Category: Thought Leadership

The New Possibilities of an ID-Redacted World


Change has come to the world of digital marketing measurement. With the redaction of DoubleClick (now Google Marketing Platform) IDs, the rollout of Google’s Ads Data Hub (ADH), and growing interest in data clean room solutions, marketers must “unlearn” what was possible in the past and become rapidly familiar with what is now possible in the ID-redacted world.

The introduction of GDPR and the general global trend toward digital privacy has pushed us into a new era in the digital media world—but it doesn’t appear as though some have accepted that reality just yet. It’s usually true of technological disruption that we first try to retrofit old paradigms onto new technology and miss the forest for the trees. Only later do we find that there has been a complete paradigm shift and new use-cases have emerged.

We at MightyHive are finding that ADH and other emerging technologies are facilitating new and interesting ways of interacting with customer data that were not previously possible. As part of our own process of discovery and experimentation, we are learning that we need to fundamentally change the way we and our clients use data as a tool and a value generator. This learning process involves moving away from trying to measure what we no longer can and focusing more and more on how to build new utility into consumer ad experiences via connected data sets.

A quick primer on Ads Data Hub

First, a bit of background on what ADH does: ADH is a “data clean room” solution that allows multiple parties to analyse the intersections among respective first-party data sets, but without revealing granular row-level, user-level, or log-level data.

At MightyHive, we’re discovering what’s possible first-hand; working with clients to generate very new and very exciting insights from multilateral data co-ops.

As an example, Brand A could upload transaction records tied to first-party cookies. ADH lets Brand A query the intersection between its own records and ad exposures recorded by Google, but not a user level. Instead, the finest granularity at which ADH reports is about 50 individuals, which for most scaled campaigns is more than sufficient for generating insights, provided the ADH user knows how to structure the uploaded data and develop queries.

Why Ads Data Hub solves more problems than it causes

Focusing only on user-level ID redaction is myopic. Because ADH is designed as a “Switzerland for data,” it solves multiple problems, which in turn creates new possibilities for marketers, platforms, partners, and publishers:

  • User privacy is protected. What comes out of ADH are aggregated data intersections. User-level data goes in, but it does not leave.
  • Intellectual property is protected. First-party data is extremely valuable. This has historically made data owners averse to letting data leave company walls. ADH solves for these commercial roadblocks by providing aggregate insights without allowing any row-level data to leave the hands of its respective owner.
  • ADH accommodates a multilateral set of participants and data sets. ADH is not just a Google <> advertiser or Google <> agency solution. Multiple parties can load data into ADH and analyse intersections and patterns. This essentially allows for the construction of user journeys (e.g., brand campaign → offline visit → conversion → customer loyalty) by analyzing blended data sets from multiple parties who have interacted with customers.
  • ADH is cloud-based. Being a cloud-based solution based in large part on Google BigQuery, ADH is extremely flexible, accessible, and scalable from a technical standpoint. It’s an easy, industry-standard environment for engineering, analytics, and data science teams to plug into.

Ads Data Hub is not just a Google <> advertiser or Google <> agency solution.

How to navigate an ID-redacted future

ADH and other emerging data clean room solutions present the advertising ecosystem with new environments in which to ask questions of data, in many cases where data sets have never been brought together before. At MightyHive, we’re seeing what’s possible first-hand; working with clients to generate very new and very exciting insights.

For example, a Consumer Packaged Goods (CPG) company (ACME Dish Soap) can partner with one of their top retail channel partners (Shop-Now Supermarkets) and understanding how media spend (collected via buying platforms and ad servers) is affecting brick and mortar store sales. A data clean room can serve as a scalable and flexible environment for ACME Dish Soap to get unprecedented insight not from outdated Shop-Now Supermarkets regional sales reports, but from actual Shop-Now Supermarkets customer data and the patterns that emerge when compared with Google advertising data.

In the right hands, the value of being able to connect marketing activity to a holistic universe of data sets and business outcomes far surpasses the past value of terabytes of user-level log files. Marketers, agencies, and partners just need to learn how to use the new tools they have at their disposal and adapt campaign, media, and brand strategy to new insights and principles.

The industry needs to do a lot of catching up. Marketers, agencies, and partners are going to need to become more familiar with the technical particulars of data joins and sophisticated database queries that can’t always rely on user-level IDs as a “crutch.”

Like any data-driven solution, “garbage in, garbage out” (GIGO) very much applies here. In order to get the full value of Ads Data Hub or other data clean rooms, many brands will finally need to adopt some sort of first-party universal customer identifier that makes data useful in a data clean room environment.

Lastly, when a solution like ADH can tie media exposures, owned interactions, and sales revenue together to create a more customer-centric and holistic view, the value of popular proxy metrics (which were already questionable) like reach or CTR are bound to decline rapidly, forcing many a campaign to be completely re-evaluated.

Let’s talk

If you have questions on Ads Data Hub, ID redactions, GDPR and digital privacy, or next-generation measurement and analytics, please reach out to us at We’re excited about these new possibilities and we think you should be too.

