content that feels integrated rather than fragmented.

A key need is to understand changes that happen within embedded content. Most CMSs don’t offer good visibility into how pervasively specific content components, structured or unstructured, are used across digital publications.  Conducting an analysis of how these components change will help your organization manage them better.

Ideally, a reliable and repeatable process for understanding change will involve something like this:

  1. a snapshot is taken of a consistent representative sample of content at different time intervals
  2. the sample snapshots are compared using file comparisons to identify what aspects of the content have changed over time
  3. the text of content that is found to have changed is analyzed as to its type, meaning and purpose
  4. patterns of change for components of content are identified according to element and the context in which it appeared, to provide a basis for developing content business rules
Example of CMS track changes functionality. It does not indicate what kinds of content components are being changed, or why
Example of CMS track changes functionality. It does not indicate what kinds of content components are being changed, or why.  Are the wording changes substantive, impacting other content, or merely stylistic?

Another area where most CMSs are weak relates to versioning content, especially at the component level.  There isn’t much intelligence relating to versioning in most CMSs.  Typically, the CMS auto-creates a new version each time there’s a revision, for whatever reason.  The version number is meaningless.  The publisher can “roll back” the version to a prior one in case there was a mistake, but you can’t see what was different about the content three versions ago compared with the current version.  Even for the few CMSs that let you track changes over time, there is no characterization of what the change represents, and why it was made.  A few CMSs let the author add comments to each version, but such free text entry is generally going to be idiosyncratic and not trackable at an aggregated level.  Comments might say something like “Revised wording based on Karen’s feedback” — meaningful in a local workgroup context perhaps, but not meaningful elsewhere.

At a minimum CMSs need to provide publication date-based version management, so that administrators can easily identify what content about a topic was published before or after a certain date.  This capability allows one to at least see how much content may be impacted by an event-driven change.  This is basic stuff, and easy to do, but it falls short of what’s actually needed.  It would be helpful to be able to apply conditions to such search, such as finding items published prior to a date with content containing some variable.

An even better solution would provide an easy way to record the business reason for the update.  These could be formalized as trackable data elements, that could be applied as a batch when clusters of content are updated at the same time.  Examples of reasons you might want to track are: product model change, warranty change, branding update, campaign language revision, etc.

Having such changes tracked will enable organizations to monitor how much updating is happening, and the status of updates.  It allows content owners to examine the status of all their content without having to read each item.

More robust business rules

As organizations begin to look at content updating as an organizational issue, instead of as the problem for individual content owners, the opportunity arises that different kinds of updates can be prioritized according to business value.  I would be surprised if many organizations today have an explicit policy on how to prioritize the updating of content.  Instead, it is common for updating to be based either on what’s easy to do, or what seems urgent based on immediate management prerogatives.

While any content that is out-of-date should be updated, provided the content has continuing value, it’s obvious that some content is more important than others.  Broken links are always a lousy experience, but unless they are on a high traffic page that’s a key part of a conversion funnel, they probably aren’t mission critical.

Different kinds of updates need to be characterized by their business criticality, and an estimation of effort involved with the update.  Errors and changes to regulatory, legal and price related content are business critical.  Changes to unstructured content, such as branding changes involving photographic imagery, often take longer when done on a large scale.  Each organization needs to develop its own prioritization based on its business factors and content readiness.

Once an organization has a better understanding of what drives content updates, it can begin to define business rules relating to content so that it is kept current.  The goal is to formalize the changes of state for content, so it can be better managed.

The content update analysis performed earlier will provide the foundation for the development of business rules. To do this, map the content changes you observe against the content contexts (larger content containers) and against a timeline.  Map what changing content elements (fragments of text, images, whatever) are associated with content types and topics.  Identify common patterns.  Some content elements will be used many places.  Some topics or types of content will have multiple changes associated with them at a given time, others will only experience minor changes.  After you have performed this analysis (using either a computer-based cluster tool, or doing it manually through affinity diagramming), you should start to see some common scenarios.  If it is not obvious why the updates occurred, work with content owners and other stakeholders to reconstruct what happened.  You should end up with a series of common scenarios that describe cases where your content requires updating.

From the scenarios, you will want to identify specific triggers that generate the need for updates.  This will be internal or external events that impact the content, or situations where some variable relating to the content has changed.

In the case of situational change (e.g., something changed, but the actor or the timing is not well defined), it is important to understand how small scale change can ripple through content.  Perhaps a product line has been renamed, or messaging has been revised slightly.  When such details impact many items of content, they should be managed through content templates where such details are structurally controlled.  There is always a trade off between the overhead of managing components and the efficiency of updating them.  Having a solid grasp of the relative frequency of items, their prevalence of use, and frequency of updates will allow content designers to strike an appropriate balance.  Even if such content elements are not all centrally managed, it is important to know where they are being used.

