Executive Summary:

Technology is of increasing importance to modern advocacy organizations. The private sector has adjusted to the innovations out of Silicon Valley but the public and political sectors still lag behind. Today, most of these efforts can’t compete effectively unless they structure a campaign to seriously utilize and leverage technology, data, and analytics.

In the not-so-distant past, advocacy organizations traditionally spent most of their resources on 30 and 60-second television commercials, direct mail pieces, and focused on major press events. But these are losing their luster. Since the first Presidential website in 1996 (see Bob Dole’s website here), people don’t engage exclusively through traditional and old-school media. Our modern world is clouded with an ever growing stream of information choices. Potential voters now watch the news live online, experience primary state rallies, previously limited to Iowa, New Hampshire etc., on Meerkat and Periscope. Recent Presidential announcements have been an all medium event, generating strong social traffic on Twitter, Facebook, Instagram well before the traditional news cycle could capture it on the 6 PM news or in the morning paper. The modern campaign has to grow with this new frontier because those they wish to influence have already moved on from their “not-so-distant past” strategies.

Modern advocacy organizations are 24/7, online and in person, and increasingly driven by large scale data analysis. Sasha Issenberg’s 2011 book The Victory Lab explains “suddenly the crucial divide within the consulting class is not between Democrats and Republicans or the establishment and outsiders, but between . . . empiricists and the old guard."

Every day, technology will continually transform the way we think about outreach: door-knockers will have geo-located data to ensure that they knock on the right doors (by matching a cell-phone with a physical address) with a message tailored to a specific person (based on associating an interest graph with a voter ID) and even when they are most likely to be home. Richard Viguerie’s pioneering of direct mail in the 1970’s was quickly advanced by Karl Rove’s personalization of direct mail for different types of voters. And personalized direct mail has gone from cutting edge, to omnipresent on every major campaign across the country. While most “old school” methods will still remain similar, the rhythm and reason is quite different. In a modern advocacy organization, the “why” is known and leveraged. A campaign that doesn’t embrace modern ways for reaching and understanding people open themselves up to misallocating resources and having their messages fall flat.

Digital is not just flashy; it’s good politicking and economical. Well-designed technology, specifically code, can scale quickly and relatively cheaply from the Iowa Caucuses to 270 electoral votes. Human-based structures have difficulty scaling and introduce more and more levels of human error (in particular organizational structures get messy going from an organization of 50 to one of 800).

On day one, technology and data need to be at the heart of an organization. This means hiring a technical team leader, like a Chief Technology Officer. But digital and analytics aren’t merely reserved for premier Presidential candidates, even local efforts can use basic digital and analytics tools to gain an edge.

During most cycles, organizations fight over hiring the staffers with the best experience. Today this experience also comes with liability and risk. Technology is moving so quickly that their experience can be quickly out of date, but they may rely upon their experience to the detriment of the organization. Like Clay Christensen’s Innovator’s Dilemma, the Political Guru’s Dilemma occurs when staffers and consultants rely on old, previously successful tactics but lacking the self-awareness to know those tactics and strategies are out of date. They are passed by upstart campaigns using newer tactics, spending less money, and upsetting the balance of power. For a modern advocacy team, it is more useful to have a staffer hungry to learn new techniques, intellectually honest enough to test their ideas (not just assume they’re correct), and seriously leverage data, as opposed to trusting their gut instincts. In the 21st century campaign world, gut instincts are increasingly out of date.

In order to utilize digital tools and analytics in an impactful way, advocacy organizations need to 1) pay technical staffers competitive salaries (good talent is not cheap), 2) create a cultural atmosphere conducive for digital people, 3) ensure the digital organization is a direct report or a senior level report within the organization, and 4) create an organization that can scale quickly, with delineated ownership and well-understood, well-followed workflow.

Employing digital tools properly means more than hiring a Chief Technology Officer from Silicon Valley; many parts must function together. Advocacy organizations must also 5) be aware of the two-sided social media sword, 6) make digital a native part of the campaign, 7) engage in an authentic and timely way on social media, and 8) most importantly, the organization needs to accept a willingness to fail. Digital strategies will often fail, so it is important to experiment early and often. You don’t want to find out when it is too late. Small scale experiments will help with focus and resource allocation to prevent failure.