Get the Slide Deck and Video that Explain Data Clean Rooms

Marketers need practical solutions to preserve measurement and insights in a privacy-first era. Get an overview of data clean rooms that are available now (e.g., Ads Data Hub) as well as what might be coming next.


Cookies Yield to New Privacy Rules 


For more on digital privacy, our free report “Digital Privacy in a Post-Cookie World” is an in-depth guide to new privacy changes and the steps marketers can take to adjust. Get the report.

A New Normal?

Digital marketers’ 25-year reliance on browser cookies is ending. It’s true. The world’s top three browsers have all announced privacy roadmaps that significantly restrict the use of cookies. Additionally, the General Data Protection Regulation (GDPR) established new privacy standards across Europe, and 11 U.S. states are currently adopting or proposing similar laws. Simply put, the cookie’s time has come.

The good news is that many solutions are available to digital marketers. Embracing them sooner rather than later will ease the transition away from cookies and foster much better ways to connect with customers.

Apple’s Line in the Sand

In 2017, Apple’s Safari browser launched Internet Tracking Prevention (ITP) to restrict third-party cookies from tracking users across multiple websites and domains.

Source: Twitter

Safari ITP will significantly limit marketers from:

  • Applying behavioral targeting tactics to Safari users or Safari-heavy segments (e.g., North American mobile users).
  • Identifying users over periods of time longer than a few days.
  • Segmenting customers and audiences based on observed behavior.

Source: StatCounter

Firefox and Chrome Follow Suit

While Safari automatically restricts cookies on behalf of its users, Google is taking a different approach by improving Chrome’s user controls. Google announced its intentions in May of 2019 but the specifics have not been shared publicly yet.

For Firefox users, Enhanced Tracking Protection will soon operate in the background to block thousands of companies from tracking users’ online activity.

Source: StatCounter

A Roadmap for Adaptation

Over the last 25 years, the Internet and consumer habits have changed — but cookies have not. Continuing to rely on cookies may be necessary in the near term, but over time, this will result in increasingly ineffective digital advertising, so marketers should follow a “now / near / long-term” approach for adapting to change.

  • NOW: Get up to speed — Adjust targeting and measurement to account for Safari ITP-related roadblocks and restrictions.
  • NEAR: Mitigate the effects — A range of proactive technical approaches to preserve digital targeting and measurement where possible.
  • LONG-TERM: Invest in new strategies — Make a strategic shift away from cookie-based ID regimes to logged-in users and value-add user experiences.

NOW: Get Up To Speed

The first step is knowledge — embrace and adjust to the new rules. Safari ITP will reduce the life of most marketing cookies to one day or less, in many cases blocking them altogether. This means that campaigns may not distinguish between a new visitor and a repeat customer.

Marketers may not enjoy the disruption, but it is happening and needs to be properly understood and accounted for before moving forward with more proactive solutions.

NEAR: Mitigate the Effects

There is no perfect solution that restores all pre ITP- and GDPR-era capabilities. Marketers should evaluate their individual needs based on the impact to site-side measurement, targeting, campaign performance, geographical exposure… the list goes on.

A variety of browser-side and server-side approaches are available marketers. When coupled with adjustments to site-side analytics, ad serving, and DSPs, marketers can mitigate impact and preserve insights. A deeper dive is available in our free report: “Digital Privacy in a Post-Cookie World.”

LONG-TERM: Invest in New Strategies

To account for a greater focus on digital privacy, marketers will have to fundamentally rethink many digital marketing strategies and start investing differently. But change is good and marketers have many avenues to explore.

The list of possible adaptive strategies is long and includes a focus on logged-in user experiences, leveraging the scaled audiences and measurement insights of “The Triopoly” (Google, Amazon, Facebook), second-party data partnerships, programmatic direct buys, smarter contextual targeting, privacy-safe measurement via “data clean rooms,” and doubling down on brand and creative.

The current trend toward privacy will undoubtedly have far-reaching consequences and no one can be certain where this will lead. You can expect more browser changes and new regulations, but this also means more adaptive solutions and new targeting tools. The agile marketer will adapt to change and take advantage of new opportunities.

For more on digital privacy, our free report “Digital Privacy in a Post-Cookie World” is an in-depth guide to new privacy changes and the steps marketers can take to adjust. Get the report.

Video: Belinda Smith of Electronic Arts Shares Her History of In-Housing, Recruiting Talent, and Instituting Change


Recently, MightyHive hosted its first In-House Alliance Dinner in San Francisco where the conversations centered around recruiting and building in-house programmatic teams. MightyHive President Americas, Emily Del Greco asked Electronic Arts’ Global Head of Media, Belinda Smith to share her experience as a change agent building teams at both AT&T and EA. Also in attendance at the dinner were senior marketers from Merck, Visa, Sprint, Electrolux, IHG, and others.

Belinda Smith, left, Global Head of Media, EA and Emily Del Greco, President of the Americas, MightyHive, speak at the inaugural MightyHive In-House Alliance Dinner in San Francisco.

“I feel like I have seen advertising from all sides… and at each junction I was so frustrated by people not willing to admit the failures of the ecosystem.”

