Mobile marketers across the globe recognize the massive importance of ad spend measurement. The ability to effectively collect ad spend data from media providers directly affects a marketer’s success on mobile.
But various events can skew your ad spend data as it travels from your ad networks into your analytics, distorting metrics, destroying the ability to target your most profitable audiences, and interfering with vital activities like creative analytics. As a result, collection of accurate and detailed spend data from ad partners is a non-trivial task that trips many marketing teams up.
It is a problem that Singular set out to solve for marketers more than seven years ago. In that time we’ve pioneered numerous technologies to automate the collection of accurate and detailed ad spend data directly from media providers in just about any form imaginable: API, export, PDF, screen-scraping, and more.
As the industry matures, and other analytics platforms start to recognize the importance of ad spend and ROI analysis, the time feels right to review the various spend collection methods being utilized in the mobile marketing industry and highlight the advantages as well as the limitations of each method.
In doing so, we hope to advance the growing dialogue on ad spend collection in the analytics ecosystem and continue pushing the industry to improve the handoff of marketing data from media providers to advertisers.
Overview of spend collection methods
Currently there are three main types of methods for collecting ad spend:
Direct: platform integrations
Semi-direct: exports and reports
Indirect: passing spend data in tracking link parameters (i.e. cost “macros”)
Indirect: passing spend data in server-to-server postbacks
Platform Integrations
In this method, media providers such as mobile ad networks report rich metadata and performance information through some form of programmatic data reporting, commonly a reporting API. In many cases, networks have multiple API endpoints that may serve different granularities, breakdowns, formats, or audiences.
Advantages
Platform integrations give marketers the ability to accurately match the media provider numbers, including cases in which data changes retroactively
Platform integrations give marketers access to a wealth of information beyond ad spend, such as additional performance metrics, creative data, targeting options and more
Platform integrations are the only way to integrate with the self-attributing networks (SANs): Facebook, Google, Twitter, Pinterest, Apple Search Ads and others
Platform integrations pass sensitive data is securely, server-to-server
Platform integrations provide data as quickly as it is available, and therefore quicker than any other method
Limitations
Platform integrations are harder to build and maintain
Platform integrations must map media provider identifiers to user data, requiring coordination between tracking links and data collected
Platform integrations can limit data update frequency – while some networks offer near real-time updates, others offer hourly or daily updates
Semi-direct
There are also cases where networks send data in email reports to complement some form of reporting that the API lacks. There are other cases in which dashboards and various types of exports (e.g., CSV via Amazon S3) complement reporting where an API is not available.
Advantages
Semi-direct at least gives you data … always a good thing
Semi-direct data is right from the ad network, so it should be accurate
Limitations
Semi-direct data may not be timely
Semi-direct data for one time period could be different in a later export as more data from extended attribution windows becomes accurate
Semi-direct methods can be brittle
Passing Spend Data in Tracking Link Parameters
With this method, marketers attach a few additional macros for cost data to the tracking links they create in their attribution platform (e.g., cost={...}&cost_model={...}). These links are built such that additional cost information is appended on top of every ad click (and ad impression, when view tags are supported).
While most larger networks support passing spend data through tracking links, many networks do not support this method. In addition, we’ve found that relying solely on tracking links to transmit cost data frequently leads to inaccuracies, which is why we recommend marketers complement data from tracking links with data from Platforms integrations, side-by-side, to ensure 100% accuracy and consistency.
Advantages
Tracking link parameters deliver a built-in capability to attach cost to individual user data
Tracking link parameters update data in near real-time
Tracking link parameters are simpler technology and relatively easy to maintain
Limitations
Tracking link parameters have inherent discrepancies with media providers – tracking links don’t ensure a 100% match with the network’s spend figures, and spend could differ from the actual invoices marketers receive
Tracking link parameters make it difficult to support cost reconciliations, retroactive data updates and discounts
Tracking link parameters are not applicable for self-attributing networks (like Facebook, Google, Twitter, Snap and others) as tracking links aren’t supported in these networks
Tracking link parameters make it challenging to support CPM & CPA campaigns:
CPM requires impression tags, which aren’t globally support yet, and due to sheer volume/inaccuracies will only increase discrepancies.
CPA is harder to support as cost is determined by a downstream metric or a set of downstream metrics, and there isn’t a clear way to define that at the link level
Passing spend data in postbacks
This method is similar to the tracking link method, however, instead of using tracking link parameters, media providers can send cost data through postbacks directly to the attribution provider. While we expect postbacks to deliver improvements over the tracking link method, other challenges (listed below) still remain unresolved.
Advantages
Postbacks deliver a built-in capability to attach cost to individual user data
Postbacks deliver data in near real-time
Postbacks offer support for all campaign types (as opposed to tracking link parameters)
Limitations
Postbacks suffer from inherent discrepancies with media providers – this method doesn’t ensure a 100% match with the network’s spend figures, and spend could differ from the actual invoices marketers receive
Postbacks make it difficult to support cost reconciliations, retroactive data updates, and discounts
Postbacks are not applicable for self-attributing networks like Facebook, Google, Twitter, Snap and others
Postbacks require development from the network, and not all networks have the resources, ability, or desire to change their ad server to fit these requirements, and as a result, coverage is still limited
Summary
As pioneers in this field, we are excited to see the increased awareness of the problem of marketing data collection. This is a problem we have been solving for our customers for over four years, and along the way we have seen the impact of our work: better collection techniques, new interfaces with media providers, and overall increases in granularity, speed and accuracy.
Our fundamental belief is that the best solution to the problem is the most comprehensive one: one that combines all available methods of ad spend and marketing data collection into a hybrid approach. Singular’s customers are some of the largest marketers in the world, and as such, we are held to the highest standards of delivery for accuracy, coverage, speed, and granularity.
Our promise to our customers and our ecosystem is to keep innovating, and tackling the problems to come. In fact, we have some groundbreaking innovations we are excited to share with the world in the upcoming months, and we can’t wait to tell you more about them.
To learn how Singular can solve for marketing data collection in your business, request a demo now.
Over the past decade we’ve seen the rise of the marketing technologist, who has one foot in the marketing department and another in engineering. And we’ve seen the data scientist role jump from almost nonexistent to being one of the fastest-growing jobs in just a decade.
Increasingly, as marketing is changing, technology is central to how marketers perform. Growth is now a key unifying function in brands and enterprise, and we’re also seeing the rise of the Chief Growth Officer.
We’re releasing a report on that in about a month.
But … our CEO Gadi Eliashiv gave a sneak peak at some of the results recently at Mobile Apps Unlocked in Las Vegas.
The rise of chief growth officers
Ultimately, the way chief growth officers lead their organizations is by using data-driven insights. Some of the most successful leaders drive those insights via marketing intelligence platforms like Singular.
The primary function of a marketing intelligence platform?
To provide insights for growth by connecting effort with outcome at granular and aggregate levels.
Ultimately, that’s how CGOs and other growth leaders get the score. Understand if they’re winning or losing. And know at both as high level and as granular as they want: how successful are our marketing, our campaigns, our ads, our creative.
Knowing that — and getting smart insights for optimization — powers breakthrough improvement in conversions and ROI. And that’s exactly what most brands, enterprises, and companies need.
