MADTECH

Articles

The Salesforce Ecosystem

· ·

What is Salesforce?

Salesforce.com is a comprehensive ecosystem of infrastructure, SaaS applications, programming interfaces, APIs, consulting partners, and learning management systems. In this article, we give you a high-level overview of these products and services. In future articles, we’ll dive deeper into the technology and specific products.

Sales Cloud

Salesforce.com started as a customer relationship management application (its stock symbol is CRM). It was founded by ex-Oracle executives, led by Marc Benioff, in 1999. What was revolutionary about salesforce.com was that instead of being application software installed on “on-premises” server hardware, businesses accessed salesforce.com via the internet, using a web browser. This approach led to salesforce.com being dubbed a “cloud application”. Another key innovation was the “platform” infrastructure approach, where the underlying database (built on Oracle technology) was abstracted to the user. Tables, columns, and rows were renamed as objects, fields, and records, and made accessible to users and administrators, allowing for customized data models for each organization.

The typical use cases for the original product was to manage the sales cycle for enterprise businesses, from lead generation through opportunity outcome, including the ability to manage accounts, contacts, products, territories and marketing campaigns.

Today’s Sales Cloud is the evolution of the original CRM product, with process automation, analytics, AI, and both declarative and programmatic custom development capabilities.

Service Cloud

Service Cloud is focused on customer service management. Whereas Sales Cloud primarily drives opportunity outcomes, the Service Cloud platform provides a customer service agent workspace, case management, CTI integration, and process automation.

The object model is similar to that of Sales Cloud, with the focus shifted from the Account and Opportunity objects to the Case object. Sales Cloud and Service cloud share very similar UI and administration interfaces, as well as process automation and custom development capabilities.

Field Service Lightning

Field Service Lightning (FSL) is an extension to Service Cloud, enabling Technicians, Dispatchers, and Contractors to manage field service operations using Salesforce. FSL offers:

  • Case, Work Order, Account and Contact Management
  • Scheduling
  • Mobile App with in-app guidance
  • Service Contract Management
  • Contact Center Console

Financial Services Cloud

Financial Services Cloud (FSC), based on the core platform, optimizes it to serve the financial wealth management vertical. Used by Registred Financial Advisors (RIA’s), the application provides:

  • Client and Household Profiles
  • Relationship Builder and Maps
  • Next Best Actions
  • Opportunity Insights
  • Advisor Analytics

Work.com

Work.com arguably was one of the quickest product launches to market at Salesforce. Spun up to address the issues faced by enterprises reopening after COVID-19 quarantines, the product was in a beta release by May (partners and developers had early access) and GA in June. Work.com features include:

  • Workplace Command Center
    • Monitors return to work readiness
    • Automates surveys to assess employee wellness
    • Enables shift management
  • Contact Tracing
    • Tools to track health-related interactions
  • Emergency Response Management
  • Workforce Reskilling

Community Cloud

Community Cloud enables businesses to create self-service portals for customers, employees, and partners, surfacing data from the core cloud applications. The AppExchange ecosystem has developed industry-specific templates for communities that enable businesses to create portal sites with minimal effort. Customized solutions can be developed with both declarative and programmatic development.

Typical use cases are:

  • Access to information about products and services
  • Account and contact management
  • Distribution of leads to partners
  • Enabling B2B commerce transactions
  • Case management
  • Knowledgebase access
  • Onboarding

Einstein Analytics

Einstein Analytics brings predictive analytics and artificial intelligence to the Salesforce platform. Businesses can analyze data both in Salesforce and in external data sources. The features include:

  • Analytics Studio: Dashboards and quick-start templates
  • Einstein Discovery: Automated discovery tools, modeling, and scoring
  • Einstein Prediction Builder: Predictive Analytics on Salesforce objects
  • Data Platform: ETL and APIs for managing data

Marketing Cloud

Salesforce has grown through numerous acquisitions. Marketing Cloud is an example of how it has integrated technology from these acquisitions to provide a fully-featured marketing automation toolset:

  • ExactTarget: Email service provider (ESP), content creation and curation, campaign management
  • Radian6: Social media listening
  • BuddyMedia: Social media publishing
  • Krux: Data Management Platform (DMP)

Marketing Cloud enables cross channel messaging, including SMS and Push. The Advertising Studio feature set includes integration with the Facebook, Instagram, Google Ads, LinkedIn, Twitter, and Pinterest digital advertising platforms.