– Belinda Smith

Ultimately, Smith changed the culture of the companies she worked for. To do this, she moved teams halfway across the county, launched robust talent recruitment programs, and successfully campaigned for millions of budgetary dollars. But perhaps the biggest challenge was finding ways to blend agency culture with the deeply-rooted brand values.

“If I think about what I want to accomplish in my career, what’s going to be worth it to me to not be at home playing with my kid…then I really want to change the industry.”

– Belinda Smith

During the interview, Smith confessed she missed the variety an agency affords, but showed no signs of wanting to return. At EA, Smith recognizes that she is in a place where she can make the most impact.

Smith and Del Greco spoke about building in-house teams in front of an audience at the MightyHive In-House Alliance Dinner.

“When you are at an agency… your client being happy is what is sustaining you. That creates an inherent conflict of interest when you have bad news to tell them…so for me, I am fulfilled by the fact that I am in a position to have the mandate to be audacious and to challenge what is going on and to think about how we do things differently.”

– Belinda Smith

The Fusion of Creative, Media, and Technology

Watch “The Fusion of Creative, Media, and Technology” (page opens in a new window)

At AdExchanger’s Programmatic I/O in San Francisco, MightyHive CEO Pete Kim took the stage to discuss the need for advertisers to adopt a unitary approach to their creative, media, and tech. Advertisers who fail to do so risk missing crucial opportunities to hold meaningful conversations with today’s ultra-savvy consumer.

Got Creative?

Jeff Goodby, Co-Chairman and Partner, Goodby Silverstein & Partners, implied in a 2018 AdWeek Op-Ed that technology debases creativity in advertising. Goodby suggests that using data and tech to inform programmatic media buys amounts to nothing more than “targeting and tonnage,” eliminating good creative altogether. He asserts that this model produces an expensive and ineffective structure for advertisers, and an unpleasant experience for consumers.

In his Programmatic I/O talk, however, Pete paints a very different picture of how data and tech actually work together to elevate creative. In fact, Pete says, not only does this model yield more relevant, dynamic creative, but it does so at scale and in a constant optimization loop, so it always gets better. 

creative big idea

The Big Idea

While there has been an undeniable sea change in the industry leading to major digital disruption, one thing hasn’t changed: great advertising still requires “the big idea.” But the machinations behind producing and disseminating creative to consumers continues to evolve.

“We believe that technology doesn’t debase creativity; if used correctly, it elevates it. We are all really different people… that’s why personalization is necessary.”

– Pete Kim, CEO, MightyHive

In the past, technological limitations required advertisers to generate one message, and “broadly cast” it to all consumers. Now, it’s possible to personalize creative and target consumers based on their demographics, preferences, and other criteria. In addition, we’re able to create, test, and iterate upon these creatives in near real-time.

Personalization is the New Table Stakes

For the first time in history, consumers can watch, read, and listen to anything they want at any given time. This new on-demand, cross-screen culture has raised the bar for advertisers to meet the quality of the entertainment their content is interrupting. Unfortunately for advertisers, this expectation for perfection and relevance across all media means that consumer sentiment around advertising has never been lower–and a one size fits all approach is unacceptable.

creative many to many

The Battle of the Good Idea

On creative teams past, there was a “gladiatorial” fight to determine which big idea actually got produced and saw the light of day. That one idea then became the basis for a static, inflexible campaign lasting weeks, or even months (in some cases years!). Now, Pete says, marketers are no longer “locked into” one creative (e.g., one commercial, one print ad, one radio ad) that gets produced and is written in stone until the campaign’s end. Granular targeting and creative optimization capabilities give advertisers the iterative flexibility to deliver the right message to the right consumer, on an ongoing basis.

What’s Next?

Because this creative, data, and technology loop is relatively new, we are just at the beginning of recognizing its true power. According to Pete, the next change is a mindset shift to catch up to our newfound technological capabilities. In addition to learning how best to leverage dynamic creative optimization and programmatic tools, we need to adapt our mindsets to suit the new ways consumers interact with media. creative mindset

Better, Faster, Cheaper

In order for advertisers to get their story across in a way that resonates, it’s crucial to move away from giant, costly campaigns once or twice a year to a constant conversation model. Dynamic creative and programmatic technology allows advertisers to have not just one, but millions of simultaneous, personalized conversations in real time.

“If you are only updating your campaigns a few times a year, it’s like having a conversation with somebody who always says the wrong thing, and takes months before they respond to what you just said.”

– Pete Kim, CEO, MightyHive

Watch Pete Kim’s full “The Fusion of Creative, Media, and Technology” session for further insights about unleashing creativity at scale.  

Get Ready for the Data-Confident Marketer


As first-party data takes on “strategic asset” status, brands are engaged in high-stakes jockeying to decide who will take the First-Party Data High Ground. Will the outcome hinge on company size? Speed and agility? Technology? Agencies and solutions partners?

The jury is still out, but new research on first-party data from MightyHive sheds some light on those marketers that feel they’re pulling ahead of the pack. MightyHive partnered with Advertiser Perceptions, surveying 200 brand marketers and conducting five in-depth marketer interviews to attitudes, goals, and expectations around first-party data (timelines, ROI, ease of access, etc).