In the simplest possible terms, a chief marketing officer’s role is to implement strategy that ultimately increases sales. A chief growth officer’s role is even simpler and more explicit: grow the company.
But how?
And what tools do they need to achieve those goals?
Singular is privileged to work with growth marketers at companies like Lyft, LinkedIn, Rovio, Wish, AirBnB, DraftKings, StitchFix, plus many more. We’ve seen what the best growth marketers the planet do, and we know what technology they use.
“Marketers are drowning in data,’ says Jo Ann Sanders, a VP at Optimizely.
That’s the problem.
“With the exponential growth of data over the past decade … it’s becoming harder daily to turn information into action,” says SurveyMonkey CMO Leela Srinivasan.
Marketers are drowning in data thanks to the unprecedented data exhaust of our digital lives.
We browse the web, we install apps, we watch four million videos on YouTube every minute, we search on Google 40,000 times a second. The world will soon have almost six billion mobile subscribers, and American adults now spend more than 3.5 hours a day on their phones in branded apps, sponsored media, and ad-supported sites.
At the same time, marketers are dealing with an exponential rise in tech tools, more digital channels than ever before, and more billion-user platforms every year.
Add in global competition, and 76% of CMOs say they can’t measure marketing performance accurately enough to make truly informed decisions.
Marketing intelligence platform
What marketers need most is actionable insights for growth. So CMOs’ (and CGOs’) biggest challenge is simply mining nuggets of gold from all that data. That requires real-time measurement and analysis at scale across potentially hundreds of platforms, partners, and channels.
That’s why Singular built what we call a Marketing Intelligence Platform.
The new marketers are different. They speak data and write code. They form hypotheses and run experiments; then measure results and optimize. These new marketers are marketing scientists, and they need tools of their trade.
With a Marketing Intelligence Platform, marketers achieve three critical things:
Unprecedented visibility at scale
On-demand flexible reporting
Full customer journey insights
That’s seeing not just your data, but your ROI on every activity. It’s slicing and dicing not just by campaign, but getting CAC per creative asset. And it’s measuring not just conversions, but cross-device and cross-platform journeys that led to customer action.
This requires at least nine components, combined into a single platform, grouped in three sections. We’ll take a very brief look at each. For a full in-depth overview, however, check out our complete Marketing Intelligence Platform report.
The three things that CGOs and CMOs need to drive and accelerate growth are …
One: Unified marketing data
You can’t get the golden nuggets of actionable insights without mining your data, and that starts by unifying it.
Unifying marketing data includes:
Data governance
Data ingestion
Data processing
Attribution
Dimensional data combining/synthesis
Data governance ensures clean data from every source, and enables processing, enriching, and combining later on.
Ingestion is getting all your relevant data from every source, and it’s not easy. Processing is essential to standardize and normalize it, at which point you can conversion outputs to marketing inputs. Combining and synthesizing top-funnel and low-funnel data reveals deeper trends and granular results.
Two: Intelligent insights at scale
At a high level, marketers need to know the score: across all their campaigns, are they winning or losing? At more granular levels, they need to know if a specific campaign, partner, publisher, or creative is performing.
Generating intelligence insights includes:
Reporting and visualization
Actionable insights
Reporting and visualization shows marketers what’s happening, and actionable insights provide clues for future profitable growth. Some of those insights are pull, but some need to be push: alerts about out-of-scope campaigns, click-through rate drops, poorly performing ad partners, and so on.
Three: Automation
The volume of data flooding marketers’ dashboards, reports, and spreadsheets cannot be handled manually at scale. Automation is required, and it includes:
Data transport
Alerts, fraud, audiences
And much more
It is not useful to have a system that only ingests data. Marketing data needs to move from systems of deployment to systems of analysis to systems of engagement, and sometimes in multiple directions. So building in the ability to do that via API, exports, or S3 to internal BI systems and hundreds if not thousands of external partner systems is critical.
And while modern scientific marketing is not a set-it-and-forget-it activity, marketers increasingly need to be able to automate actions within set parameters.
That includes automated creation and distribution of audiences for retargeting, look-alike campaigns, or suppression lists. It also includes built-in on-by-default configurable mitigation of fraud, along with both whitelisting and blacklisting of sources and publishers in paid media campaigns.
And at higher levels, it includes automation of bids and buys for ad campaigns at scale.
Results: what a marketing intelligence platform delivers
What does a marketing intelligence platform deliver?
Find out soon in part two of this blog post, coming next week.
How do you grow ROI while maintaining CPA and scale?
This is a question marketers face every day. And answering this question has become more complex as they advertise on more platforms across more devices than ever before. When conversions happen, it’s a struggle to connect the dots and understand what caused them.
Back when Singular was founded in 2014, we focused on solving this challenge first for the complex, highly fragmented, mobile ecosystem: providing a single solution that automatically collects and combines spend data and conversion data to expose mobile marketing performance, including ROI, at unrivaled levels of granularity.
That is powerful. And we quickly became the de facto solution for unifying campaign analytics and mobile attribution to expose ROI.
But in 2019, the game is different
Top brands advertise over a wide range of platforms to users on multiple devices. A customer may see an advertisement for a product on her desktop, and later buy that product on her mobile app. With today’s analytics, it’s hard to connect the two experiences and measure the customer journey accurately.
For mobile-first brands, this often leads to two separate teams, one web, one mobile app, using different tools, and even different metrics, to measure the customer journey. For web-first brands, it results in limited investment in mobile apps, preventing them from diversifying their marketing efforts to bring in incremental users, leaving untapped growth potential on the line.
Moreover, inaccurate measurement leads to misguided decision-making. Matter of fact, poor data quality costs brands an average of $15 million annually, according to Gartner. Making an investment and creative decisions with inaccurate and incomplete datasets is just plain costly.
In true Singular spirit, we sought to solve this new challenge for our customers so they can drive growth more effectively and efficiently in this multichannel world. And I’m happy to say that we have leveraged our vast experience in attribution and marketing analytics to do just that.
Cross-device, cross-platform attribution
Today, Singular is announcing the first-ever cross-platform and cross-device ROI analytics solution for growth marketers.
With the release of Cross-Device Attribution, Singular’s Marketing Intelligence Platform connects marketing spend data to conversion results across devices and platforms. First, we ingest granular spend and marketing data from thousands of sources. Then we connect it with attribution data from our easy-to-implement in-app and web SDKs as well as direct integrations with customer data platforms, analytics solutions, and internal BI systems, bringing the full customer journey into a single view. Finally, we match the two datasets.
The result is the most accurate cohort ROI and CPA metrics available to marketers, at the deepest levels of granularity including campaign, publisher and even creative.
That’s ground-breaking. It’s revolutionary.
But bringing cross-device and cross-platform ROI into Singular and measuring it accurately, at granular levels, is only the beginning to driving impactful growth.
Granular data for growth
Marketers can now access granular ROI cohort reporting that is more accurate than ever, as you can get clear, combined revenue for users across all devices. This is critical to achieving profitable growth and only possible with Singular – a complete platform that innovates beyond a single attribution solution.
