This article is a natural follow-up to the “Unifying Brand Marketing and Performance Marketing with Data” whitepaper published in January 2023 (*1 – link to source) and will try to explain how location intelligence should be used fundamentally in this inevitable unification. Location Intelligence is not the one panacea for misaligned measurement and interoperability between brand and performance marketing; however, it has paramount importance and should be implemented as soon as is possible to wherever it is needed. Again this is not a technical article, but one loaded with common sense. The aim of this is to make location intelligence undeniable, to provide enough volume of proof so as to make its adoption certain, axiomatic and with haste. Though this article will again predominantly feature marketing, location intelligence has multiple uses across different business processes such as retail strategy, organization modelling and sustainability intelligence. Here we will focus on marketing and customer intelligence, prove to you why your business needs this now and show you that its adoption will not only supercharge your marketing efforts, but keenly, give you the insights needed to grow your business with confidence as your customers evolve and their brand relationships change.
Importance of Unifying Brand & Performance Marketing
*2 : Marketing mix modelling by Meta Marketing Science 2022 – Meta, The Short- And Long-Term Impact Of Advertising
“Meta now says that it exists to build brands more than it does to drive sales.”
To begin with, a quick reminder on why unifying brand marketing and performance marketing is so important. In the recent whitepaper we highlighted the study by Analytics Partners which showed that by adding traditional media to digital media activation (or vice versa) the combination of effort would improve ROI by up to 35%. Here instead, let’s explore the recent Meta study on this, “The short and long-term impact of advertising” (*2 – link to source). In their study, 3500 campaigns were analysed running TV, Facebook & Instagram, online video, print, search and outdoor, and the conclusion says that digital media (FB & IG) actually accounts for 60% long-term ROI, while only 40% short-term. Let’s consider this for a moment, Meta made $113.6 billion last year with approximately 98% coming from its advertising. For over 15 years, Facebook relentlessly pushed its reach, targeting capabilities and social graph to consolidate a world leading position, dominating in performance marketing. Now, with changing times ahead (ATT, 3rd party cookie depreciation, GDPR and user privacy as standard) Meta has discovered that in fact its platforms are predominantly brand building ones, and that for every $1 spend there, $0.60 is actually predicted to drive long-term brand growth. This is huge news. Meta now says that it exists to build brands more than it does to drive sales. But why? Because brand marketing and performance marketing are symbiotic, because they do not exist in separate silos journeying toward different objectives, and because it is only when they are both fully understood, mapped perfectly in relation to one another, and accurately defined by data and each real effect on your business, does marketing truly succeed. This complete and final understanding is currently mostly empowered by marketing mix modelling and econometrics, where data sets are extrapolated out and confidence ranges are applied to keep the conclusions valid and true. But what if location intelligence could be applied here to take up some of the data heavy-load; what if we could map real results in real-time and access the marketing results across both brand marketing and performance marketing in an endlessly learning and results-improving model? Location intelligence is just this model. One thing remains an unerring constant in marketing activation and that is location; location of customer, location of supply/service, and location of advertising. Understand these three factors completely and you simply have to adjust course based on any changes to any one of these three fundamentals, changes that are knowable long in advance of the fact, meaning course adjustments are made with confidence and clarity.
History of Geographic Information Systems (GIS)
But this is all very theoretical still, where’s the actual working model? The model has been 60 years in the making, when in 1963 the first GIS (Geographic Information System) was created by Roger Tomlinson for the Canadian government as a means to manage Canada’s inventory of natural resources. Then in 1969, Esri was founded (Environmental Systems Research Institute) and up till today they provide the global standards for which global location intelligence services are judged. Jakala’s key strategic partnership with Esri was cemented in 2018 when we were awarded as their global partner for insights intelligence, an accolade we went on to receive again in 2019 and 2020. As one of the first MarTech companies in the world, Jakala developed tools and systems that would benefit clients when it come to their marketing, and one of the first products we developed was specifically for location intelligence in its relation to marketing, a product we went onto name J-Hexagon. But before we get ahead of ourselves, let’s review where we are right now. Brand marketing and performance marketing need to be united, and any tools to assist with this would be very welcome. Meta agrees and goes onto say that it is now predominantly a brand-building advertising platform. Ok, but what do we really mean when we talk about location intelligence? All the ad platforms, Meta, Google, TikTok etc have granular geo-targeting capabilities and brand lift surveying options, and when combined with your marketing mix modelling, you should get a pretty good understanding of what is really working in your media. But a really good understanding is probably not what you are looking for, instead, why not have a perfect (or near-perfect) understanding of all your marketing efforts. We will now explore a real-world example of when “really good” analytics can still leave a lot to answer for.