If organizations don’t embrace a digital and analytics oriented strategy, their respective issue or candidate will be left behind. The rest of the world has moved on and are retraining their minds to adopt ideas far differently than before. What worked a few cycles ago (or sometimes even last year) is increasingly irrelevant to make new inroads and cause people to take action. The future resides in the hands of those who embrace new strategies. Experienced based strategies are not to be discarded, but are simply one hypothesis to be tested against the data.

In this guide, we will explore how to adopt and build your team into a technologically competent advocacy organization. This guide is written for modern campaigns; a bridge between empiricists and the old guard, for those who may want to work on these campaigns, consultants who will assist these campaigns and the media who will be covering these campaigns. The sections are broken out by the number of people needed to reach or the size of a given organization. Our recommendations are a collection of experience from hundreds of minds all across America. We hope this is the start of an edifying public discourse on sharing best practices because we know this information will be out of date sooner than we desire.

The reality of our new world is before us; how will you respond to see the change you want to see in the world? Outline:

This guide will layout the basic structure of how a campaign and issue advocacy organization should utilize technology within their organization. We cannot cover all possible cases; however, we did our best to be exhaustive. The primary goal of this guide is to offer a general rule of thumb for your reference.

Executive Summary: Outline: Organization Skills and Specialties UX Design and Creative Software Engineering Product Manager Digital and Social Media Advertising Data Analysis Attracting Tech Talent Appropriate Compensation Conducive Work Environment and Organization Campaign Levels National Campaign Characteristics Overview Team and Organization Pointers Scaling requires a good foundation Plan and Strategize Hiring Let Experimentation and Data Lead the Way Recommended Tools Recommended Products to Build Statewide Campaign characteristics Overview Team and Organization Pointers Avoid “Code-Monkey” Syndrome Avoid Glitzy Technology Be Led by Data But Don’t Be Overly-Confident Recommended Tools Recommended Products to Build Local Campaign Characteristics Overview Team and Organization Pointers Digital Marketing is Relevant: Forget Product Development Digital Isn’t Your Golden Ticket Recommended Tools Recommended Products to build Case Study Romney 2012 Obama Campaign 2012: Conclusion:

Organization

The following details the scope and roles within the digital organization. Skills and Specialties The following specialties should exist within a Digital and/or Technology team: UX Design and Creative Data Science and Analysis Software Engineering Product Development Digital Advertising / Social Media Advertising Product Management and Project Management

We explain each in further detail below: UX Design and Creative Responsibilities - Owns visual and User Experience (“UX”) design.

Designers vary in both skill and areas of subspecialties. A graphic designer does not have the same skill set or abilities as a user experience designer. Here are a few examples of sub-specialties within design:

User Experience Designers – Spilt by mobile and web platforms, user experience designers specialize where visual design and interface design meet. UX designers focus on how users flow through a product. They are typically quite capable visual designers but primarily specialize in channeling the user’s point of view to design the best flow possible to solve a specific problem. Visual Designers – From colors to icons, visual designers create the appearance, the look and feel, of design. To a visual designer, solutions are usually based on aesthetics rather than functionality (which is the strength of a UX designer). Product Designers – Combining the role of a product manager and a designer, a product designer enjoys coming up with the ideas and priorities of what to design and build. This type of designer understands business goals and product priorities. Product designers can typically work without the need of product manager or product requirements. Product designers are rarer and are typically weak on some other portion design. User Interface Designer – Every design starts with a basic outline, also known as a wireframe. A user interface designer specializes in taking ideas from thoughts into a tangible wireframe for designers and engineers to cost and fully understand the scope of a project. User interface designers are not strong in visual or interaction design but are somewhat capable at user experience. Graphic Designers – Very similar to visual designers, graphic designers are usually very strong visually and aesthetically-focused designers. Visual designers typically have more experience with technology products than graphic designers. Graphic designers usually specialize in “flat” oriented design such as print materials, icons, email, etc. Animation Designers – Animation designers makes technology feel alive and fun. An animation designer takes all the parts made by an user experience and visual designer and crafts brings a design alive. Animation designers typically work closely with user experience designers. Most designers of any type have experience with animation but don’t specialize only in animation. With the popularity of mobile apps, animation design has started to become a subspecialty. Animation designers are very important to make a design feel fluid, usable, and connected.