Moreover, marketers can also utilize the wide set of capabilities that Singular’s Marketing Intelligence Platform offers to make smarter decisions and optimize their growth efforts with additional cross-device visibility; plus, they have more visibility into essential context such as the exact creative customers engaged with and the audience segments they belong to.
For example, you may find that a web channel’s impact is much higher than expected for specific types of customers. And now you can analyze the impact of the same creative across mobile and web.
In fact, we won’t be surprised if marketers start shifting investments with this new level of clarity. We are excited to see how growth strategists are going to rise above the crowd using this new solution to become part of the future wave of sophisticated marketers. Gone are the days of attribution feature wars – Marketing Intelligence has arrived.
Launching Cross-Device Attribution is just another step towards achieving our goal: to be every marketer’s indispensable tool in driving growth. We keep working not only to ensure that you can innovate your growth processes and have access to the highest data accuracy but also to ensure that we bring you the right insights at the right time to help you make timely strategic and operational decisions.
Are you ready to take part in the future of growth?
What do brands need most out of their marketing data in 2019?
Actionable insights, consumer privacy protection, and full marketing data unification, chief marketing officers say.
I recently asked 199 CMOs, VPs of marketing, and other marketing leaders what their biggest challenges for marketing data will be in 2019 for a story in Inc. (There was far too much to write there, so this post became necessary.) Tops on marketers’ lists of priorities? Actionable insights in an avalanche of data. But just behind it in today’s climate of consumer privacy breaches was privacy – and trust.
Here’s how Felicity Carson, CMO for IBM’s Watson division, put it:
“Among all the marketing data challenges, the biggest in 2019 will be how marketers instill trust in data – both for the marketing discipline and customers – balanced with the need to improve customer experience.”
The 800-lb gorilla in the room?
Marketers have far too much data already. That’s a consumer privacy risk, but it’s also a potential marketing intelligence nightmare.
“Marketers are drowning in data from various analytics systems,’ says Jo Ann Sanders, VP of Product Marketing at Optimizely. “What marketers are going to have to do going forward … is to go beyond analytics data … and adopt new, agile test and learn practices.”
Marketers don’t need more data.
What they need are actionable insights drawn from the data they already have. Marketers’ third priority, unifying all their marketing data, will help.
“With the exponential growth of data over the past decade and into the new year, it’s becoming harder daily to turn information into action,” says SurveyMonkey CMO Leela Srinivasan. “While more data has the potential to deliver more meaningful insights, prioritizing an action plan to address it is critical.”
Consumer privacy and data security
Insights are essential for growth, that’s clear.
But a strong brand untainted by consumer privacy breaches is also essential for growth. Anyone who feels otherwise, just ask any company that experienced a privacy breach in 2018 … and look at its stock price impact.
That’s why, almost shockingly, marketers’ second-biggest concern has now become consumer privacy, the security of consumer data that brands now possess, and regaining the trust of their customers.
“The single biggest challenge B2B marketers face in the coming year will be balancing privacy and personalization to regain the trust of their audiences,” says Penny Wilson, CMO of social media marketing platform Hootsuite. “That starts with respecting [consumers’] privacy, being open and transparent about when and why data is collected, and then leveraging the data that customers are willing to share to create personalized one-to-one experiences that deliver unique value.”
“Going forward, brands must focus less on maximizing reach, and more on generating transparent, quality engagements that add value to their customers,” Wilson adds.
This is not business as usual for marketers and advertisers, who have typically wanted as much data as possible. In fact, a new “social contract” between brands and consumers will become so important, says Lloyd Adams, SVP at SAP North America, that data ethics will become more important than data analytics.
Unifying marketing data: a top-3 priority
What else do marketers care about?
Not far behind privacy/security/trust, marketers rank unifying marketing data as a top-three priority.
The challenge is obvious.
In a universe of 7,000 marketing and advertising technology tools, marketers are both doing and learning so much more from their prospects and customers. But most of those actions and insights are being generated in siloed, dispersed systems.
“The problem is, we’ve put too many tools in place to collect and analyze marketing data that are too hard to use and it’s causing a lot of frustration,” says Tim Minahan, CMO of Citrix. “Marketing professionals are spending way too much time searching for information and clicking through multiple pages in applications to gather the insights they need to design, execute, and measure effective campaigns.”
The result is not pretty.
“Everyone’s data is a mess,” says Peter Reinhardt, CEO of Segment.
Identifying insights from your marketing data and then unifying them for a single view of customers – and a unified understanding of marketing success – is critical to cleaning it up.
“Data lives in different places — sales, customer service, digital marketing,” says Selligent Marketing Cloud CEO John Hernandez. “The biggest data-related challenge [for 2019] will be consolidation and a full 360-degree view of the customer relationship.”
That’s a difficult challenge, Hernandez says, and CMOs agree.
And in fact, not only is it hard right now … it’s getting harder.
“The biggest issue with marketing data is federating it into a meaningful whole picture,” says Eric Quanstrom, CMO of Cience. “As CMO, I live in (literally) a dozen different dashboards, daily. And that number is growing.”
Marketing data used to be fairly simple: survey data, market data, customer data, product use data, and probabalistic reporting on ad performance in traditional channels. New digital channels offer deterministic reporting possibilities, but with web and mobile and apps and wearables and IoT – to say nothing of platform proliferation like email and social and messaging and search – it’s getting harder.
And all of that proliferation leads to siloed data sets.
The problem with siloed data is that fragmentation subverts complementarity, says Rebecca Mahoney, CMO at MiQ. When data isn’t complementary and doesn’t add up to a complete picture, the marketing results is an inability to detect new opportunities, or see weak links in existing marketing campaigns, she says.
Data lakes may not save marketers, says Daniel Jaye, founder of Aqfer, a data lake provider.
In fact, they can actually exacerbate the problem because most data lakes inevitably become data swamps, Jaye says. Widespread data proliferation, chaotic file partitioning and sharding practices, and the lack of traditional data management tools all cost marketers the opportunity to achieve integrated insights.
Marketing intelligence unifies data insights
But there is hope.
Good marketing data practice does result in growth.
“With a holistic view of data, powered by marketing intelligence, campaign performance will drastically improve, and otherwise unidentified business opportunities will become unlocked,” Mahoney says.
It’s true that not every marketer will have a single marketing cloud for all their marketing technology and data needs. And even most marketing cloud customers also use additional tools to engage and understand their customers.
That suggests that centralization of marketing insights, particularly on paid but also on organic marketing efforts, is what will help marketers the most. Engagement happens where the customers are and data lives in the tools a brand uses to connect with them. Marketing intelligence aggregates then insights from the entire gamut of customer engagement into one single unified view.
Marketers are also concerned about the quality of the data that they have, and its accuracy. 13% said that accuracy was a top concern in 2019. Another 12% said they have too much data.
“In many ways, marketing has too much data on its hands,” says David Meiselman, the CMO of corporate catering company exCater.
As Citrix CMO Tim Minahan said above, we’ve put too many tools in place to collect and analyze marketing data. The result is frustration.
A potential savior?
Artificial intelligence.
“We … believe marketing and customer engagement will be an excellent first use-case for enterprise AI,” says Patricia Nagle, CMO at OpenText. “AI systems can analyse structured and unstructured data to identify opportunities for marketing outreach, customer support, and other actions that enhance overall customer experience.”