In the case of event triggered change, it is useful to characterize the types of events and associated actors, and the elements typically updated as a result.  Triggers can be internal, such as a new marketing campaign, sale of a division, the introduction of a new product line, or a new partnership.  Triggers may also be external: a new regulation, a dramatic market shift, or the adoption of third party guidelines.  Such events potentially impact multiple content elements, and involve more complex coordination.  By identifying typical events that impact content, as well as major corporate-level changes that may be less frequent but have huge consequences, you can build workflows needed to assure necessary updates happen.

These recommendations may appear simply to follow the principles of good content design.  But effective content design also needs to be transparent, so all stakeholders can understand the linkages, and status of updates.  Such visibility is essential to being able to revise the model as business requirements change.  Unfortunately, even in well designed content implementations, it is often difficult to understand what’s under the hood, and know how the pieces fit together.

Implementing a more intelligent approach

Content administrators, content owners, and the executives who depend on content to deliver business outcomes have common needs:

  • knowing what to do when updating is needed
  • knowing the status of updates
  • being sure their effort is efficient and effective

An effective process needs to accommodate the various parties who are involved in content updating.   One approach would be to empower a central team with lead responsibility for major update initiatives.  It might involve a command center or newsroom, where company initiatives that impact company content are identified, and the updates needed cascade through the organization.  Suppose the company announced a new initiative, or a change in policy. The central command center would query a database of content to identify impacted content.  If the changes were global, they could make the updates themselves.  If the changes impact selected content, the team would identify the specific content and send a notification to the content owner to make revisions. The notification would include a message about the business criticality of the update, the reason for needing the update, and an estimation of effort.  As updates are made, the team would monitor progress on a dashboard.  This approach assumes a degree of central management of content within an organization.

Another situation is when large scale content changes are unplanned.  Such changes might be harder for a central team to identify, especially if they arise from a peripheral division that doesn’t have a close relationship with the central team.  Suppose a content owner initiates an update that has an impact on other content she does not own.  Assuming this owner has authority to make such an update, there needs to be a way to alert other parties of the change.  Ideally, the content system will be smart enough to have a file conflict detection capability, so that it could spot a conflict between the revisions the content owner has made, and other similar instances of content.  The inspiration for this approach is the conflict detection capability in repositories such as GitHub, though the user experience would need to be radically simpler and more informative.  Complex, marked-up content is unquestionably more elaborate that the flat files managed in file repositories.  The task is not trivial, and there could be a lot of noise to overcome, such as false alarms, or missed alerts.  Having good taxonomic structure would be imperative.  But if it could work, the alert would serve two functions.  First, it would make the content owner aware that the change will make her content out of sync with other content, and ask for conformation of intent.  Second, it would trigger notification of the central team and affected content owners that updates are necessary.

Costs and opportunities of an intelligent process

The vision I have outlined is ambitious, and requires resources to realize.  No doubt some will object to its apparent complexity, the expense it might entail, the uncertainties of trying an approach that hasn’t already been thoroughly tested by many others.  Some CMS vendors might object that I undersell their product’s capabilities, that I am exaggerating the severity of the problem of keeping content up-to-date.  I can’t claim to be an authority on all the 1000+ CMSs available, but most I see seem to emphasize making themselves appear easy to use (“drag and drop inline editing!”), aiming to convince selection committees that content management should be no more complicated than an iPad game.   Vendors deemphasize harder questions of enterprise level productivity and long term strategic value.  Once installed, few endusers find their new CMS nearly as fun as they had hoped.  The emphasis on eye candy is an attempt to deflect that enduser unhappiness.

As I noted in the my earlier post, relying on existing approaches is simply not an effective option for large organizations.  It’s costly to always be playing catch up with content updates and never be on top of them.  Organizations that are always behind on updating their content miss business opportunities as they exhaust their staff. It’s risky not to know that all your content is even up to date: an expensive lawsuit could result.  Playing catch up impairs a business’s ability to operate agilely.

Yes resources are required to develop the capability to proactively update content before it becomes out-of-date.  But content has no value unless it is up-to-date, so there is little choice. In this era of mining big data and precision enterprise resource planning, it’s not unrealistic to expect more granular control over one’s content.  It’s not acceptable for large organizations to be presenting information to their customers that’s not the newest available.

I don’t assume my suggestions are the only approach to making the process more intelligent, but radical change of some kind seems needed.   If you agree this is a problem that needs new solutions, I encourage you to share your views on your favorite social media channel and encourage the development of something better.

— Michael Andrews