Hello world! I’ve been dormant for nearly three years. I’ll be honest, I got tired of being in the public light. There are a lot more of you than there are of me. Even at our peak I still kept to a small team of six to twelve employees and devoted our time to a small select group of customers. I’m now going back through all my backlog of articles to refresh my thoughts on various topics, like marketing automation and what my thoughts are on the best practices of marketing automation in 2019 and 2020.
Ethical concerns about marketing automation
My concern is this, marketing automation is easy to lean too heavily on or the abuse. It’s been my experience over the past decade that various employers or clients will abuse to capabilities of e-mail marketing for instance. Or if establish overly frequent remarketing campaigns.
Marketing automation best practices
Marketing automation should enable scale and growth both vertical as well as horizontal. Are we able to easily duplicate efforts at scale? Can we effectively branch out new customer journeys or paths? What complex challenges are presented in the businesses marketing that could be automated? In the last three years, I’ve discovered that some of the best marketing automation is done in data preparations. For example, flagging customer records by date acquired and date of last activity to create a journey. There are a ton of applications that help facilitate this work, and more than one way to skin a cat so to speak. Marketing automation should be managed and handles by a person who is highly proficient in data sciences, as well as extensive marketing application knowledge. The most qualified person or employee in 2019 into 2020 is not only someone who has a strong affinity to the data sciences but can also understand the customer lifecycle and craft a creative campaign journey from start to finish. This ideal person should be able to give creative employees some direction on what assets are needed. In many instances, I’ve found that companies assets are underwhelming at best. You need a variety of product (or service) specific display ads, in virtually every size. You’re going to need to be able to multi-variant test video pre-roll creatives at different lengths (45 second, 30 second, 15 second, and 6-second bumpers). Companies also need to begin thinking about how to sequence their shorter form of media to create a personalized and dynamic experience. I think that many of the best practices in marketing automation that I had from 2016 to 2017 are still relevant. For example, the importance of having a machine-learned bidding strategy is essential. Your business must be building good quality and accurate marketing data models to excel in marketing automation.
I can never understate the importance of excellent quality creatives, very good brand standards are both a must have as a pre-requisite or starting point for taking advantage of marketing automation properly. You need to work on fleshing out where there are highly repetitive tasks and use data to inform your software. Make your programs more intelligent with the insights your business garners from Google Analytics or custom fields created in your CMS such as Salesforce. Harness the power of data layers for your marketing automation if you’re using a homegrown application. Find ways of joining data that would otherwise be fragmented or exist in a silo. I’m going to dedicate some portion of my time this year into documenting some of the research I’ve done over these past three years I’ve been somewhat radio silent. I hope you’ll all continue to read my blog, excuse my absence and tune in to my upcoming video series I’ll be publishing to DESIGNA’s YouTube Channel starting in 2020.
Thoughts on marketing automation from 2016
We’re there. It’s been over three years in concept, research, and development. Business technology and marketing automation are changing a lot. We now have more advanced access and easier use of API with webhooks than ever before. It’s an incredible feat because a year ago we were barely able to connect some software and services. Now API is native to practically every popular application for business. The truth is this, even with all these incredible hooks and features, marketing won’t be entirely automated any time really soon. What we can do however is exchange more data and information then utilize machine learning and self-awareness supported by input and feedback. So what exactly does this look like and what does it mean for companies? Well using data such as projected/estimated value on new business opportunities. For example, what if we could map your new opportunities precisely back to their sources of influence and amplify those. That’s exactly what machine learning is capable of.
Marketing automation and Google ads (bidding)
We adopted the use of Google’s target ROAS (return on ad spend) bidding in 2013 and have been working with it and our products to provide that type of optimal value to our customers. In the past three years, we’ve worked on building a road map of not only how software should exchange and use the information that relates to each part but also monitor the outcome of new opportunities created. To that end, we’re happy to announce that we’ll be rolling out CRM support in 2016. The support and product needed will be included to all our customers for use with their commerce metrics and insights to better optimize campaigns. Customers who have been early adopters and users in the past year have seen significant upticks in their overall return on investment from the marketing. Returns have doubled and even tripled in some cases. Marketing automation is using additional data and essential insights in business to act as signals for pushing and pulling levers in your AdWords campaigns – primarily controlling bids and budgets to influence impression share, increasing the overall share of top performers while highlighting deficient areas of your accounts and campaigns within. This allows us to optimize by knowing where more human capital effort needs to be applied, which is why I insist that for the time being marketing is not and will not be fully automated soon, but it is coming. As machines become more self-aware of design, styles and what is providing a greater return on investment from certain industries, many of those practices and successes carry over into other companies. It’s competitive in advertising but generally speaking, once you’ve managed millions of dollars you get the general gist of what works well for one company, can typically be applied in some fashion for another. This is a universal concept, not to say it always works or applies exactly the same way for one business as the next but let’s face it – companies have been using templates in all sorts of design and marketing for decades. Yes, some of the best have high production values and are certainly original, that remains important. Marketing automation will level the playing field for all businesses investing in advertising in 2016. DESIGNA is on the forefront of those discoveries because we’re willing to adapt quickly, implement changes from feedback and stay the course of bringing the best machine learning available; battle-tested tried and true.
Marketing automation results in 2016
The results are clear, our customers end up saving between 20% to 40% on advertising that is less influential and could otherwise go unnoticed. We also are seeing a better trend to reach target ROAS in a shorter amount of time than traditional methods. The reason is Google’s innovation with data science as it relates to advertising is a massive part of their business interest. In short, Google wants to provide a better experience for its users and businesses want to be connected with customers looking for their goods and service. We can utilize clusters of data collected through a variety of automated and manual input to determine what drivers help assist and cause conversion or new customer acquisitions. Typically we see a clear path visitors take, usually involving multiple marketing networks and channels. This data is referred to as an attribution model and is essential for machine learning and marketing automation because 9/10 times it’s a combination of placements or advertising networks that need to be amplified in order to increase ROAS. We will be supporting new, advanced and automated bidding methods in the new year; while growing our existing study using new software and data resources to help businesses effectively increase returns. Out of thirty case studies so far, 27 have been successful. Three were startups who may not have been the best for the studies due to constrained marketing budgets; we kept them part of the study for posterity sake. Part of our research revealed that impression share is still very important and has an indirect correlation to rising or falling conversion rates. Because impression share is so important, it’s important you’re not spreading budgets too thin and therefore we suggest consulting a marketing analyst to determine if you have a proper budget for your marketing, geography, and goals.