Can trust rust? I believe it can (pun intended), particularly if marketers over-automate their content. Authenticity is maintained in large part by humanizing marketing and customer service, and overly automated content can carry that scent of robotic in-attention which so quickly turns people off, regardless of their former product experience.
I come to this caution honestly. We recently posed a challenge with a mid-size technology firm’s marketing department: Which of the content marketing you are doing is most effective at each stage of the buyer journey? Our original intent was to study the success points, and build on or mimic them in other areas of the business. What we found, though, masked a larger problem of unhealthy dependency on automation tools.
The audit we conducted looked at response rates in absolute numbers (what content engaged best overall, regardless of life stage) and in context (which content was most engaging by key audiences at those key moments of truth which moved buyers along a journey). We analyzed the data by content type (video, webinar, whitepaper, blog post, email, social post, collateral, telephone, sales call, etc.), audience (buyer cohort, loosely defined by personas), product and buying stage.
Turns out that consistently, the best success resulted from careful listening. Anytime we got hazy about the customer attitude or perception, we got snagged. Automation was great to support an experience – lead someone to discovery, arm them with tools to enhance the experience, give them social outlets and encouragement to share, and follow up suggestions and feedback collection.
This level of automation worked well because it was specific to each experience itself, and tailored to the audience. Where it fell down was when there was not enough data to effectively predict the response, and the marketer relied on generic content or content created for some other, only somewhat relevant behavior.
This happens when marketers guess, assume or frankly haven’t yet created content specific to the emotional or cognitive life stage at the decision point, and push people toward a next best action that is not universal. Using predictive analytics helped develop those next action suggestions, but when they were applied to everyone, it fell down. Too much automation was not effective.
It makes sense, although trusting the data to guide us even when it doesn’t make sense is one of the most difficult marketing activities. When the automation was done based on individual preferences and behavior, it worked great. When it wasn’t, it creates a tin ear.
That sudden lack of understanding between brand and customer/prospect fails hard and fast. Similar to when we feel that suddenly a listener doesn’t “get me.” The faltering connection shows up in emails that promote products out of synch with the person’s interests. It is revealed when long form content is presented to someone who has heretofore only engaged with video or social. It’s the stuff of mockery when marketers presume to understand a person’s interests, only to turn them off fully.
“Look how wrong they were!” is a common #FAIL post on Facebook and Twitter. It seems people love to mock brands that get the content wrong.
There is always a risk of automation off inaccurate data. Marketers who automate out of ignorance of the customer or laziness in customization significantly increase that risk, and put their brand position in jeopardy. Sometimes we don’t have enough data to understand, and so we need to resist the urge to automate a specific buyer path until we do. Certainly all of us have resource constraints. We can’t address every individual twist and turn of the buyer journey and have to pick and choose the pathways that drive the most revenue and loyalty. However, at the least, it’s important to acknowledge that most buyer journeys are not linear or predictable, so aligning marketing messaging to one path or a set of rigid paths only means most of the rest of the recipients will get content that doesn’t match their non-linear journey.
In this client situation mentioned here, we learned quickly that over automation based on presumption does not work and actually seemed to erase the good branding impact built to that time. In response, we pulled back on some of the automated programs and focused on gathering more data. We watched behavior more closely during those life stages, and improved our targeting with fresh content sets. We also went back and re-evaluated some of the content we were using early in the cycle, creating new audience cohorts where there was significant breakage in the pathways to a sale.
Trust is the currency by which automation trades. People who trust a brand with their contact information or public support in the form of like or follow, are creating both opportunity and vulnerability in the new relationship. Marketers who assume too much, too fast, or fail to understand people throughout the buyer journey, do so at their own peril. A rusty relationship drives few sales.
What think? Does this jive with your own data? Where do you see the biggest opportunity this year to undo out-of-control automation, or to invest in more complete predictive analytics?
## ## ##
Reblogged 1 year ago from www.clickz.com