Interesting Patterns in the Data

When we crunched the numbers from the survey data, we saw some interesting lines get drawn between marketers. Not only were some marketers more confident in their timelines and expected ROI, but these “data-confident” marketers exhibited clearly different characteristics than their less-confident counterparts. Our research found that data-confident marketers:

  • Use a more diverse range of technologies to manage and activate first-party data
  • Are more likely to use partners such as agencies and consultancies to help unlock first-party data
  • Put a significantly higher priority on certain first-party data types such as ad serving data and mobile app analytics



What Do Data-Confident Marketers Stand to Gain?

Admittedly, our survey asked marketers to make a self-assessment, so one can’t assume that just because some marketers said they expected higher ROI from their first-party data that those marketers will actually achieve higher ROI. But…when you consider that, for example, this higher-ROI cohort also indicated significantly greater adoption across every one of the four different partner types we asked about (Agencies, Management Consultants, Systems Integrators, and Ad Tech Vendors), it gets more difficult to chalk higher ROI expectations up to marketer hubris.

Even our survey responses seemed to reinforce the stakes in advertising’s current era of upheaval. When asked what the primary reasons were for using first-party data, these were marketers’ top four answers:

  1. Improved performance / ROI
  2. Accuracy / data quality
  3. Lends to more precise targeting
  4. Improved measurement and attribution

Perhaps not coincidentally, those four answers can be rearranged into a loop or cycle:

It is urgently important to understand that that cycle is a positive feedback loop. Meaning that once marketers “crack the code,” the feedback loop will kick in and they will amass an ever-larger and richer trove of first-party data that they will do a better job of activating in campaigns and customer funnels that throw off yet more first-party data. We’ve already seen this happen with DTC brands that have, in a few very short years, sprung from out of nowhere and are giving incumbent brands a run for their money.

Download the Report and Stay Updated

Download “The Data-Confident Marketer” here.

Over the next several weeks, expect more analysis digging deeper into the report. We’ll also reveal additional data and insights that were left on the cutting-room floor. So when you register for the report, opt in for email updates, or follow MightyHive on LinkedIn and Twitter.

If you have questions on the report, please email us at We look forward to hearing from you.



MightyGuide: Tokyo


MightyHive recently announced the opening of our new office in Tokyo, which will serve to connect Japanese marketers with local MightyHive talent, supported by our network of digital experts across the globe.

Our team in Tokyo is looking forward to partnering with marketers in the Japanese market, helping to guide them as they take control of their digital futures through unified media and analytics. The team brings a wealth of programmatic knowledge to the APAC region, and we are very excited to welcome everyone in Tokyo to the MightyHive family.

We wanted to get to know both the team and Tokyo better, so we asked Toshihiko Honda, a Digital Analytics Specialist based in our Tokyo office, to answer a few questions about the best parts of living and working in Tokyo.

Check out the interactive map at the end of this post for more information on each of Toshihiko’s recommendations.

Q: What brought you to MightyHive?

The Tokyo office just launched in February 2019, and we are establishing MightyHive’s footprint in a new market. The Japanese digital market is huge and unique in many ways, making this an exciting opportunity to build a new branch from scratch.

“After meeting with different members of the MightyHive team, I was really impressed with MightyHive’s collaborative culture and that’s what really drew me to the company.”

Although we are based in Tokyo, we are collaborating with the entire MightyHive team globally, which allows us to tap into MightyHive’s collective expertise to support our Japanese clients.

We have a challenging task ahead of us, but working cross-functionally makes it a very exciting opportunity!

Q: What’s your favorite part of working in Tokyo?

Everyone here is very professional and projects usually proceed on time, as everyone is punctual.

The city itself is very clean and traffic is managed efficiently, so it’s easy to get around and you don’t need to worry about traffic delays.

Another bonus is the food. You can find really tasty food at a reasonable price. Good Japanese food and sake are everywhere in Tokyo!

Q: What are your favorite restaurants in Tokyo?

One of my favorite places to eat in Tokyo is in Shinbashi. It’s a business area, but it’s also famous for its many delicious Japanese style bars and restaurants (They are called Izakaya:居酒屋).

MightyGuide Tokyo- Restaurants

SOURCE: Uokin Facebook, Jidoriya Facebook, Coco Ichibanya Trip Advisor

My favorite restaurants in Shinbashi are:

  • Uokin: If you come to Japan, you can’t miss the seafood. You can find fresh seafood at a reasonable price at Uokin. They have some branches in Shinbashi, but you’ll have great seafood, no matter which branch you choose.
  • Jidoriya: Jidoriya is one of the best traditional yakitori (grilled chicken) places in Shinbashi. As the restaurant is pretty small, I would recommend you to go with a small group.  
  • Coco Ichibanya: Coco Ichibanya is a Japanse curry rice restaurant. Japanese curry is a little different than Indian curry, but it’s a Tokyo staple. Coco Ichibanya is a chain restaurant, so they have locations all over the city.

Q: Favorite cultural activities?