Google’s Brand Lift Methodology
*3 : Google Brand Lift – Measure Every Moment that Matters | YouTube Advertisers
In the interests of balance, let’s explore Google’s Brand Lift methodology (*3 – link to source). This service boasts the following: “robust methodology, easy implementation and near real-time results”, so let’s tackle these claims one by one. For “easy implementation” this is very typical of all Google services and very much part of their company DNA, this claim is therefore predictably accurate. For “robust methodology”, the main term here is robust, defined as strong and healthy, so let’s delve deeper. Google offers its advertisers the following unalterable brand study questions for YouTube:
- Ad Recall – Which of the following have you seen online video advertising for recently?
- Awareness – Which of the following have you heard of?
- Consideration – Which of these brands would you consider purchasing?
- Favorability – For which of the following brands do you have a positive opinion?
- Purchase Intent – Next time you have to choose from the following, which one are you most likely to choose?
This survey is then shared to two groups: 1. people who have seen your ads and 2. people who were eligible to see your ads, but didn’t see them. Let’s initially analyse the questions from the second group’s standpoint. You can choose up to 3 competitors and your own brand as answers, so regarding online videos recently seen, the user has 4 options max. Normally, due to the basic human psychology that we see our own company as more important than other companies, when choosing 3 competitors, we are likely to pick three (actually) bigger and better-known brands. Ideally at this point, Google wants the user to answer with the three competitors as they have not been served any ads, and fortunately for Google (and due to basic human psychology) this is normally what happens. This same pattern repeats out for all the subsequent questions. For group 1 (people who have seen your ads) Google now wants the reverse to happen, so let’s explore the language used more closely:
- Ad Recall – Which of the following have you seen online video advertising for recently?
The key word here is “recently”. Google knows exactly when each user will have seen an online video advert for each of the 4 brands, so by delivering the survey to the user at the right time, can increase the likelihood of their choosing the actual brand that is responsible for the survey. The same applies to the next question on Awareness. For Consideration and Favorability, this starts to add elements of creative, CTAs, targeting etc, but again is ultimately influenced by the mental narrative inculcated during the first two questions. Finally, the last question of Purchase Intent requires just one answer, so all things being equal, there would be a 25% chance of picking the surveying brand. However, the design of the proceeding questions need only to have engaged the user in such a way so as to make this likelihood slightly higher than it would be for group 2. A tiny increase in this likelihood will then be magnified out across the multiple surveys, making the results more likely to shine brighter on YouTube’s ability to improve brand uplift with its advertising offering. In practice, Google only lets you select 3 of the above questions, and due to the similarity of Awareness, Consideration and Favorability questions, the surveys normally include Ad Recall and Purchase Intent (as arguably the most useful) and then one of these three most similar questions. So is this methodology robust? In so far as it is “strong and healthy” yes, but is it perfect? Far from it. And finally to near real-time results, the survey results are generated “as early as 7 days”, but real-time indicates to me that something is ready now, immediately, in actual real-time. So brand lift studies supply back imperfect data, and data that is a week old at best, but what’s the alternative?
What To Map & Why?