Software Engineering Responsibilities - Owns the development, management, and execution of technical product development.

Software is the heartbeat of any product. Engineers bring life to products by writing code to take designs and make them work in the real world.

As with design, engineers come in various stripes. Front-end engineers – Design and product oriented, a front-end engineer focuses on what designers and product managers care most about: what users see and touch. These engineers are on “the easiest” end of the code base and typically don’t work on complex engineering logic. Front-end engineers consume back-end software logic and put a user-facing design to it. Their primary languages are HTML and Javascript. Mobile engineers – A derivative of front-end engineers are mobile engineers who work in native mobile development - Apple and Android apps. They only work on mobile native code typically don’t work on more than one platform. Their primary language is Objective C or Swift. Back-end engineers – Using the metaphor of building a house, front-engineers come in after the drywall is up and finish the job with molding and fixtures. Back-end engineers build the frame and roof of the house. Back-end engineers focus on what the application does (like upload photos or take payments) and builds the logic on which front-engineers can build a design. Infrastructure engineers – Continuing with the house metaphor, infrastructure engineers lay the foundation on which everything else is built. Infrastructure is primarily concerned with scaling and the architecture (“design”) of software product. They take goals of the product and build a foundation to support it. Data engineers – With technology, you can track just about anything users do. Data engineers focus on developing the pipeline of information and ensuring its accuracy. They constantly measure the accuracy and scalability of a database and product. Quality assurance engineers – Quality assurance engineers specialize in finding human mistakes and fixing them. Though only needed at large scale engineer teams, quality assurance (“QA”) is important to prevent unnecessary work and prevent small mistakes to turn into disasters.

Product Manager Responsibilities - Owns the development, management, and execution of a product.

If design determines what something will look like and engineering determines how it will work, product determines what you need to build in the first place. A product manager’s role is to deliver maximum return on investment for the business and users. It is often said that product manager is the “user advocate”, speaks on behalf of what users want. Product managers develop the ideas, the requirements for how a feature will work and metrics for success. Product managers are important because their expertise is delivering more for less and managing performance of technical talent.

Product managers take concepts from ideation to execution. They understand how to leverage fully design and engineering talent to move numbers for the organization. Without a product manager, execution can be inefficient which usually leads to strife between engineering, design, and higher ups. The product manager is the objective voice to keep everyone on track. They balance all of the needs of an organization to keep all stakeholders happy and manage expectations. They know how to motivate technical talent, manage expectations of stakeholders, and develop a vision that as an impact.

In most digital and technology organizations, product managers own design, but engineering is a separate organization. This is helpful as building products can create tension between the teams. Engineering typically wants to do the least amount of work possible to deliver a feature, while design wants everything to be included from Photoshop file to code. Product managers help balance this tension to build a superior product. Competition between the teams (with product managers being the arbiters) lowers cost from design and engineering while still delivering the product to end-user.

Digital and Social Media Advertising Responsibilities - outward digital communication, such as Social Media Advertising, Digital Advertising, Email Marketing, Paid Advertising

Once a software product is built, the digital advertising team takes over and spreads the word to the world. Digital and Social media advertising drives engagement to a product using external channels, such as social media, email, paid advertising, and blogging. Digital marketing does not include technical product development, which is design, product management, and engineering.

Digital Marketing is a huge field. Here is a high-level summary of the most common specialties Social Marketing – “Tweets” to “Snaps” that go viral are intentional. Social media marketers own the strategy and manage the execution on engaging on social media channels, such as Twitter, Facebook, blogging, etc. Social media marketers are usually quite creative and entertaining. Their ultimate drive is to create engaging content that spreads like wildfire. Most social media-marketing experts have a brand and community marketing background. Community Management – When you build anything online, you will have user feedback and complaints. Community managers focus on developing and managing a community of users. Typically, this role is very relevant for social media products. Their goal is to keep users happy. Email Marketing – As the number one communication channel on the Internet, email marketers specialize in driving results from email. These marketers are often very metrics and data oriented. Email marketers own the design of emails and cadence of when they send. Email marketers should also be experimentation focused and enjoy running A/B tests. Email marketers are typically very metrics oriented and ROI driven. Direct / Performance Marketing – As the most metrics-oriented in the group, direct marketers specialize in anything that requires a measuring a conversion rate. They are usually flexible on the channel they work on as long as the goal is to drive the improvement in a metric, like number of donations or selling products. Performance marketers work with advertising products like banner ads, video ads, and search ads. Brand Marketing – What you feel and think when you hear of a company is the brand. Brand marketers handle the message and the purpose of a marketing campaign. They also usually include press relations and blogger relationships.

