The best thing I ever did as a marketer was make friends with my company’s head of data science.
I began my career managing the email marketing channel of a mid-stage e-commerce start-up, Craftsy (acquired and rebranded as Bluprint). Craftsy is a digital platform for tutorial videos in additional to physical products across 17 (and counting) different “craft” categories from knitting to woodworking. Consequently, I inherited an email list of nine million users divided into over a hundred segments. As with many scrappy start-ups, Craftsy tended to both manage and develop all back-end tools in-house. This led to countless late nights spent manually pulling personalized email lists and recommendation tables from our internal databases to optimize each marketing touchpoint.
Assuming not every marketer has a data scientist as a work friend, Dynamic Yield created an AI-powered personalization engine enabling a plethora of companies to better target customers. Dynamic Yield, closing its $38m Series D this month, uses the power of machine learning to enable “the rest of the digital world be just as optimized and personalized as Amazon.” 
Personalized marketing comes from a simple hypothesis. Showing people what they like increases the likelihood for engagement and eventual purchases. Keeping all other variables constant, the more personalized the marketing campaign or website experience, the higher the conversion rate. This customization requires the prediction of user behavior based on the actions of previous users; therefore, presenting an excellent use-case for machine-learning.
Machine learning utilizes data analysis to uncover patterns between user characteristics and preferences to further enable process improvement. Dynamic Yield has developed an entire product around the use of big data for targeted omnichannel marketing campaigns.
In recent years, many are grappling with rising customer acquisition costs as digitally-native brands and traditional brick and mortar giants compete for the market share. For many successful companies like Warby Parker or Blue Apron, customer lifetime value is declining as customer acquisition costs average in the hundreds of dollars.  Boasting over six hundred million unique users per month and across over two hundred brands, Dynamic Yield uses behavioral data across all of its clients to provide hassle-free omnichannel personalization. In the near term, the plan is to continue to build out the suite of APIs targeting every digital channel and customer touchpoint. 
Dynamic Yield’s product will inevitably get better over time, as more data leads to more accurate specifications and recommendations. Its long-term goal is to become the leader of the newly developed personalization market. Furthermore, Dynamic Yield plans on being everywhere the customer goes. This means “personalization APIs beyond web-based environments to kiosks, point-of-sale systems, call centers, IoT devices.”  Dynamic Yield advertises a majority of clients with a strong brick and mortar presence. Even amongst digitally-native brands, the current trend is a move towards experiential retail. Customers consider the physical store experience and the online presence prior to making a purchase. Dynamic Yield should definitely focus on the personalized integration of e-commerce and brick-and-mortar.
In the short-term, I recommend that they target clients struggling in the traditional brick-and-mortar space. I think failing retailers have the most to gain from Dynamic Yields offerings, and Dynamic Yield could use the access to the large databases. Long term, personalization goes well beyond pricing, web pages, and product recommendations. An individual’s preferences also stem from visual triggers. Dynamic Yield could potentially get ahead of the curve by using image recognition to personalize visual recommendations and the overall user experience.
Despite the promising capabilities of machine learning in consumer personalization, many companies are slow to adopt them. Dynamic Yield is working on bringing state-of-the-art personalization tools to growing companies, enabling them to compete with the likes of Amazon. From marketing to supply chain, retailers are under pressure to adopt machine learning marketing capabilities or risk becoming obsolete.
How should Dynamic Yield address potential conflicts of interest between two clients playing in the same market?
Dynamic Yield currently enables any e-commerce brand to have access to AI-powered omni-channel marketing. Assuming any brand without this capability will be short-lived, how will companies differentiate themselves once the user becomes accustomed to such robust personalization?
What kind of companies would benefit the most from Dynamic Yield’s omni-channel marketing capabilities?
 TechCrunch. (2018). Dynamic Yield, which builds Amazon-like personalisation for the rest of us, raises $38M. [online] Available at: https://techcrunch.com/2018/11/02/dynamic-yield-which-builds-amazon-like-personalisation-for-the-rest-of-us-raises-38m/ [Accessed 13 Nov. 2018].
 Fedyk, A. (2016) ‘How to Tell If Machine Learning Can Solve Your Business Problem’, Harvard Business Review Digital Articles, pp. 2–4. Available at: http://ezp-prod1.hul.harvard.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=120606181&site=ehost-live&scope=site (Accessed: 8 November 2018).
 Dennis, S. (2018). Unsustainable Customer Acquisition Costs Make Much Of Ecommerce Profit Proof. [online] Forbes. Available at: https://www.forbes.com/sites/stevendennis/2017/08/31/unsustainable-customer-acquisition-costs-make-much-of-ecommerce-profit-proof/ [Accessed 13 Nov. 2018].
 Yield, D. (2018). Dynamic Yield Raises $32M, Expanding its Platform to the Physical World. [online] Dynamic Yield. Available at: https://www.dynamicyield.com/2018/08/32m-series-d-funding/ [Accessed 13 Nov. 2018].
 Schiff, A. (2018). Dynamic Yield Snags $32M In Series D To Up The Ante On Personalization | AdExchanger. [online] AdExchanger. Available at: https://adexchanger.com/ecommerce-2/dynamic-yield-snags-32m-in-series-d-to-up-the-ante-on-personalization/ [Accessed 13 Nov. 2018].