The alternative is to map, track and report full funnel marketing results in actual real-time. Simpler from an online marketing perspective, all readily available in Google Analytics, to be tagged, filtered and pulled from dawn till dusk, each and every day. However, much harder from a traditional media standpoint, those offline activations of “broad reach ads that people find interesting and enjoyable” (*4) or is it? The one constant demonstrated by 60 years of location intelligence development is that when data is mapped, it is easier to understand and to forecast future behaviour with. By pinpointing a value in its real-world location, its relationship with other pinpoint values in real-world locations is more easily understood. It really is then a case of joining the dots, seeing how different behaviours affect others, and optimising each activation for the benefit of the whole. So what do you map and why?
- CRM data – this could be where your customers live, in order to see how a greater propensity of customers affect your marketing performance and vice versa
- Real Estate data – typically your store locations, to analyse marketing performance based on customer proximity to these locations
- Customer personas – no one understands your hero customer better than you do, so why not map these lookalikes for the benefit of your total marketing
- Service availability – make certain you can actually sell to your marketing prospects first and foremost, or perhaps add a radius to show “one hour delivery” potential say
- OOH ad locations – map these real-world locations to watch how it affects campaign performance nearby
- TV / radio regions – add these into your insights and add weightings for program type, time of day, channel etc
- Print adverts – add publisher circulations to match increased performance back to print ads and vice versa
- Event locations – map all your previous and upcoming events to see how they impact your advertising nearby
- 3rd party data – consider adding census data, Experian segments, Jakala targeting, or anything else you want to add in order to improve the whole
- Competitor data – store locations, service availability etc to compare their effects on your advertising
- ABM targets – map your target companies to match performance by your marketing tactics
J-Hexagon allows you to map all these variables and many more. In essence, whatever can be mapped, will be, and the more data layers added, then the more comprehensive the understanding and subsequent optimisation.
J-Hexagon Location Intelligence
*5 : J-Hexagon mapping example
“The results are compelling, improving campaign performance metrics by an average of 30% within the campaign’s first month”
J-Hexagon begins by analysing and optimising your Google campaigns, using the cumulative knowledge of every data layer that is added, it is then expanded out across all digital media advertising accounts. The results are compelling, improving campaign performance metrics by an average of 30% within the campaign’s first month, but so much more than that, seeing the performance in your online ads in relation to your print campaign, or the proximity to your retail locations, or your customer data, these insights make a 30% improvement look basic, which it fundamentally is, as this stage is only just the beginning. Looping back to Google Brand Lift, this service is in essence a means of plugging a knowledge gap. In the absence of a conversion, Google wants to show that an advert still has value, improving brand lift by a factor of “x” say. But this knowledge gap only exists when it is looked at from the silo of one platform. Instead, this gap should be filled with additional data touchpoints (CRM data, store locations, customer addresses etc) and then the business need to gauge “uplift” becomes secondary, as more actionable insights are already available and instantly put to work, repeating this process ad infinitum.
Instantly Actioned Optimisation from Geo Experiment Insights
Discovering the incremental value of your marketing is one of the most important learnings today for marketers across all industry types. Without an understanding of how an addition of marketing effort in one area will improve the performance, then it is very difficult to discern exactly what budgets should be committed where. Relying entirely on marketing mix modelling for this key strategic output is not advisable, nor is it necessary. There exist multiple methods of analsying and testing your marketing, so that learnings can be gleaned faster and acted upon sooner rather than later. Modelling incremental ROAS from a geo and time-based perspective can be done using models developed by Kerman, Wang and Vaver (*6) this understanding is the gold standard in comparative analysis and can be applied almost instantly at a low cost point. The learnings from these models should then be used to compliment your marketing mix modelling, as the data used is normally more recent and therefore more robust. From a pure platform perspective, geo experiments like this can be run quickly and efficiently. The brand Secret Escapes published an example of how it setup geo experiments on Google to discover if its econometrics were accurate in its CPL prediction for generic Search (*7 – link to source). The results showed that marketing mix modelling was incorrectly forecasting a CPL that was x2 higher than the actual geo test results, and this just goes to show that having multiple methods and models of tracking your incremental ad performance can help stave away costly misjudgments in budget allocation, misjudgments that are only the result of an absence of learning and the incorrect assumptions created by this absence. J-Hexagon applies these geo-based models in a real-time platform, not only filling any gaps in the marketing mix modelling, but also optimising the campaign there and then, improving the overall performance with this deeper location intelligence. In practice it applies multiple time-based and geo-based regression models, at a hyper-local level that is infinitely more granular than the advertising platforms. Not only are campaign learnings more profound and more robust, they are enacted to the benefit of the campaign, constantly optimising itself and creating new models with which to improve future learning. This is the methodology of J-Hexagon and why location intelligence is the most powerful tool to unify your marketing, not only unifying brand and performance marketing, but also the methods to track their successes and failures, individually, cumulatively and incrementally.