Marketing Cloud is not built on the core Salesforce platform, but it can integrate with it and external data sources via platform interfaces, APIs, automated FTP, and file import/export.

Additional features include campaign automation, audience segmentation, and integration with Google analytics.

Tableau

Tableau, the industry-standard data visualization and analytics toolset, became a member of the Salesforce family in August 2019. While it is now a Salesforce company, it continues to operate independently,

Datorama

AI driven analytics vendor Datorama was acquired in July 2018. An innovator in marketing analytics, Datorama is used by ad agencies and in-house marketing analysts to measure and optimize marketing campaigns. Integration with Marketing Cloud creates a unique marketing automation stack that can measure campaign ROI in close to real time.

Pardot

While Marketing Cloud can serve both B2C and B2B use cases, Pardot is primarily intended to be used as a B2B marketing automation platform.

Pardot offers:

  • Lead capture
  • Landing page creation
  • Email marketing (ESP)
  • Content creation and curation
  • Content personalization

Similar to Marketing Cloud, Pardot is not built on the core Salesforce platform, but it can integrate with the Sales and Service Clouds to share data between the applications.

Commerce Cloud

Commerce Cloud represents another acquisition by Salesforce. Previously known as Demandware, Commerce Cloud brings B2C and B2B eCommerce to the ecosystem. A full-featured eCommerce platform, Commerce Cloud functionality includes:

  • eCommerce website templates for mobile and desktop
  • Campaign management
  • Product catalogs
  • Automated payment processing
  • Personalization
  • Integration with Marketing and Service Clouds

Customer 360

Customer 360 enables the Salesforce Customer Data Platform (CDP) solutions. The technology links platform entities (Accounts, Contacts, Persons, Subscribers, and Customers) across the ecosystem to provide a singular reference to the entities. Customer 360 allows for a single view of a customer across Sales Cloud, Service Cloud, Marketing Cloud, and Commerce Cloud, with integrated hooks for implementing controls to comply with local, national, and global privacy regulations.

App Exchange

AppExchange is the “App Store” for Salesforce. Listed on the exchange are resources and solutions for implementing and extending the functionality of Salesforce applications. All of the application packages are subject to a rigorous security review, and certified consultants are vetted through the Salesforce partner program. Resources and solutions include:

  • Apps: Ready to install third-party application packages
  • Lightning Components: Building blocks for Lightning pages and apps
  • Bolt Solutions: Industry-specific templates for apps and communities
  • Flow Solutions: Pre-built Flows for process automation and connections with external systems
  • Lightning Data: Third-party data sources with native integration to the platform
  • Consultants: Salesforce certified system integrators and independent software vendors

Integration

Integration with enterprise and SMB applications is often required with Salesforce implementations. The robustness, reliability, and accuracy of data across back-office applications can often make the difference between a successful implementation and one with poor user adoption metrics. The are several types of integration patterns:

  • Programmatic: Using the platform’s programming language, Apex, custom code can be written to source and consume data via APIs. Industry-standard REST and SOAP APIs as well as HTTP protocols are supported. Data formats include XML and JSON.
  • Declarative: Workflow Rules and Flow process automation can be used to consume APIs and services to and from external applications
  • Integration Platforms: Integration Platforms as a Service (IPaaS) that connect Salesforce to enterprise applications. Salesforce recently acquired MuleSoft, which arguably has become the “native” Salesforce integration tooling. Other integration platforms include:
    • Jiiterbit
    • Boomi
    • Informatica
    • IBM Cast Iron
    • DBAmp
  • AppExchange Packages: AppExchange packages can provide integrations for enterprise and SMB applications and services. The AppExchange covers a wide variety of industry verticals and use cases.

Heroku

Heroku is a Platform as a Service (PaaS) that has been integrated with Salesforce.com platform databases. This allows application containers running industry-standard application technologies to interact with Salesforce data and applications.