That’s true, and AI is a tool that marketing is already seeing results from in fraud reduction, creative reporting, and other areas.
But it does some with some dangers as well.
“Deep learning models have been shown to be vulnerable to imperceptible perturbations in data, that dupe models into making wrong predictions or classifications,” says Prasad Chalasani, Chief Scientist at MediaMath. “With the growing reliance on large datasets, AI systems will need to guard against such attacks on data, and the savviest advertisers will increasingly look into adversarial ML techniques to train models to be robust against such attacks.”
And finally … all the other quotes
When you ask 200 top marketers for their insights, you get a lot of insights. And they’re too good to bury.
So here are many of the additional quotes that marketers provided, broken down into categories that I’ve chosen. Some of them are partially referenced above, but are given in complete form here. Each of the responses is answering a simple question:
What are brands’ biggest challenges with marketing data in 2019?
Marketers need: Data accuracy and quality
Peter Reinhardt, co-founder and CEO of Segment
The biggest challenge for marketing data in 2019 will be data correctness. Everyone’s data is a mess. Consumers are bombarded with tons of noise, much of it based on wrong data, names, and locations. As a result, customers are burned out. It doesn’t matter how much a company invests in personalization if the underlying data is incorrect. For businesses to truly succeed in 2019 and beyond, they need to prioritize making sure their data is clean and accurate.
Martha E Krejci, The Tribefinder
The biggest challenge with marketing data in 2019 will be determining how good the data really is. Before this rise in cookie awareness people weren’t really flushing the cookies or clearing their cache as much, which lent itself to long-standing good demographic data. Now, the data isn’t as deep, therefore not as reliable. In 2019, businesses will need to learn to re-target.
Not surprisingly, war against fraud will be the biggest challenges mobile marketers will have to face. In addition to common fraud cases like SDK spoofing and click spamming, more and more new types of fraud will stop developers from obtaining real users. Our job as mobile marketers is to keep educating ourselves about different types of fraud and the pattern of each fraud cases, so that we can take a proper action when we find them.
Marketers need: Actionable insights and marketing intelligence
The best tools solve the disconnect between data, insight, and action, incorporating multiple sources of data to execute, allocate, attribute and optimize digital campaigns across channels.
Marketing data still struggles with insights and it would be amazing to see more of a focus on this essential craft. There are endless tools for gathering the WHAT – numbers and histories and basic information about your customers – but very little that helps us figure out the WHY. The big challenge in 2019 (and likely for a few more years) is going to be training people to understand how to read the what to get to that why.
Phil Gerbyshak, Digital Selling and Marketing Strategist
With all the data collected, the biggest challenge with all the marketing data is finding the most meaningful data, and then figuring out the most actionable insight from that meaningful data. Too often reports for reports sake are created, even with AI to help us find the patterns. Taking the time to think about what you want to accomplish and setting up your data accordingly will challenge marketers and delight stakeholders in 2019 and beyond.
There is so much data out there that ‘big data’ is no longer the priority; there is a need for actionable data that means something to marketers. The other challenge is that the biggest winners on the platform side are increasingly closed and stingy with their access, which may be necessary for consumers and benefits the platforms but hurts marketers. Finally, marketers will have to grapple with a savvier base of consumers who are constantly reading mainstream press coverage about data abuses; marketers will need to determine how cautious they want to be with collecting and accessing consumer information.
What is the biggest data challenge for marketers in 2019?
Building actionable results based off of accurate data. We continue to see an inability of our clients to properly read analytics and come to assumptions. I hope continued AI and machine learning will add tools to assist.
Felicity Carson, CMO, IBM Watson Customer Engagement
Among all the marketing data challenges, the biggest in 2019 will be how marketers instill trust in data – both for the marketing discipline and customers – balanced with the need to improve customer experience by identifying meaningful patterns buried deep within the deluge of data. Compounding this challenge is the need to break down compartmentalized martech and adtech stacks that house this information, coupled with the need to have contextualized understanding of aggregated customer data across the organization such as commerce and digital teams. Marketing teams will need to rely on AI to achieve this level of high performance at scale, particularly in the new era of the ‘Emotion Economy’ that requires organizations to engage with customers in relevant ways on issues that personally matter to them.
The biggest challenge with marketing data in 2019 will be to decipher which of the outlier [datapoints] are leading indicators into new market growth.
Leela Srinivasan, CMO of SurveyMonkey
Today, we have access to more data than ever before, but with the exponential growth of data over the past decade and into the new year, it’s becoming harder daily to turn information into action. A study by IDC bleakly projects that by the end of 2025 only 15% of global data will be tagged; of that, only 20% will be analyzed and approximately 6% will be useful.
While more data has the potential to deliver more meaningful insights, prioritizing an action plan to address it is critical. In 2019, B2B marketers will be laser-focused on finding a way to cut through the massive troves of data available and identify the insights that matter most.
The biggest challenge with marketing data is the ability to find the useful insights to create concrete actionable next steps. We have so much data, but not enough of an efficient way to sift through the noise accurately for truly useful data.
The biggest challenges we have as a startup are: having easy-to-use data that doesn’t require a ton of training like a Nielsen or IRI platform and being able to triangulate all of our data among a two-person team. Right now, we are using about 8 different vendors to analyze one or more stages within the customer lifecycle, and while it’s great to have all of this data, it’s not easy to triangulate it. It would be great to have 1 platform that could assess all or most of our marketing program’s performance, but those platforms usually come with a large price tag that we can’t afford.
One of the biggest challenges marketers will face in 2019 is how to better analyze consumer behavior and turn those insights into effective marketing. Consumers spend 40 hours a month and three hours a day in apps, mobile time spent will surpass time spent in TV in 2019, so marketers need to change their traditional planning behaviors for this brave new world. They must integrate their traditional teams with their digital teams, combine their video teams into one cohesive team, and integrate the data across all channels so that they can be smarter about how they find their most valuable customers. It is also getting harder with privacy, however, companies with strong technology especially predictive algorithms can predict users intentions based on less data. We are just scratching the surface on data analysis and with new data privacy laws, this challenge will only get harder.
Tim Minahan, Chief Marketing Officer and SVP, Citrix
Every marketing challenge can be whittled down to a mathematical equation – whether it’s measuring customer sentiment, tracking conversions, or weighing the return on a particular campaign. Data-driven marketing can eliminate much of the he-said/she-said friction that has historically muddied sales and marketing relationships. It can cut through emotional biases and drive the right course of action to reach and win the market and deliver the best results. The problem is, we’ve put too many tools in place to collect and analyze marketing data that are too hard to use and it’s causing a lot of frustration. Marketing professionals are spending way too much time searching for information and clicking through multiple pages in applications to gather the insights they need to design, execute, and measure effective campaigns.
To tackle this problem, marketing organizations need to tap into intelligent technologies like machine learning that can make data-driven marketing smarter and easier to execute. Machines can recognize patterns and analyze things with greater speed and efficiency and automatically deliver insights and intelligence that humans can use to make more informed decisions and engage customers and prospects in the most optimal way. And beyond tools that automate tasks and make marketing more efficient, we need to equip our teams with solutions that enable them to push the envelope. Like using artificial intelligence and machine learning to see data in new and innovative ways. Or leveraging augmented reality to create entirely new worlds where we can interact with customers in insanely personal ways.