One of the best ways to experience the culture of Tokyo is to simply walk (or sometimes run) around the different neighborhoods. In fact, I usually spend the weekend walking around the city with my wife.

Since Tokyo is such a large, populated city, each neighborhood has its own unique personality. For instance, Otemachi is a very upscale business neighborhood around the Imperial Residence where you will find a nice view and a lot of runners. Shibuya is a very busy and crowded tourist area with shops, nightlife, and restaurants. Ginza is an upscale tourist area, and Harajuku is a popular cultural and shopping area for teens.

Just walking around the different areas in Tokyo will give you a rich cultural experience.

Q: What are some of your favorite off the beaten path activities?

Good question! Because there are so many people in Tokyo, it’s important to find a non-touristy location to spend your weekend.

My recommendation is Koganei park. Although it is located a little far from the central area, the park is not crowded and you can spend a slow afternoon enjoying the day. The park is also home to the Edo-Tokyo Open Air Architectural Museum where you can explore architecture from the Edo era.

MightyGuide Tokyo - Edo-Tokyo Museum

SOURCE: Edo-Tokyo Museum Facebook

Q: Where are the best places for a great cup of coffee (or tea)?

You should go to the Okushibu neighborhood (located near Shibuya) for the best cafes.

MightyGuide Tokyo - Fuglen Tokyo

SOURCE: Fuglen Tokyo Facebook

My favorite is Fuglen Tokyo. The interior of the cafe is from Norway, and the mood is very relaxing. There are great cafes all over Okushibu, so no matter which one you choose they all serve high quality drinks.

If seats are not available at the cafe, take your coffee to Yoyogi Park. It’s the biggest park in Tokyo and a good place to relax.

Q: Finally, what’s your favorite neighborhood in Tokyo?

While not technically in Tokyo, you should definitely visit a hot springs area while you are in Japan. It’s a big cultural activity here and is very relaxing, so one of my favorite places is the Izu Penninsula where you will find many of these hot springs.

One of my favorites is Izunagaoka Onsen. As it takes about 2 hours by train, I recommend making your visit a weekend trip.

We are excited to bring MightyHive’s services and solutions to Japan to offer full regional support, helping Japanese marketers take control of their programmatic futures and providing unified media and analytics solutions across regions. MightyHive is currently looking to fill a variety of roles in the Tokyo office. Visit our careers page to learn more and apply

Event: ANA In-House Agency Conference

The MightyHive team is thrilled to be attending the sold out ANA In-House Agency Conference. MightyHive CEO Pete Kim will be joined by Josh Palau, VP Media Strategy and Platforms at Bayer to deliver a pre-conference session on Paving the Path to In-Housing Success.

Other speakers from PwC, Nationwide, Verizon, Clorox, GlaxoSmithKline, Bank of America, and more will discuss topics like driving cost efficiencies, project prioritization, building a strong culture and fostering talent, and the future of in-house agencies.

Paving the Path to In-Housing Success

March 13, 2019 | 3:45 PM EST

The decision to go in-house should not be taken lightly, nor is a 100% in-house model right for every organization. Marketers who jump in without building a thorough business case for change are likely to struggle; a smooth transition requires planning, prioritization, and patience. In this session, Bayer and MightyHive will break down the organizational choices that, when examined preemptively and thoughtfully, will pave the path to in-housing success. We’ll consider how to decide what to bring in-house, cost calculations, managing expectations, partner selection, hybrid resourcing models, and more.

Add to Calendar: Google, iCal


Meet the Team

Interested in learning more about how you can take control of your digital future? Schedule a time to meet with the MightyHive team onsite by emailing We look forward to seeing you there!

See What Makes Confident Marketers Tick

If you’ll be onsite, be sure to visit the MightyHive table to get a cozy gift and a first look at our new report: The Data-Confident Marketer. We partnered with Advertiser Perceptions to survey 200 marketing decision-makers to find out what makes data-confident brands tick. Pick up our report to see how you measure up.

The Ritz-Carlton, Orlando Grande Lakes
4012 Central Florida Parkway
Orlando, FL 32837

Wednesday, March 13 – Friday, March 15, 2019

Machine Learning Basics that Marketers Should Know


Making Machine Learning Concepts Accessible

Cutting-edge advertising technology is often marketed as “powered by Artificial Intelligence.” Some marketers get scared away, believing that they aren’t prepared or educated enough to leverage the latest tech. However, many marketing and advertising applications of Artificial Intelligence (AI) lean heavily on a subset of AI called Machine Learning. Essentially, Machine Learning (ML) is about improving predictions from the data you already own. As a decision maker within your organization, you should understand what basic ML applications entail and how they can be properly employed.

This blog post will cover three basic ML applications:

  1. Regression
  2. Classification
  3. Clustering

We’ll explore some use cases and avoid grandiose concepts that are too complicated to execute when getting started. I believe that beginning any Machine Learning endeavor requires baby steps. Like all technology development, it’s an iterative process that requires due diligence, realistic expectations, and never-ending trips back to the drawing board. Most importantly, this information should provide marketers with a reasonable understanding of ML concepts so that you feel empowered to make educated decisions when investing in projects.

