Big Data. Trending topic in the marketing arena for some years now. Encouraged by an industry that are making a nice buck, Big Data is supposed to be the goose that lays the golden egg. Their credo: Big Data builds future-proof marketing. Which, at least at this moment, is at least questionable. For why focus on Big (= 3rd party) Data when you are sitting on your own pot of gold, i.e. small (= 1st party) data? Apart from the chance of ending up getting lost in your data warehouse or drowning in your very own data lake? Working with your own small data allows for making serious steps in rendering your customers their personalized journey.
Damn, data really pays off!
That is what I found out back in 1992 when working for IBM’s International Logistics Center. Our mission: 100% customer satisfaction among our EMEA clientele through timely order delivery and information. A 200 headcount venture with people gaining their information from separate systems of all sorts. “There’s got to be an easier way”, were my first thoughts. So, after office hours I took the time to dive into the world of databases and queries, building reports to make my own life easier. After 3 months it had grown into a management information system enabling us to manage by exception. With all of my colleagues getting their specific reports in their mailboxes overnight. Just imagine what valuable information that yielded and, consequently, what huge time savings. From that moment onwards we spent that time on raising customer service levels. This aroused my appetite for more.
Activity-based costing was the next project: measuring the time employees needed for individual activities in our operational processes. Recently, ILC gained, with a lot of effort, a contract for product distribution in a specific market segment. An alleged cash cow. The reality the activity-based costing project revealed was that these activities did not make a positive contribution margin. Instead, they were loss-making. The divestment decision was easily made.
The next step I made was visualizing the incoming and outgoing supply chain. A report I made explained crystal clear that our suppliers’ shipment accuracy left a lot to be desired. This was the trigger to renegotiate all service level agreements. The result: an average inventory reduction by 35%, equaling structural cost savings in the amount of tens of millions of US Dollars. A welcome ‘present’ in a market with short-cycled product introductions resulting in obsolete products with, consequently, huge inventory write-offs.
The last example from the supply chain discipline: analyzing the transportation method for purchase orders I proved that, based on the volume/weight ratio, there ís a preferred shipping method. Above a specific ratio, it is cheaper to ship products by air instead of ocean. The result: yearly structural cost savings of millions of US Dollars. A nice side-effect was a considerable reduction of the total number of expedited purchase orders with related high transportation costs through the implementation of the principle that the responsible party pays.
All in all, this was ample proof that meaningful insights retrieved from 1st party data in one’s own systems allow for making a tremendous improvement with regards to competitive edge and yield. So, if this works in supply chain management why not in marketing?
Fruit salad or ‘apples vs. apples’
Years later, we talk 2005, I worked as an account director/strategist for a couple of car brands on a pan-European scale. CRM pur sang. Not only strategic advice but also omnichannel model introductions, brand activations and the implementation of a technical integrated CRM infrastructure with a Customer Data Platform as the centerpiece: one single database as the source to fuel all marketing and communication activities from 360° customer profiles. We worked, of course, in the triangle client-media agency-ourselves. Looking at campaign results at some point in time, it struck me the media agency reported figures that deviated from what we saw in our own Customer Data Platform.
I advised my clients to orchestrate their analytics activities in-house. We implemented Adobe SiteCatalyst, a true relief: no more discussions about which figures being correct, knowing that all analytics were based on the exact same definitions. No longer an indefinable fruit salad but being able to compare apples with apples. And, not insignificant: we were able to put targets on individual channels while measuring results (i.e. channel attribution). This allowed us to re-allocate budgets and pursue conversion optimization.
From insights to relevance
Some 10 years later I worked in a similar role for one of the big DIY players, where our main activities focused on email marketing. A powerful medium with the capabilities to make a substantial contribution to brick and mortar as well as online sales. Íf you make use of all of the capabilities and tailor content to the preferences and behavior of individual customers. To optimize email marketing activities I proposed an analysis of the database where we connected email behavior to online and offline cash register transactions through a personal ID. The client was not so much interested in their customers’ response to promotions, so the said. Out of curiosity, we decided to look into it nevertheless.
