No marketing professional wakes up in the morning and says, “Today is the day I’m going to alienate my customers.” Yet we do it every day, by leaning on old marketing techniques that have long outlived their usefulness and ignoring what our customers really care about. The frustrating part is that many of these mistakes are based on “proven” marketing techniques from the past. But times change, and what may have been the right thing to do twenty years ago can be the wrong thing to do right now. So, without further ado, here are nine mistakes that marketers make and that customers hate.
#1. Making data-based decisions on the wrong data.
Remember when getting a “single view of the customer” was the marketer’s mantra? It turns out that the value of that view depends a lot on the data you’re looking at. If you’re just focused on generic demographic data like age, income, and location, you could be missing what really makes your customers unique – their behavior and interests. A data attribute such as “likes sugary snacks” could be far more important if you’re selling M&Ms than the customer’s age.
#2. Feeding bad data to good technology.
Machine learning (ML) and artificial intelligence (AI) tools have the power to transform your marketing campaigns. But, like a car, if you feed them bad fuel you’ll get bad performance. Nowhere is this more visible than with third-party data sources. Time and again, I’ve seen companies struggle with underperforming campaigns only to trace their problems to outdated or inaccurate data that they bought. Be very cautious about what kind of data you use to train your ML/AI systems, or those initiatives could stall before they even get started.
#3. Putting too much hope in the next campaign.
Marketers spend a lot of time thinking about new campaigns and content, and too little thinking about how their actions affect customer lifetime value (CLV) and the sales funnel. The reality is that a new campaign isn’t always the solution to a campaign that isn’t working. Oftentimes, re-aligning the campaign goals to drive CLV or promote sales funnel movement can have a more positive impact than expecting that the next campaign will get you better results. Tweaking and improving current successful campaigns can have an even bigger impact as well. Too often marketers are afraid to play around with what’s already working and lose an opportunity to continue building on success.
#4. Fear of experimentation.
A corollary to #3 above is what I call a fear of experimentation. We are always looking for the next marketing guru who has all of the answers instead of looking for a person with an understanding of how to take data and create knowledge. Try hiring marketers who are willing to say that they don’t have the answer but know how to experiment and learn. If you make an assumption in a campaign and the assumption turns out to be false, try another assumption. Maybe you presumed an affinity where there wasn’t one, but you could discover a new affinity you didn’t see before on your second or third hypothesis. Too often, marketers assume that ineffective campaigns are the result of bad segmentation or poor messaging, but the segment and the message could be highly effective if matched differently. Of course, the ability to do this quickly is the key. The faster you can test these new hypotheses, the more likely you can seize on the opportunity while it’s still hot.
#5. Believing that more tools build better campaigns.
Marketers have accumulated a mountain of monolithic (multi-channel) tools in their MarTech stacks, but all it’s done is make personalization and content recommendation more complex. What marketers need to be doing is automate the MarTech stack by using technologies such as “headless” content management systems (i.e., CMS platforms that can be utilized by different groups easily), decisioning engines, and orchestration. In this way, marketers can automate the process as data is sent from the CMS, repurposed for recommendations, and delivered through experience tools (e.g., SendGrid, Twilio). You can think of this as a microservices-based marketing architecture, with best-of-breed services replacing best-of-breed tools but instead, marketers are buying bigger and more complex tools defeating the original purpose.
#6. Focusing on scores instead of features.
Demographics don’t drive decisions, and scores don’t reflect customer sentiment. Having a rich database describing the features of your products or services will help you understand the motivations behind the behavior of your customers. For example, if you’re a supermarket, knowing whether an item is vegan, low calorie, or eco-friendly will tell you more about a person and their buying habits than age or income range. These labels can then be fed into marketing decision engines to drive better recommendations and experiences.
#7. Mistaking personalization for something it isn’t.
Most marketers will tell you they’re already doing some type of personalization. But personalization isn’t putting someone’s name at the top of an email or the top of a webpage. Instead, it’s what Netflix does. And Spotify, Amazon, and Facebook too. They recommend content that is relevant to each individual user and deliver an amazing experience that consumers love. Amazon doesn’t use a “personalization web campaign tool” to customize their experience. They know what you like based on the actions you take and the content you consume and feed that knowledge into every interaction.
#8. Clinging to the “safety” of segmentation.
Marketers were trained that segmentation was the shortest path to personalization. When segments first arrived they were based on facts. Create a segment for all people who signed up for the newsletter. But then creating segments morphed into opinions – I believe people who downloaded these 4 assets would most likely want product X. This is not using data or machine learning to make decisions. This is using gut feel. There has been no testing of hypotheses. In practice, applying segmentation to a newsletter marketing campaign for one million customers is a recipe for failure. It doesn’t scale and it isn’t personal in any meaningful way. Real personalization, one-to-one messaging, requires a recommendation engine, experimentation, machine learning, and other “modern” marketing concepts. It’s a different mindset than the past, but it’s critical that marketers break the bad habit of leaning on segmentation as we know it today, as a way to get closer to customers.
9. Looking (not looking?) in the rearview mirror.
Stop looking at campaigns as being done or over. Instead, focus on those that are working and experiment to continue optimizing. Implement data-driven feedback loops as there may be huge opportunities in existing programs. Campaigns are not binary. Most companies run too many campaigns and instead should have fewer with more experimentation.
We all make mistakes. If some of these hit close to home, take heart: The only real mistake is to keep making the same ones over and over. Digital transformation has changed the rules of engagement for brands and their customers. But you don’t need to be a digital-native brand to behave like one. After all, data is data. It’s what data you’re using and what you do with it that makes all the difference. So what are you going to do with your data tomorrow?