Applications can be written in Node, Ruby, Java, PHP, Python, Go, Scala, or Clojure. Postgres, Redis, and Apache Kafka can also run on Heroku.

Training and Certification

The Trailhead learning management system (LMS) is the Salesforce platform for training and certification. When used with the comprehensive documentation published publicly by Salesforce, users, administrators, and developers have the tools to learn the skills necessary to administer and develop applications on the platform.

Certification tracks are available for administrators, developers, and architects. The certification process includes in-depth testing to assess the knowledge of candidates. These tests emphasize hands-on skills as well as theoretical knowledge.

Enabling Technology Transformation

In summary, “Salesforce” refers to a multi-faceted, diverse ecosystem of related technologies:

  • Fostering enterprise revenue growth by enabling process automation, democratizing data analytics, and integrating AI technologies
  • Automating customer service
  • Enabling marketing automation with integrated data from first, second and third-party sources
  • Providing analytics that focuses on the key metrics driving profitability
  • Predicting next best decisions
  • Platforming eCommerce tooling with 360-degree views of the customer
  • Declarative and programmatic development
  • Integration with enterprise applications

With the robust integration between the technology stacks atop the platform and external systems, Salesforce is uniquely positioned to enable enterprise technology transformation.

How Dynamic Ad Insertion Works

· ·

What is dynamic ad insertion? How are ads delivered to OTT/CTV devices? Can linear, VOD and SVOD have different ad loads? Can linear advertising be geo-targeted from a centralized management infrastructure? We’ll attempt to answer these questions, at a high level, in this article.

After finally cutting the cord, we started to notice differences in how ads are being served in the OTT/CTV ecosystem. Yes, sometimes we see the same ad over and over, but in other cases there are a variety of ads, including retargeting ads. The experience reminded us more of online advertising than traditional linear television advertising. Researching the technology to understand how these ads are delivered, we found that the enabling technology for cross platform, targeted, audience based video advertising is Dynamic Ad Insertion (DAI). What follows is a description of one method for accomplishing DAI.

Just a few years ago, the lack of DAI technology was the roadblock to audience based targeting. Fast forward to today and the technology is not only available for online video, it is in use across platforms for linear broadcast, VOD, mobile and OTT/CTV, enabling targeting by platform, device, audience and geography.

To understand DAI, let’s examine “Blackout” programming in the linear television ecosystem. Broadcasters must occasionally pre-empt programming in a local market to comply with contractual agreements. Traditionally, this was accomplished by “splicing” the analog video stream to insert the alternate content. However, the level of effort and broadcast hardware configuration required by this technique limited its use for advertising insertions.

With the advent of Adaptive Bitrate Streaming (ABR) for online video, a digital, software based approach to alternate content insertion has evolved. ABR provides a method for delivering multiple video quality levels within a video stream. Video data within the stream is separated into fragments, and each fragment contains blocks containing payloads at different bit rates. This allows a client device or video player to choose the appropriate bit rate for the performance level of the connection at a given point in time. The video client determines the connection performance level by measuring the time the blocks take to load.

In order to manage these fragments of content, a manifest (or playlist), is sent alongside the stream. The manifest contains URLs (pointers) to the video files contained within the stream. In addition to identifying the bit rates available, the manifest contains metadata which identifies the type of content (programming, advertising, PSA, etc.) being delivered to client player. To enable DAI, an Alternate Content Decision Service (ACDS) parses this metadata, and provides an alternative manifest which identifies the bespoke advertising, blackout, or geographic content to playout. In this way, targeted advertising can be delivered to client browsers and devices. Note that this also can also allow for centralized management of both national and regional programming.

A placement opportunity information service (POIS) can be used to host metadata about the location and rights associated with alternate content opportunities (ADCS) within linear and on-demand video. API connections between demand side platforms (DSP) and POIS can facilitate programmatic workflows for advertising placements.

Connected devices that receive programming via unicast transmission (i.e. mobile phones), can return data on a per device basis for tracking and attribution. For QAM, ATSC and IP connected devices, the manifest contains data regarding playout of stream content. ATSC 3.0 promises to provide back channel data via the internet for over the air statistics.