Making Predictions with Regression

Regression is the process of estimating the relationship between variables. For example, a basic Return on Ad Spend (ROAS) model: given $100 of ad spend (X), how much revenue can I expect to generate (Y)?

Another useful example is predicting morning traffic. How much does the length of my commute increase for every inch of rain, day of the week, or time of departure? Traffic and ROAS predictions for that matter depend on more than just one single input variable. These models can combine tens and possibly hundreds of variables. Predicting ROAS could be a function of budget, time of day, average frequency caps per user, site content ratings, ad visibility, ad size, button color, ad copy, etc. This is known as MULTIPLE REGRESSION.

A multiple regression model might tell you how inches of rain, day of the week, or time of departure affect commute time. (Photo credit: Wikipedia)

The beauty of multiple regression is that the model dictates the relative importance of each input in determining the output variable or prediction variable (e.g., ROAS or commute time). Simply stated, regression is best used when attempting to forecast a result.

You might hear analysts or data scientists talk about regression analysis to predict future performance. Regression uses historical data points to create a model that allows a program to plug values into the weighted equation and get back a predicted result.

Using the example above, let’s say we’ve determined that our ROAS metric is on-site revenue ($OSR). We’ve trained our regression model with some of the possible variables considered above. Let’s assume our model is:

$OSR = (1.03 * budget) – (0.05 * time of day) + (0.63 * avg. frequency caps per user) + (0.005 * site content rating)

In making a prediction for on-site revenue generated, you could plug any combination of values into this equation and get a predicted amount of on-site revenue ($OSR).

A simple example of linear regression. In practice, ML regression models are often significantly more complex. (Image credit: Wikipedia)

Any machine learning system should constantly re-train on new data, as results are always changing and the relative importance of each input variable can change significantly based on actual results. The model should adapt as your data changes. Your predictions are only as good as the data you feed the system.


Classification is a form of categorization via pattern recognition. At its simplest form, logistic regression is the model of choice for determining the likelihood that something is either Category A or Category B.

A common example might be an email spam classifier. A more relevant example for advertising would be a classifier that determines the probability of a user converting after seeing an ad. This type of classifier powers the automated bid optimizers found in many DSPs. In theory, such a classifier would take an array of inputs related to a user —often gleaned from detailed cookie data— and provide a probability or likelihood of conversion. The higher the likelihood of conversion, the more your DSP is willing to pay for the impression and thus a higher bid. However, not all classification has to happen within an auction.

If you’re curious to get a deeper dive into how supervised classification algorithms work, this video from Wolfram walks through several examples:

On-site experiences can be significantly improved by employing classification. Understanding who is visiting your site by age, gender, location, estimated income level, interests, and so forth can help you tailor on-site content such as text, images, calls to action, recommended products, and much more. No classification system is going to be 100% accurate, but even a slight improvement in UX that can be driven by a classification system can move the needle in terms of revenue.

Enterprise analytics platforms (such as Google Analytics 360 or Adobe Analytics) often offer demographic reporting that marketers find useful. These reports are often driven by a combination of classification and clustering algorithms employed together.

What’s the Difference Between “Supervised” and “Unsupervised” Machine Learning?

In the case of classification, “supervised” ML means that the algorithm is provided with preset classification labels or categories. The classification algorithm can then work through a data set and classify items according to the available labels and categories.

On the other hand, “unsupervised” ML is required when there is a data set, but the categories, patterns, or meaningful relationships that might exist in that data set are unknown before the algorithm (e.g., clustering algorithm) has run.


Classification and regression are typically SUPERVISED learning models, meaning the input data is labeled for a particular set of outputs. Clustering algorithms are UNSUPERVISED—the process of finding patterns and hidden structure within unlabeled data. Think of unsupervised models this way: “What patterns can I find in a set of data that I don’t already know much about?”

Audience or customer segmentation are prime examples of applied clustering. Your Customer Relationship Management (CRM) system is filled with a breadth of customer-related information. Wouldn’t it be valuable to figure out who you should reach out to for a customer loyalty promotion? Or, what kind of client is likely to churn and what can you do to retain their business?

Clustering algorithms can identify the similarities between your users/customers and group them accordingly. K-means clustering is a simple but powerful algorithm that can be implemented with minimal inertia. While k-means clustering is often a stepping stone to more sophisticated clustering algorithms, it can whet the appetite for what can be further uncovered.

This animation shows a hypothetical K-means algorithm working iteratively to identify clusters of data points. (Animation credit: Wikimedia Commons)

Distribution models employ a more statistical approach to clustering, where membership is probabilistic and members can belong to more than one cluster. Distributive clustering is actually quite useful in audience segmentation. Your users may exhibit behavior that is commonly found in high-revenue customers but may also exhibit behavior in customers that are likely to leave for a competitor.

Machine Learning Doesn’t Just Work Right Out of the Box

No machine learning system is complete without cross-validation, feature reduction (removing variables that don’t impact the model), and other techniques that improve the accuracy of predictions. Ensemble machine learning systems (that blend multiple ML approaches) are increasingly more complex and sophisticated and often require the help of experts to develop and maintain. It’s important to understand and start with the basics and decide if machine learning is worth the investment for your organization.