As marketing tools and automated solutions continue to flood the market, the biggest challenge marketers will face is applying data to create timely, emotionally-reciprocal experiences. More and more consumers desire a human to human connection and want to communicate with an empathic human rather than a bot or an algorithm. The year ahead will be a pivotal milestone for marketers and brands, the ones that use their data to better understand consumer behavior and leverage it to create more personalized, human connections will succeed, while the ones that do not will risk losing loyal customers.
In 2018, if you’re a marketing professional that listened to recommendations of marketing experts everywhere, you collected a lot of customer data and used it to formulate campaigns. The problem of moving marketing forward into 2019 is two-fold; security and recognizing changes in data before solid patterns are defined. Data security, obviously, will be most important for maintaining trust between marketing and their audience, whereas recognizing emerging patterns in the data deluge will mean the difference between cutting-edge marketing or just ‘catching up’ marketing in the new year.
Sid Bharath, growth marketing consultant for tech startups
The biggest challenge with marketing data is figuring out what signals to pay attention to and how to prioritize them. With an explosion of data, the bottleneck moves to how fast you can execute on what the data tells you, and unless you have unlimited resources, you need to prioritize them.
The biggest challenge with marketing data in the coming year will be gaining actionable insight from a flood of data generated via a diverse and numerous set of online (and offline) channels, including social media, website, email, events, PR and advertising.
The Marketing Data Scientist will be focused on deriving detailed insight about customer behavior and producing reliable predictive and prescriptive insights based on complex data models and machine learning. These models will evolve from historical analysis into real-time applications that transform how products are delivered to customers.
Gennady Gomez, Director of Digital Marketing, Eightfive PR
As marketing data becomes not only more accessible but also much more bountiful, there will be an exponential increase in analysis paralysis. As a result, we’ll start to see the focus in martech shift from data mining to insights reporting, driven by data science and machine learning. These new breed of tools will be critical for marketers as they sort through, identify, and filter actionable data.
Until our ability to glean insights from all this data catches up to our ability to capture it, we’ll face the same issue as a city with plenty of cars and not enough roads: traffic. Don’t confuse having more data with having more insights.
The biggest data-related challenge will be consolidation and a full 360-degree view of the customer relationship. As it stands, data lives in different places — sales, customer service, digital marketing — and migrating it into a single platform and making sense of it all is going to be difficult. I hope that in a year’s time, we’ll see a lot of progress and proof that leveraging data to focus on delivering personalized, more relevant experiences is the optimum path for better engagement, stronger sales pipelines, and more meaningful marketing results.
Emerging technology has improved marketing strategy, but the challenge marketers are facing is the daunting task of managing a large number of applications. In the next year, more CMOs need to take a platform approach. Investing in a platform that can be integrated into an existing CRM allows organizations to easily unify their revenue teams, and with the addition of AI incorporated into the platform, unified teams have insight into the behavior of modern buyers with the use of real-time data.
Connecting the dots between marketing touch points and giving proper attribution has been and will remain a major challenge for marketers in 2019 … it can be hard to get a full 360-degree view of the true marketing and sales funnel … it is still nearly impossible to combine data from multiple touch points … to paint a full picture of marketing efforts and sales results.
Many marketers have so much data from multiple domains on hand, but no way to streamline and manage it in one centralized location to gain valuable insights.
Eric Quanstrom, CMO, Cience
The biggest issue with marketing data is federating it into a meaningful whole picture. As CMO, I live in (literally) a dozen different dashboards, daily. And that number is growing.
Daniel Jaye, Founder, Aqfer
2019 will be the year enterprises discover that serverless data lakes are a thing, and that they inevitably become data swamps due to widespread data proliferation, chaotic file partitioning/sharding practices, and the lack of traditional data management tools. As marketers are still floundering to piece together the data and figure out whether or not campaigns truly succeeded—they will realize that they can’t keep their heads in the sand on data any longer, and must work to get a better grasp on data management in order to get to the truth about their customers.
The biggest challenge will be how to tie all the data together to clearly identify what marketing channels are working or not working. There are multiple touch points to a buyer’s journey and it’s very common to see multiple marketing channels involved in the buyer’s decision, which makes it hard for analytics tools to attribute accurately what marketing channel contributed the most.
The biggest challenge in 2019 will be the ability to bring together … multiple sources of data to connect the dots, make informed decisions and act quickly on those insights.
The marketing technology landscape is increasingly fragmented and that’s not going to slow down. But marketers will need to find a solution to stop isolated data sources from negatively impacting their marketing capabilities in 2019. By integrating key marketing technologies such as CRM, marketing automation and ABM platforms, marketers can start to share data across these applications and get the complete customer view that they crave.
Businesses have a wealth of valuable marketing data available to them, but complications arise when this data remains in siloes pertaining to the different departments within that business. This prevents the data from being complementary, and businesses cannot detect potential weak links or new opportunities. With a holistic view of data, powered by marketing intelligence, campaign performance will drastically improve, and otherwise unidentified business opportunities will become unlocked.
Brian Czarny, CMO, Factual
In 2019, marketers will be faced with the challenge of data implementation. Marketers know how valuable data is, but struggle to make sense of it as they’re faced with the challenge of navigating numerous fragmented platforms and systems to get accurate and quality data. The goal is to gather data from multiple sources that work together to achieve optimum success, but there isn’t one standard way to streamline data. Eventually, unlocking this will give marketers the capability to improve context, relevance, and develop creative that resonates.
The key is to centralize … data and connect every interaction to a single customer ID. Then you can actually understand how your customer behaves across channels and devices. But even more importantly, that data has to connect to your marketing execution platforms directly, so you actually use those insights to power your marketing.
The increase and abundance of data that is available now due to integrated marketing platforms will demonstrate the flaws in various deep learning models. Deep learning models have been shown to be vulnerable to imperceptible perturbations in data, that dupe models into making wrong predictions or classifications. With the growing reliance on large datasets, AI systems will need to guard against such attacks on data, and the savviest advertisers will increasingly look into adversarial ML techniques to train models to be robust against such attacks.
Marketers are equipped with more consumer data than ever before, which can give them valuable insights into their customer base. Many are eager to use AI to automate and personalize communications, but lack the proper infrastructure and data know-how for AI to work properly. There are countless marketing AI platforms available, but until brands are able to properly segment their datasets and make their data truly work for them, they won’t have the ability to conduct innately intelligent marketing. What does this kind of marketing look like? All the marketer needs to do is set the framework, and the AI takes it from there to create personalized messaging for consumers. In 2019, we can expect to see a push from brands to organize their data within a framework that allows them to hyper-personalize communications.
Analytics continue to be a critical way to review the impact of marketing on business objectives. In 2019 the continued adoption of dashboard reporting and analysis systems will improve how we measure marketing programs and tactics. With better understanding of all marketing functions, organizations can take a more strategic approach and focus on what’s performing best. We also believe marketing and customer engagement will be an excellent first use-case for enterprise AI. AI systems can analyse structured and unstructured data to identify opportunities for marketing outreach, customer support, and other actions that enhance overall customer experience.