An associated technology innovation is HTTP to UDP conversion. Traditional broadcast infrastructure uses Quadrature Amplitude Modulation (QAM) over cable or fiber distribution networks and ATSC over the air. Typically, OTT/CTV infrastructure uses the HTTP protocol over IP to deliver video via DASH or HLS. Therefore, two parallel sets of infrastructure have been required to fulfill distribution across devices. HTTP to UDP conversion enables replacement of the analog, on premise infrastructure used for linear and VOD with the cross platform video ad server and CDN technology used for digital platforms. This eliminates the costly legacy equipment and maintenance and allows for a common architecture across platforms. Additionally, the CDN enables and supports “Cloud DVR” functionality for client devices and browsers.

These technologies stack up to provide an “end to end” ecosystem for targeted, programmatic video advertising, delivered across platforms, with tracking and attribution pointers embedded within the content. As broadcasters, MVPDs, service providers, advertisers and agencies embrace these platforms and toolsets, the vision of “the right message, at the right time, to an interested consumer” will become table stakes for advertising in the medium with the highest mass reach of all: Television.

What Facebook and Google Know About Us

· ·

It’s no surprise that the duopoly collects data about us as we use their applications. But you might be shocked by how much they actually store and the way they use it.

Both the iOS and Android operating systems allow access to the sensors and data on our smartphones, across the majority of the apps installed on them. And, since most of us use our phones constantly, this data paints an extremely detailed picture of our lives.

What data do these platforms collect?

  • Location: The apps track your phone’s location constantly. If you use Google maps, Google knows exactly where your phone is, down to a few feet. Search uses location to provide useful results. If the facebook app is loaded on your phone it is tracking the device even if you are not using the app.
  • Search history: It goes without saying that Google knows every search you’ve ever made on a device. And, if you are logged into Gmail on several devices, they know your search history across all of them and can synchronize that data.
  • App use: Google can inventory the apps you use on your phone or tablet and the browser extensions in use on your laptop or desktop. Facebook stores an inventory of every app you’ve connected to your facebook account.
  • Email: Although Google no longer uses your gmail for ad targeting, they are still reading it.
  • Messaging: facebook keeps copies of your facebook messenger history.
  • Contacts: Both Google and facebook can read and store your contacts.
  • Relationships: facebook correlates interactions with your network of friends and utilizes machine learning to categorize you.

Yes, you can limit some (but not all) of this data collection. However, many of the application’s features are disabled or limited if you do. And most users don’t take the time to discover the numerous screens and settings that must be navigated and disabled to secure their data.

So what does this mean for consumers and marketers? Advertising revenue for media companies “subsidizes” free content and marketers can theoretically target consumers with offers that offer better ROI than the “spray and play” tactics used in traditional media. The platforms (Google, facebook and others) are the conduits for this strategy and have built extremely profitable business models around it. Marketing technology has evolved rapidly over the past few years, and the manner in which all this data are being leveraged is staggering. Just a few examples:

  • Consumers are grouped into hundreds, even thousands of segments, based on search history, sentiment (“likes”), browsing history, location (this can unveil related information such as socioeconomic status for instance), which ads have been clicked on, the history of their friends and numerous other data points.The platforms then sell these segments to marketers for targeted advertising.
  • Marketers parse their “first party” data (name, address, email, age, sex, socioeconomic status, purchase history, etc.) and anonymize it. The platforms, within their “walled gardens”, allow marketers to match this data with the platform’s users. For example, if Google thinks I am a single male, living in a relatively high net worth geographic area, it might categorize me as such and a marketer could send me ads for luxury sport sedans. On facebook, a “soccer mom” in a less affluent geographic area would get an ad for a minivan.
  • Because the Google and facebook applications are on both smartphones and laptop/desktops, the platforms can track users across devices using proprietary identifiers. So if for instance, you browse a marketer’s website using your laptop on your lunch hour, then make a purchase via your phone on the train ride home, the platform can monetize the laptop interaction with the marketer.
  • Marketers who have POS (point of sale) data can match credit card information on purchases to ad campaigns run on the platforms.
  • By leveraging wireless beacon and location technology, marketers can judge the effectiveness of their ads based upon where you’ve been and what you purchase. For example, let’s say you’ve walked through Times Square and were exposed to digital signage for a luxury beauty brand. You enter the brand name into the Google search app on your phone. 30 minutes later you walk into a retail location and make a purchase. Cha ching.