Taking the Next Step with Machine Learning

You should now have a better grasp of common Machine Learning algorithms and their everyday applications in marketing and advertising. Getting started may seem daunting, and with the volume and depth of data at your disposal, hiring an expert to get you off the ground can pay major dividends to your business now and in the future. Exciting breakthroughs by ML practitioners and the availability of open sourced, enterprise-quality SDKs (such as and StatsModels and scikit-learn), along with scalable cloud infrastructure at your disposal, has made it easier than ever to discover what insights your data may hold.

How Salesforce Journey Builder is Set to Power the Next Generation of Programmatic Advertising


Salesforce and Google, the tech giants joining forces to solve the traditional gap between offline and online customer journeys could have an unexpected impact on the digital advertising industry.

Uniting the Online and Offline Customer Experience

First announced at the end of 2017, the Salesforce and Google global partnership has come to fruition with an enhanced integration between Google Marketing Platform and standard connectors for Google Analytics 360. One of its greatest benefits is making online behavior and insight data available across Salesforce Sales Cloud and Marketing Cloud, enriching traditional CRM customer data (offline) with live (online) metrics.

Simultaneously, Salesforce has been advancing Journey Builder, the customer engagement tool that is central to its Marketing Cloud package. The result is a powerful, robust tool capable of delivering responsive and highly personalized data-driven automated campaigns.

Composing a Symphony of Data

Think of Journey Builder as an orchestra conductor making sure the musical ensemble is playing perfectly and following the partition. Previously, the conductor was only hearing from one ear (offline data) and adding Google Analytics 360 to the mix let the conductor hear with both ears (online). While performing well before, suddenly the musical ensemble is able to play a much larger repertoire.

In this sense, the integration between Google Analytics 360 and Salesforce gives marketers a more complete view of their customer data, with Journey Builder enabling a path for improved engagement. 

As Google and Salesforce describe it, it’s a match made in marketing heaven. This match has also created additional benefits for marketers looking to upgrade their technology but wanting to balance their needs against those of their organization. This results in alignment between the goals of the CMO and CIO. The CMO benefits from a powerful actionable tool mapped on the entire B2C customer journey and provides the CIO a cost-efficient alternative to custom integration and implementation.

Salesforce and Google Marketing Platform Today

Marketers today have the ability to view Google Analytics data within Marketing Cloud, associating those site usage goals and e-commerce data points to email metrics.

In the latest release, Marketing Cloud includes improvements to its Google Analytics 360 connector allowing marketers to capture online behavior by linking to Google Analytics 360 audience activation. Using unique IDs shared between the two solutions, it allows marketers “to build audiences based on Google Analytics data for web and mobile app activity, email engagement, and offline behavior.”

Capturing online customer behavior and activities allows marketers to trigger automated, personalized journeys through the Marketing Cloud engagement channels via Journey Builder.

Salesforce Journey Builder Engagement Channels Through Marketing Cloud and Google Marketing Platform

While the latest release is packed with new features, it’s hard not to see the exciting future ahead of us as the product roadmap is expected to include deeper links with the Google Marketing Platform stack including Google Optimize 360, Surveys 360, and Data Studio.

Looking Forward: The Impact on Programmatic Advertising

Journey Builder always seems to have been destined to enter the digital advertising arena. With the goal of fulfilling a marketer’s dream of enabling 1:1 marketing, Journey Builder is set to become a valuable tool for helping advertisers streamline and personalize their campaigns in real-time for a better customer experience using online data collected from Google Marketing Platform.

Different People Receiving Personalized Journeys on their Devices

Allowing Journey Builder to connect with the other platforms in the Google Marketing Platform stack such as Campaign Manager or Display & Video 360 would open the gates for even smarter and more efficient programmatic ad buying.

The integrated ecosystem will ultimately encourage and help facilitate brands who are thinking of bringing their digital advertising in-house. With the hyper-targeted audiences activated via Marketing Cloud, marketers will also be able to drive overall acquisition costs down, making the business case for in-housing their digital advertising more compelling.

MightyHive is a leading partner for brands looking to integrate Google Marketing Platform and Salesforce. With a focus on uniting media and analytics, we help our clients own their analytics and insights to inform a data-driven campaign strategy.

To learn more about how MightyHive can help you activate a unified data and media strategy with Google Marketing Platform and Salesforce, please contact us.

Attribution Solutions: Where to Start When Digital Metrics Fall Short


“It’s important for us to help our clients build the framework to identify how their KPIs tie back to their marketing and business objectives.”

– Cullen Urbano, Enterprise Consulting Lead, MightyHive

Attribution is an integral component of any successful marketing plan. But where should marketers start with attribution modeling, and what happens when digital metrics fall short? On January 17, practitioners from across the industry met at MightyHive NYC over pizza and drinks for a deep dive into the triumphs and challenges of digital analytics, and where advertisers often go wrong.