But this is not a sales push, this is a common sense push, as sometimes the two go naturally hand-in-hand. We are not saying you need to onboard new tech, nor are we saying your marketing mix modelling is amiss. We are not criticising your current marketing strategy, nor the methods with which you currently track it. We are recommending that you think about location intelligence as the means to map every dot that matters to your business and in a way that makes joining these dots much easier. At the height of the pandemic, we would tune in daily to see slide after slide presenting map after map of how the pandemic was then developing. Trying to imagine this presentation without the application of location intelligence is very difficult; perhaps the data could have been delivered in one super spreadsheet or in a spectacularly detailed infographic, either way, this would have been slower to understand and therefore slower to act upon. Location intelligence provides us real-time access to this data; data which is then analysed and acted on in a way that increases the probability of the desired outcome. In marketing, everything has a location; the customer has a location, as does the product/service, the marketing is engaged with at a location, and it is this location that has subtle (yet profound) influence on how that customer interacts with a brand. Only by layering in a deeper understanding of what is happening at a location can we begin to appreciate better all the factors that are at work there. If your brand marketing is not mapped in relation to your performance marketing then you will never be able to see (and influence) the myriad of inextricable factors at work between them. If your weekly analytics provides insights at city or town level, we say go deeper, factor in every possible customer touchpoint at actual location level and only then will a more complete understanding be available. J-Hexagon does just this, it is the location intelligence tool for marketing and marketers. It exists to make your marketing truly data-driven as the bridge between brand and performance. It is the unifying force that makes your marketing whole again to the benefit of your bottom line, empowering your growth tomorrow and your understanding today, in the here and now.
For a demo of J-Hexagon, please contact firstname.lastname@example.org
- *1 : Unifying Brand Marketing and Performance Marketing with Data – Jakala, Jan 2023 – link to source
- *2 : Marketing mix modelling by Meta Marketing Science 2022 – Meta, The Short- And Long-Term Impact Of Advertising – link to source
- *3 : Google Brand Lift – Measure Every Moment that Matters | YouTube Advertisers – link to source
- *4 : “Broad reach ads that people find interesting and enjoyable” – Quote from Les Binet and Peter Field, The Long and the Short of It
- *5 : J-Hexagon mapping example – J-Hexagon Location Intelligence
- *6 : Estimating Ad Effectiveness using Geo Experiments in a Time-Based Regression Framework – Jouni Kerman, Peng Wang, and Jon Vaver, Google Inc., Mar 2017
- *7 : Secret Escapes – Case Study, Think with Google, Aug 2017 – link to source
Jakala is 1,900 professionals working across 15 offices worldwide. We operate in four core service areas:
- On-life – including Brand and Performance advertising; influencer marketing; consumer engagement.
- Data – including ABM; DMP & CDP; analytics & big data; BI & data visualization; CRO.
- Creative & Content – including creative production; UX & UI; content creation; SEO.
- Reputation & Social – including branding; social media; employee ambassador; PR & events.
We are a Google Premier Partner (winning their award for Digital Transformation in 2021), Facebook Marketing Partner, Microsoft Select Partner, Salesforce Gold Partner (winning EMEA Fast Growing 2019-20), Magento Partner Award winner 2019-20 and ESRI Insights Award winner 2018-19-20 for location analytics.