So is there anything wrong with this? Depends on your point of view. From a marketing perspective it’s not much different than how direct marketing has been conducted historically, it’s just more sophisticated. However, from a consumer’s privacy perspective, how do they know who is using their data and for what purpose? The Cambridge Analytica scandal is an example of what can go wrong.

In the EU, the enactment of GDPR is intended to help consumers be more aware of data collection and to enforce “Opt In” vs. “Opt Out” permissions. In the US, the state of California has drafted legislation with similar privacy regulations. If Governor Brown signs the bill, there may be legislation proposed in additional states. However, state regulation of online privacy may be a moot point, because the FCC is arguing that state law is superseded by federal jurisdiction in matters regarding interstate communication networks.

We’ve just scratched the surface on this topic. In future articles, we’ll be taking deeper dives into the technology, Stay tuned.

Cordless

· ·

So, we finally “cut the cord”. We had the full MVPD package with HBO, Showtime, STARZ, you name it we had it. All because I am addicted to NFL Redzone during fantasy football season and the provider (FIOS) only includes it with the premium package. Then one late summer morning we got “the letter”, informing us that HBO and Showtime were being removed from our package, and if we wanted to continue using them we could do so at an additional fee. Coincidentally, this communication came just days after the AT&T/Time Warner deal was approved.

I was livid. This was the final straw.

So, I called the provider to discuss the situation, and after two conversations, resulting in two different price quotes that increased my costs by 25% and 40% respectively, I told them: “Terminate my service. Tell me where to dump your equipment.”

I had done some research on OTT/CTV (same thing) and decided that Direct TV Now looked like the best fit. They were giving away an Apple TV if you signed up for 3 months. I already had a Roku box and a first gen Apple TV so I was all set. Or so I thought.

I’d watched Netflix on my Roku devices and had a decent experience, so I was not prepared for the drama of using OTT only… whoa. One word: buffering…

My first reaction was “the #&@%$$ MVPD is throttling the connection! That sucks!” After I calmed down, I actually paid attention to the modal message on the screen: “your network connection is underperforming….” Oh, maybe it’s the WiFi. One quick Target run later (lucky for me there is one literally around the corner) and I had a Netgear WiFi extender setup. Better, but still occasional buffering.

Because I have both a Roku and Apple TV, I switched back and forth between them to see if there was any performance difference. Negligible. But the user experience on the Apple TV, as you might expect, is vastly superior. Love that box. Interestingly, Netflix performs noticeably better than Direct TV Now on either device. I guess the death of net neutrality is showing its effects…

What about the ads? (after all this is the Advertising Perspectives blog…). Well, it’s interesting. Direct TV Now has a “Cloud DVR” feature. You can browse content by network and choose from a selection of on demand content. From this interface you can also record series. The “recorded” content works just like a hardware DVR. You can rewind and fast forward (skipping commercials). However, after a while, I found there was no need for me to record my favorite shows because they were (mostly) available on demand anyway.

On the flip side, live programming cannot be fast forwarded. You. Have. To. Watch. Commercials. All of them. Interestingly, everytime you switch to a different channel a new ad pod starts before the content plays.

So after about two months of OTT only I’ve come to these conclusions:

  1. The OTT user experience needs improvement to be on par with traditional MVPD products.
  2. Ethernet connections to the OTT/CTV hardware may be required for the best performance.
  3. My iPhone or iPad on AirPlay to the Apple TV actually works more reliably than the Apple TV or Roku on WiFi.
  4. You will see more ads.
  5. The ad targeting is interesting, and sometimes uncannily accurate. I saw different ad loads, on the same content, depending on the device.
  6. Unbelievably, there are empty ad pods, even on premium networks (ESPN Sunday Night Football… seriously?).
  7. Access to my entire content catalog across all my devices is awesome.
  8. Apple TV rocks.
  9. 1990’s era MTV reality shows are a hoot.
  10. I miss channel numbers.