Rachel Adams, MightyHive’s Director of Accounts, moderated a panel entitled Attribution Solutions: How to be a Hero when Attribution Models Fall Short. On the dais for the evening were Ben Rudolph and Emma Tessier, both MightyHive Project Leads, as well as Cullen Urbano, an Enterprise Consulting Lead at MightyHive. Ben, Emma, and Cullen drew on their experience with clients and their in depth industry knowledge to provide a well-rounded perspective about the best strategies for attribution modeling.

attribution panel

Left to Right: Rachel Adams, Director of Accounts; Cullen Urbano, Enterprise Consulting Lead; Emma Tessier, Project Lead; Ben Rudolph, Project Lead

We Are Gathered Here Today…

Attribution is a hot topic. The MightyHive office was packed with people who voluntarily passed up a night of binge watching Netflix to talk about it. But what makes attribution so challenging for advertisers in the first place?

Tech Companies Aren’t Great at Sharing

Our panelists all cited tech limitations and walled gardens as major factors that make it tricky to figure out how marketers should optimize their media spend. As Ben pointed out, “In the ecosystem a lot of players want to demonstrate their specific value, so they put up walls and are unwilling to share data that will give marketers a more complete picture.” He continued, “There’s also no north star for advertisers to look to that isn’t last-touch, so it can feel risky or intimidating to look at different models.”

Also, Robots Aren’t Perfect

Another often-overlooked factor when dealing with high-tech, automated solutions is human oversight. Cullen stressed the importance of evaluating every model with a human gut-check. “If you do that, you can avoid instances where an MTA model might recommend a mass reach partner with sub-20% viewability and +30% frequency.”

Rachel echoed: “Even though all the decisions we make are highly data-driven, at the end of the day there’s an element of ‘does this feel right?’ Does something seem off? Are these impressions even viewable? Is there an instance of fraud? Are we utilizing all of the verification tools we have at our disposal, etcetera, that only a human can evaluate.”  

But Humans Aren’t Perfect Either

However, before even talking tech, all agreed that marketers need clearly defined KPIs and a full view of business objectives before they can tackle attribution modeling. Emma noted, “So many clients freeze when you ask, but they need to understand their KPIs and the platforms they’re running across. Someone who understands multi-touch attribution realizes the need to address the full funnel, whereas someone who’s not as intimately exposed might think it makes sense to allocate budget to last-touch because that’s what they see as driving conversions. They don’t understand how different touchpoints work together.”

Cullen and Emma both pointed out a common problem: clients who optimize towards the wrong goal. For example, if an advertiser makes most of its sales offline, they may be better served by optimizing for top funnel brand awareness than web traffic.

Attribution Challenges IRL

When asked about interesting attribution problems they’ve helped clients crack, Ben referenced a client with the “pragmatic goal of trying to be less wrong.” This comment elicited a chuckle from the audience, but he continued, “It’s a great way to think about attribution because there’s never really an end-state. This client wanted to move away from last-touch, so we shifted to a fractional model which was very successful for them.” Although the fractional model helped the client optimize budget, it’s still an iterative process as the ecosystem and consumer behavior are constantly in flux.

Cullen and Emma both pointed out a common problem: clients who optimize towards the wrong goal. For example, if an advertiser makes most of its sales offline, they may be better served by optimizing for top funnel brand awareness than web traffic.

Similarly, Emma recently worked with a client who was focused on brand awareness, but stated KPIs as ROI and sales. Emma guided the client towards an MTA model that incorporated offline and third-party vendor data to create a less biased full view across channels.

There’s Hope Yet

Though tech limitations and walled gardens can present roadblocks for marketers, tools like Amazon Attribution, Google Analytics 360, and tech stacks that move away from proxy metrics are a step in the right direction for advertisers trying to marry disparate data sources.

MightyHive is Up to the Task

The MightyHive team, well-versed in Salesforce and Amazon Advertising solutions and globally certified across the Google Marketing Platform and Google Cloud, is uniquely qualified to help marketers solve complex attribution challenges. In fact, when asked what’s most exciting about working at MightyHive, the panelists all indicated that they love tackling new and interesting problems.

“We’re in a new frontier where we get to come up with processes to help companies figure out problems no one else is helping them solve.”

– Emma Tessier, Project Lead, MightyHive

“One Size Fits All” Isn’t MightyHive’s Thing

One such problem, which Cullen and the MightyHive consulting team recently addressed, was working to onboard a client’s first-party data to make it available to different attribution systems. While the most straightforward answer was to use an onboarding partner, sometimes that can yield match rates below 30-40%, which makes it difficult for advertisers to glean useful insights.

“We did some interesting things where we combined several approaches,” Cullen described. “We started with an onboarding partner, and for unmatched records, we used third-party and DMP data to model out lookalike audiences. We also included second-party data partnerships, and rounded them out by amplifying first-party login info, married to offline data where possible.” Using this strategy, the team was able to produce much more indicative results, giving the advertiser a 70-80% match rate.  

We’re Hiring… Big Time

If learning about new ways to approach complex digital analytics challenges excites you, MightyHive is hiring around the globe! Check out our careers page to learn more.