What struck us: as much as 40% of their turnover was promotion-driven! An insight that you may consider in the periodical re-evaluation of your brand positioning. The analysis also provided valuable insights in combinations of products purchased, seasonality, day in the week and time per day. At which specific store, both online and offline. This is all valuable information as an input to make your email marketing efforts more decisive and create more relevance on an individual customer level. Unmistakeably the value of 1st party data.
From crystal ball to data-driven predictions
Geese with golden eggs may exist in fairy tales, in real life I have not encountered them so far. Despite what software vendors make you believe there is no such thing as a comprehensive, all-over software solution that covers all your marketing needs. Hence, for Tommy Hilfiger, the choice for a technological best-of-breed solution, i.e. a Customer Data Platform in their existing system infrastructure, was easily made.
What in 2007 started with sending out email newsletters (with of course insight in open and click behavior and specific interests) has grown into the omnichannel MyTommy loyalty program. One of the first steps was adding purchase behavior in online and offline stores. Which provided interesting insights in historical apparel preferences. And a basis to adapt the content in email newsletters and banner campaigns to reflect individual customers’ preferences.
From that moment onwards we added well over a dozen additional data sources, thus creating rich and relevant customer profiles. Think data from a personal shopper app, data from customer satisfaction questionnaires and data gathered from a product recommendation engine. All 1st party data. This enabled us to tailor the Tommy Hilfiger collection even more to individual preferences and needs. Based upon insights in socio-demographic preferences, omnichannel purchase behavior and frequency (RFM), purchase behavior throughout time and interests in specific types of customer events.
This provided us with the ‘tools’ to develop different types of campaigns, tuned to specific and actual customer behavior. Such as onboarding programs, activation, and churn campaigns. Also, we are able to reward individual customers based upon their purchase behavior and brand engagement with special promotions and treats, such as VIP pre-sales and invitations for member-only events. A step that has brought awesome results: well over 4 mio active members that spend on average some 50% more than customers who did not register for the MyTommy program!
Recently, we implemented AI-based predictive models that feed event-driven campaigns and ongoing communication programs on a day-to-day basis. Thus bringing interesting insights: a clear view of customer lifetime value (CLV) helping us to give individual customers an even better brand experience. But we can also predict when leads are inclined to make their first purchase. When existing customers are likely to make their next purchase within a specific category. Or the chance for customers to return their purchase. The latter insight may be used to implement a business rule that shows shipping costs in the webshop check-out process for those customers who frequently return their purchases. With narrowing down numerous target groups to 4 distinct personas derived from the data and clear insights in product affinity Tommy Hilfiger both realized marketing cost savings and is capable of tuning their product suggestions, promotions, and invitations even better to individual customer needs.
Hence, we created a look into the future that allows Tommy Hilfiger to be increasingly relevant for individual customers. Connecting data from the recently launched MyTommy app – again 1st party data – is the next step in the evolution of Tommy Hilfiger as a love brand. And … of course we do not deny the value of 3rd party data: with selected data points from the DMP, customer profiles gain in value and, herewith, provide a basis to target look-a-likes of Tommy’s most loyal customers. In other words, spend the online advertising budget as efficiently as possible.
Shift to 1st gear
Despite a huge amount of attention for trending themes as Big Data and DMPs, I want to make a case for shifting focus from 3rd to 1st party data. Not only for reasons that your own data is still more reliable than 3rd party data, but also for the tremendous value you will be able to create for your customers when combining data from individual online and offline data sources to create a single customer view.
Move on from words to actions and shift to 1st gear. With a focus on 1st party data, you will be able to make valuable progress in creating brand preference among existing and in the end new customers. The right results will be yours. That’s a promise. The value of 1st party data: seeing is believing!
(this article was previously published on marketingfacts.nl)