This experience has left us fascinated with OTT ad tech. Stay tuned for more articles about it here at Advertising Perspectives.

 

 

Ad Exchanges and Real Time Bidding Update

· ·

So it’s been over four years since we first posted “Are Ad Exchanges and RTB the Next Big Thing?“. We are amazed that the article is still being referenced as a resource for this topic. Let’s take a look and see what’s developed since 2010.

Real Time Bidding (RTB), is now often associated with “long tail” remnant inventory and being risky in terms of transparency around placement of impressions. Buyers cannot always determine exactly what site their ads run on. Publishers have often referred to the technology as “Race to the Bottom” as automated auctions drive CPM’s lower.

Now for the good news. The technology developed for automated auctioning of inventory has been re-branded as “Programmatic”. Programmatic is as buzz worthy in 2014 as RTB was in 2010 and has the potential to be even more influential. Here’s why:

  • Data: The technology stack enables the ability to target audiences in real time based on first party (publisher or marketer proprietary), second party (syndicated) and third party (aggregated) data. This technology has not only driven digital marketing growth rates, it will likely revolutionize the linear broadcast, cable television and on demand ecosystems as well. Television advertising is shifting from targeting based on the demographics of the show to targeting specific audiences based on data. Sound familiar?
  • Workflow productivity: Traditional digital buying revolved around the same workflow as traditional media: humans negotiating the terms of an insertion order. That means phone calls, emails and lots of excel. Top talent coming out of college, often with student loans to pay off, are not flocking to media agencies to do low paid grunt work. Conversely, top engineering talent that once headed to the financial services industry to build trading desks are now building Demand Side Platforms (DSP), Supply Side Platforms (SSP) and Data Management Platforms (DSP).
  • Removing the auction elements of the stack and replacing them with a means to allow publishers to set prices, establish guarantees on placement and leverage premium inventory, enables targeted marketing at scale with benefits to publishers, agencies and marketers alike.
  • These exchanges can be public like RTB, open to all marketers, or private, where publishers can control who has access to the inventory and at what price.

Note the reference to “marketers” vs. “advertisers”. The technology has evolved far beyond just buying and displaying ads. With rich data integration, marketers have converted the “Marketing Funnel” into the “Consumer Journey”. They can put the right message to each interested consumer at he right place at the right time. Awesome, right?

Not so fast kemosabe. Challenges remain:

  • The ecosystem is fragmented: Venture capital and the desire for profits has fueled literally hundreds of start ups and not all of them will survive. There are so many technology vendors and inventory sources in the space that marketing campaigns at scale are difficult to manage. There is a wave of consolidation happening and this will likely accelerate.
  • Lack of transparency around price and placement: Agencies and exchanges profit by being principals in transactions, often not revealing the prices paid for the inventory. Marketers are starting to push back, in some cases bringing the technology in house. Marketers are also concerned with brand safety issues. They often do not wish their messaging appearing alongside objectionable content.
  • Fraud: The technology that enables the ecosystem is vulnerable high tech criminals intent on manipulating it for financial gain. Fraud is the biggest threat to the growth of the programmatic ecosystem.
  • Cookies: The cookie, which is the linchpin of most targeting mechanisms, is long overdue for replacement. The technology was never meant for what it is being used for and is arguably the weak link in the system. Additionally, cookies are not pervasive in mobile and mobile is the fastest growing digital platform. Marketers want to be able to target audiences across all platforms.

So, is Programmatic the next big thing? No, It’s already a really, really big thing. Depending on who you ask anywhere from 23% to 85% of marketers have embraced programmatic buying. As is often the case, the real number is probably nearer the middle of that range, but it’s still a big number and growing.

What’s next? Marketing Automation. Plugging customer relationship management (CRM) and internet of things (IOT) data into the ecosystem. Yes, your refrigerator will send a push notification to your car to tell you that you are low on milk and present you a coupon for your favorite brand at the retailer you are about to drive by. And since you aren’t physically driving the car, you’ll just tell it to stop and park.