We are collecting and analyzing real time data using AI powered platforms … we are telling the story, giving shape and voice to the billions of data points that would otherwise be a bottomless inkwell of unrealized potential.
Marketing data and insights [will become] cheaper due to marketing AI platforms. It will no longer take big budgets, multiple data vendors, and a team of analysts to get actionable insights.
Marketers need: Data security, consumer trust, privacy
Lloyd Adams, SVP, SAP North America
Data ethics will become more important than data analytics.
In 2019, marketers will struggle with the social contract of data exchange between consumers and brands. They’ll wrestle with these questions: How will users be compensated beyond personalization? How can marketers do this without being “creepy”? And, as more biometric data emerges, how can marketers use in an ethical manner?
Making better and more critical decisions about what to collect, how to protect it, how to combine it, and how to use it. The idea of a “360-degree view of the customer” has encouraged brands to collect as much data as possible, but this should never be a goal because it raises risks such as privacy concerns, data breaches, GDPR compliance, and customer distrust. Marketers and customer experience leaders need to focus on prioritizing their data needs, better assessing risks, and developing a data strategy that prioritizes and centers on what data is essential and how it will be used over accumulating more of it.
The single biggest challenge B2B marketers face in the coming year will be balancing privacy and personalization to regain the trust of their audiences. In many ways, 2018 was a tumultuous year for brands, marketers, and customer experience leaders. Concerns around fake news, fake followers, and data privacy led individuals to question their trust in politicians, media outlets, social networks, and businesses alike. Those same concerns extended to how brands — both B2C and B2B — forge relationships with customers, and the data they use to do so. The priority for B2B marketers in 2019 must be to reassure customers – and their customers’ customers – their data is safe and secure. This has to be achieved in the changing climate of customer expectations. Increasingly customers — be they businesses or individuals — are expecting content that is important, interesting and timely to them. That starts with respecting their privacy, being open and transparent about when and why data is collected, and then leveraging the data that customers are willing to share to create personalized 1:1 experiences that deliver unique value. Going forward, brands must focus less on maximizing reach, and more on generating transparent, quality engagements that add value to their customers.
The biggest challenge marketers will face in 2019 is activating their first party data and growing its use. Brands are facing a privacy paradox as customers increasingly expect personalized service, but both data regulations and consumer privacy controls whittle away at third-party data and tracking. To remain compliant and provide great customer experiences, brands will increasingly rely on data customers willingly provide in the course of direct interactions across engagement channels–websites, apps, messages, social and in-store interactions and more. The best brands will go beyond gaining opt-ins, subscriptions and followers, and use these interactions to collect context and content preferences for each individual.
Tifenn Dano Kwan, CMO, SAP Ariba
A greater focus needs to be on being data compliant as well as on the ease of leveraging data.
The biggest challenge with marketing data stems from the fact that marketers have so much data available to them today, which they’re able to use to reach customers in a very precise way, yet there’s currently a huge lack of communication between brands and customers as to how and why that data is being used. Without more transparency with customers around why personalized outreach is happening and how it’s benefiting them, the immediate reaction is one of distrust, uncertainty, and even fear or anger. In 2019, brands need to focus on clear messaging that explains to customers why they’re using personalization tactics, how their privacy is protected, and what they stand to gain from it.
With ad-blockers becoming ubiquitous and privacy concerns reaching a fever pitch, marketers need to rebuild the trust that has been lost with consumers. Some of the most successful marketing campaigns in recent times have been honest and authentic, sometimes to the point of distress: KFC’s apology for running out of chicken and Nike’s partnership with controversial quarterback Colin Kaepernick. The biggest challenge in 2019 and beyond will be to create a new normal between marketers and consumers.
May 25, 2019 will mark 1 year since GDPR came into force in the EU. This privacy law sets concrete standards on how EU citizen data has to be treated in addition to strict guidelines on consent. Successful senders will have taken stock of this law and enacted internal processes to ensure compliance with European law.
As data protection regulations like GDPR become increasingly prevalent in 2019, marketers will struggle to target customers with individually customized online advertising. The most successful marketers will be those who can deliver individualized experiences, without individual user data. Computer vision and other contextual analysis technologies will be necessary to anonymously align ads with the likeliest potential customers.
Esteban Contreras, Senior Director, Hootsuite
We have more data than ever before in history and that in itself is a challenge. One of the most important problems to overcome is how to effectively handle and leverage data – data engineering, data analysis and data science – with legitimate consumer empathy. We need to consider privacy by design. The ethical use of big and small data is ultimately about creating value (e.g. personalizing and contextualizing experiences) without misleading or dehumanizing anyone.
Marketers need: Less data (or at least, smarter data)
The biggest challenge with marketing data will be having too much of it, specifically for B2C organizations.
Marketers crave access to information, but we’re drowning in a virtual quagmire of data. We want to get granular, digging deeper into every data point, but we’re stuck with analysis paralysis — unable to prioritize actions that will have the greatest impact on our KPIs.
Marketers will need a cutting-edge customer experience platform that recommend strategic actions tied to specific KPIs.
Noise. There is a lot of it and it is increasing exponentially. As data volumes increase, the tools we will need to filter it all and extract the valuable components will have to increase in their abilities accordingly.
The sheer volume of data being collected is itself posing a challenge to marketers. As we move into 2019, the ability to separate out meaningful patterns and relevant trends from the vast quantities of background noise will be a real test for many marketing teams.
Because there’s just more and more data with disparate KPIs coming from different sources, the biggest challenge will be in deciding what to ignore and what to value.
I predict a big challenge for marketers in 2019 will be remembering to use data as only part of the story and keep an eye on the bigger picture. How many times have any of us been hit with poor targeting because of something else we’ve liked or clicked on? It will be important for marketers to remember that the people they are marketing to are real, complex people who should be considered as more than just a few data points.
The biggest challenge for marketers is focusing on the data that matters. There is a lot of noise out there and each team member needs to know which data is significant to their success. But, it’s not just about having the data but putting it into context to make strategic revenue-driving business decisions. If you aren’t able to execute on your marketing data, it is useless.
I think the core problem moving into 2019 where marketing data is concerned is that with ever increasing amounts of data, marketers are still looking at each marketing silo individually or in small affiliated clusters as opposed to cross-analyzing insights across the entire marketing, sales and customer service value-chain. Often silo based marketing teams will look individually at social data, demand-gen, PR, sales and customer service rather than studying where they intersect and interact. This has always been faulty and is continuing to be a greater problem as each and every silo continues to yield more data.
Jo Ann Sanders, VP of Product Marketing, Optimizely
The biggest challenge marketers will face in 2019 is getting access to the right data to know definitively what their digital users want. Marketers are drowning in data from various analytics systems that provide a historical view of the past. They then ideate ways to improve based on this past data, spend resources to deploy updates, and then re-measure to see if their ideas worked. This process of guessing at what will improve conversion metrics can take weeks or months.
What marketers are going to have to do going forward to succeed so they can keep pace with rapid innovation is to go beyond analytics data that tells them where they have been and adopt new, agile test and learn practices. This will take the guesswork out of what users want, and better ensure that they are rolling out winning user experiences quickly.
With the proliferation of marketing data the challenge will be how to use the latest tools to narrow 1000 points of data down into the few key quality ones that are needed to improve business operations and to better communicate with your customer.
David Meiselman, CMO, exCater
In many ways, marketing has too much data on its hands. The challenge is to figure out which data helps the most to optimize targeting, messaging, and conversions. Traditionally, marketers would work their way through countless A-B tests to determine what works. Today, machine learning and artificial intelligence is helping us accelerate that process to detect correlations and causal factors to improve marketing outcomes across the board, from which people to target for highest conversion to what message to send when for greatest effect.
Tim Minahan, CMO, Citrix
The problem is, we’ve put too many tools in place to collect and analyze marketing data that are too hard to use and it’s causing a lot of frustration. Marketing professionals are spending way too much time searching for information and clicking through multiple pages in applications to gather the insights they need to design, execute, and measure effective campaigns.
For many years we’ve been trying accumulate as much data as possible, and we’re now ironically in a difficult position of potentially having too much data. We need to remember to sit back and discover what is actually happening at different points in time, so that we can figure out how it all fits together. We also can never forget that the most important data point when building attribution is still “Where did you hear about us?”
Marketers need: Many more things
These quotes don’t fit an exact category. But they’re too good to not use.
Jenni Schaub, Strategic Planning Director, DEG Digital
But always still remember, a mountain of data is not a replacement for empathy
Stephanie Smith, Co-Founder and President, MOJO PSG
Thoughtfully exploring and formulating the question you’re actually trying to use data to answer is a key challenge that marketers must face in 2019. Without taking the time to define the problem we’re solving for, we end up wasting a lot of time swimming in seas of data and even potentially misusing the data we uncover. Marketers must also find a balance between using data to inform, rather than dictate, decisions, as the marketing craft will always be a blend of art and science.
Measurement is hard. For SMBs and mid-market companies, it is harder, and the stakes are often higher. Although last click measurement is an archaic way to measure performance and impact, many marketers still rely on it because they don’t see accessible alternatives, as sophisticated tools are often difficult to set up or not flexible enough to work within their data model.. Yet, marketers are hungry for change and searching for a better way to provide visibility and optimize their campaigns. I see 2019 as the year modern marketers stop relying on vanity metrics and outdated measurement models and start looking at what is actually driving sales. Further, marketers need to embrace multi-channel adoption and prioritize creating connected stories across all touchpoints.
The biggest challenge for marketers will be navigating the evolution of what marketing is in a post-truth world. In 2019, marketers will need to ask themselves the difference between truth and propaganda.
Summary
Thanks to all the marketers who participated in this research, which was initially for my column in Inc, but grew beyond that.
Clearly, there’s a significant change in marketing data policy coming. Marketers know that it’s not about quantity of data but quality. They also know that insights on next best actions is the thing they need their data to reveal. And they are more than cognizant now that consumer privacy matters, and companies that violate their customers’ and prospects’ trust do get punished, both financially and in reputation.
We have to give the last word to Jolene Rheaulot from The Bid Lab.
She said this, which every marketer should remember:
The biggest challenge with the breadth of marketing data available to a company is to keep the data human.
. . .
. . .
Looking for a platform to unify your marketing data and derive smart insights on how to grow?
Eric Benjamin Seufert is the owner of Mobile Dev Memo, a popular mobile advertising trade blog. He also runs Platform and Publishing efforts at N3TWORK, a mobile gaming company based in San Francisco, and published Freemium Economics, a book about the freemium business model. You can follow Eric on Twitter.
Note: if you’re looking for ad monetization with perhaps less effort than Eric’s method below, talk to your Singular customer service representative (and stay tuned for additional announcements).
Various macro market forces have aligned over the past two years to create the commercial opportunity for app developers to generate significant revenue from in-app advertising. New genres like hypercasual games and even legacy gaming genres and non-gaming genres have created large businesses out of serving rich media video and playable ads to their users by building deep, sophisticated monetization loops that enrich the user experience and produce far less usability friction than some in-app purchases.
But unfortunately, while talented, analytical product designers are able to increase ad revenues with in-game data by deconstructing player behavior and optimizing the placement of ads, user acquisition managers have less data at their disposal in optimizing the acquisition funnel for this type of monetization. Building an acquisition pipeline around in-app ads monetization is challenging because many of the inputs needed to create an LTV model for in-app ads are unavailable or obfuscated. This is evidenced in the fact that a Google search for “mobile app LTV model” yields hundreds of results across a broad range of statistical rigor, but a search for “mobile app ads LTV model” yields almost nothing helpful.
Why is mobile ads LTV so difficult to calculate?
For one, the immediate revenue impact of an ad click within an app isn’t knowable on the part of the developer and is largely outside of their control. Developers get eCPM data from their ad network partners on a monthly basis when they are paid by them, but they can’t really know what any given click is worth because of the way eCPMs are derived (ad networks usually get paid for app installs, not for impressions, so eCPM is a synthetic metric).
Secondly, app developers can’t track ad clicks within their apps, only impressions. So while a developer might understand which users see the most ads in their app and can aggregate that data into average ad views per day (potentially split by source), since most ad revenue is driven by the subsequent installs that happen after a user clicks on an ad, ad view counts alone don’t help to contribute to an understanding of ads LTV.
Thirdly, for most developers, to borrow conceptually from IAP monetization, there are multiple “stores” from which ad viewing (and hopefully, clicking) users can “purchase” from: each of the networks that an app developer is running ads from, versus the single App Store or Google Play Store from which the developer gathers information. So not only is it more onerous to consolidate revenue data for ads, it also further muddies the monetization waters because even if CPMs for various networks can be cast forward to impute revenue, there’s no certainty around what the impression makeup will look like in an app in a given country on a go-forward basis (in other words: just because Network X served 50% of my ads in the US this month, I have no idea if it will serve 50% of my ads in the US next month).
For digging into problems that contain multiple unknown, variable inputs, I often start from the standpoint of: If I knew everything, how would I solve this? For building an ads LTV model, a very broad, conceptual calculation might look like:
What this means is: for a given user who was acquired via Channel A, is using Platform B, and lives in Geography C, the lifetime ad revenue they are expected to generate is the sum of the Monthly Ad Views we estimate for users of that profile (eg. Channel A, Platform B, Geography C) times the monthly blended CPM of ad impressions served to users of that profile.
In this equation, using user attribution data of the form that Singular provides alongside internal behavioral data, we can come up with Lifetime Ad Views broken down by acquisition channel, platform, and geography pretty easily: this is more or less a simple dimensionalized cumulative ad views curve over time that’d be derived in the same way as a cumulative IAP revenue curve.
But the Blended CPM component of this equation is very messy. This is because:
Ad networks don’t communicate CPMs by user, only at the geo level; [Editorial note: there is some significant change happening here; we will keep you posted on new developments.]
Most developers run many networks in their mediation mix, and that mix changes month-over-month;
Impression, click, and video completion counts can be calculated at the user level via mediation services like Tapdaq and ironSource, but as of now those counts don’t come with revenue data.
Note that in the medium-term future, many of the above issues with data availability and transparency will be ameliorated by in-app header bidding (for a good read on that topic, see this article by Dom Bracher of Tapdaq). In the meantime, there are some steps we can take to back into reasonable estimates of blended CPMs for the level of granularity that our attribution data gives us and which is valuable for the purposes of user acquisition (read: provides an LTV that can be bid against on user acquisition channels).
But until that manifests, user acquisition managers are left with some gaps in the data they can use to construct ads LTV estimates. The first glaring gap is the network composition of the impression pool: assuming a diverse mediation pool, there’s no way to know which networks will be filling what percentage of overall impressions in the next month. And the second is the CPMs that will be achieved across those networks on a forward-looking basis, since that’s almost entirely dependent on whether users install apps from the ads they view.
The only way to get around these two gaps is to lean on historical data as a hint at what the future will look like (which violates a key rule of value investing but is nonetheless helpful in forming a view of what’s to come). In this case, we want to look at past CPM performance and past network impression composition for guidance on what to expect on any given future month.
Estimating mobile ads LTV in Python
To showcase how to do that, we can build a simple script in python, starting with the generation of some random sample data. This data considers an app that is only serving ads to users from Facebook, Unity, and Vungle in the US, Canada, and UK:
[code]
import pandas as pd
import matplotlib
import numpy as np
from itertools import product
import random
geos = [ 'US', 'CA', 'UK' ]
platforms = [ 'iOS', 'Android' ]
networks = [ 'Facebook', 'Unity', 'Applovin' ]
def create_historical_ad_network_data( geos, networks ):
history = pd.DataFrame(list(product(geos, platforms, networks)),
columns=[ 'geo', 'platform', 'network' ])
for i in range( 1, 4 ):
history[ 'cpm-' + str( i ) ] = np.random.randint ( 1, 10, size=len( history ) )
history[ 'imp-' + str( i ) ] = np.random.randint( 100, 1000, size=len( history ) )
history[ 'imp-share-' + str( i ) ] = history[ 'imp-' + str( i ) ] / history[ 'imp-' + str( i ) ].sum()
return history
history = create_historical_data(geos, networks)
print(history)
[/code]
Running this code generates a Pandas DataFrame that looks something like this (your numbers will vary as they’re randomly generated):
One thing to consider at this point is that we have to assume, on a month-to-month basis, that any user in any given country will be exposed to the same network composition as any other user on the same platform (that is, the ratio of Applovin ads being served to users in the US on iOS is the same for all users of an app in a given month). This almost certainly isn’t strictly true, as, for any given impression, the type of device a user is on (eg. iPhone XS Max vs. iPhone 6) and other user-specific information will influence which network fills an impression. But in general, this assumption is probably safe enough to employ in the model.
Another thing to point out is that retention is captured in the Monthly Ad Views estimate that is tied to source channel. One common confusion in building an Ads LTV model is that there are ad networks involved in both sides of the funnel: the network a user is acquired from and the network a user monetizes with via ads served in the app. In the construction of our model, we capture “user quality” in the Monthly Ad Views component from Part A, which encompasses retention in the same way that a traditional IAP-based LTV curve does. So there’s no reason to include “user quality” in the Part B of the equation, since it’s already used to inform Part A.
Given this, the next step in approximating Part B is to get a historical share of each network, aggregated at the level of the Geo and Platform. Once we have this, we can generate a blended CPM value at the level of Geo and Platform to multiply against the formulation in Part A (again, since we assume all users see the same network blend of ads, we don’t have to further aggregate the network share by the user’s source network).
In the below code, the trailing three-month impressions are calculated as a share of the total at the level of Geo and Platform. Then, each network’s CPM is averaged over the trailing three months and the sumproduct is returned:
Running this snippet of code should output a DataFrame that looks something like this (again, the numbers will be different):
[code]
geo platform CPM
0 CA Android 5.406508
1 CA iOS 4.883667
2 UK Android 4.590680
3 UK iOS 5.265561
4 US Android 4.289083
5 US iOS 4.103224
[/code]
So now what do we have? We have a matrix of blended CPMs broken out at the level of Geo and Platform (eg. the CPM that Unity Ads provides for US, iOS users) — this is Part B from the equation above. The Part A from that equation — which is the average number of ad views in a given month that we expect from users that match various profile characteristics pertaining to their source channel, geography, and platform — would have been taken from internal attribution data mixed with internal app data, but we can generate some random data to match what it might look like with this function:
Running the above snippet should output something like the following:
[code]
geo platform source_channel ad_views
0 US iOS Facebook 73
1 US iOS Unity 463
2 US iOS Applovin 52
3 US Android Facebook 60
4 US Android Unity 442
5 US Android Applovin 349
6 CA iOS Facebook 279
7 CA iOS Unity 478
8 CA iOS Applovin 77
9 CA Android Facebook 479
10 CA Android Unity 120
11 CA Android Applovin 417
12 UK iOS Facebook 243
13 UK iOS Unity 306
14 UK iOS Applovin 52
15 UK Android Facebook 243
16 UK Android Unity 106
17 UK Android Applovin 195
[/code]
We can now match the performance data from our user base (gleaned using attribution data) with our projected CPM data to get an estimate of ad revenue for the given month with this code:
Running the above snippet should output something like the following:
[code]
geo platform source_channel ad_views CPM month_1_ARPU
0 US iOS Facebook 73 5.832458 0.425769
1 US iOS Unity 463 5.832458 2.700428
2 US iOS Applovin 52 5.832458 0.303288
3 US Android Facebook 60 5.327445 0.319647
4 US Android Unity 442 5.327445 2.354731
5 US Android Applovin 349 5.327445 1.859278
6 CA iOS Facebook 279 6.547197 1.826668
7 CA iOS Unity 478 6.547197 3.129560
8 CA iOS Applovin 77 6.547197 0.504134
9 CA Android Facebook 479 4.108413 1.967930
10 CA Android Unity 120 4.108413 0.493010
11 CA Android Applovin 417 4.108413 1.713208
12 UK iOS Facebook 243 4.626163 1.124158
13 UK iOS Unity 306 4.626163 1.415606
14 UK iOS Applovin 52 4.626163 0.240560
15 UK Android Facebook 243 5.584462 1.357024
16 UK Android Unity 106 5.584462 0.591953
17 UK Android Applovin 195 5.584462 1.088970
[/code]
That last column — month_1_ARPU — is the amount of ad revenue you might expect from users in their first month, matched to their source channel, their geography, and their platform. In other words, it is their 30-day LTV.
Putting it all together
Hopefully this article has showcased the fact that, while it’s messy and somewhat convoluted, there does exist a reasonable approach to estimating ads LTV using attribution and ads performance data. Taking this approach further, one might string together more months of ad view performance data to extend the limit of the Ads LTV estimate (to month two, three, four, etc.) and then use historical CPM fluctuations to get a more realistic estimate of where CPMs will be on any given point in the future (for example, using a historical blended average doesn’t make sense in the run-up to Christmas, when CPMs spike).
The opportunities and possibilities for making money via rich ads at this point of the mobile cycle are exciting, but they don’t come without new challenges. In general, with the way the mobile advertising ecosystem is progressing towards algorithm-driven and programmatic campaign management, user acquisition teams need to empower themselves with analytical creativity to find novel ways to scale their apps profitably.