Data Analysis Responsibilities - owns what users do and how to best reach users, such as Data analysis, User insights, Recommendation engines

Just because you can track something doesn’t mean you fully understand what is going on. Analytics / Data Science consumes what you track and measure then creates something understandable and applicable to the business. Data analysts turn a long CSV or large Excel file into something actionable. Their analysis helps the organization to make better decisions in the future. Typically, data analysts and data scientists are not responsible for making the decisions but informing decision-makers. Most of the time data analysis is never black and white, true or false, correlation-or-causation. Therefore, wisdom and experience are needed to know what analysis should be done within the organization and how to apply the analysis.

Data scientists are more advanced and experienced than data analysts. Typically, a data scientist has a Ph.D. and specializes in very complex algorithms. Data analysts are more comfortable with analyzing the performance of a new feature release, rather than developing a recommendation system.

Attracting Tech Talent

Campaigns generally need tech talent much more than tech talent needs them. This is atypical and most campaigns haven’t adjusted their strategies. Most staffers for an advocacy organization are veterans of similar efforts. But most technologist who work on campaign and causes are not political at all. A recent study found : “Out of 339 total staffers in digital, data, and analytics on the 2012 Obama campaign, only 22 percent gained their primary professional experience in politics. Many came from the commercial sector (17 percent) and technology or data analytics firms (14 percent).” That study also found a discrepancy between digital staffers on the left and right finding that “From 2004-2012, Democratic campaigns hired 503 staffers in digital, data, and analytics, compared with 123 Republican staffers.

Why non-technical and technical talent join campaigns is quite different. Non-technical talent are looking often for a stepping stone to higher position in a future campaign or at a consulting firm. Digital talent rarely seeks political influence post-campaign, and they likely have various other terrific opportunities. Technical talent is drawn to working on difficult challenges, the mission of the organization, and learning from unique experiences. Technical is almost always recruited. The demand for technical talent outside the advocacy and political world is immense, while non-technical politicos are not drifting from their main industry.

With so much competition, what strategies can a campaign or cause implement to attract technical talent?

Appropriate Compensation

For starters, digital and technical talent should be fairly compensed, close to their salaries in the private sector. Though pay is not a huge driver, it opens the door and starts the discussion. Most technical talent will know they are taking a pay cut, but the salary should not be insulting. The opportunity cost for technical talent to join a campaign is massive and campaigns and causes need to be considerate of the sacrifice - this is a sea change in political recruitment.

To most in the political industry, paying staffers in their 20s high salaries makes them squeamish; however, several non-technical roles are compensated very handsomely, like fundraisers, outside lawyers, consultants. Further, several roles that are well-compensated in traditional advocacy world now look quite dubious in a modern era. When digital is done correctly, the financial benefits are compounding and scale.

Google embraces this concept of outlying returns with good technical talent. One good engineering hire can output 10X of an average technical hire; not all engineers are created equal. Good talent should be compensated at a much higher rate since their work scales.

Conducive Work Environment and Organization

Comparing the work environment of a technical company or a startup to a cause or campaign is night and day. Google and Facebook have their own private bus system. It is the norm for a startup to pay for breakfast and lunch for everyday during the workweek. Working from home is considered a right rather than a privilege.

It is important to know in what type of an environment technical talent works. Offering similar perks isn’t necessarily the answer, but it is crucial for a campaign to mimic the mindset and culture from the technology world. This is important to create a thriving team. Conducive atmospheres are often very open, relaxed, and casual (often resulting in employees putting in long hours). The Dresscode for an average startup is anything and everything opposite of business attire. Floor plans are usually completely open with no cubicles.

What makes tech companies successful isn’t just cool perks but the way work gets done. This is probably the most important and significant structural change for a campaign to adopt. Technical talent prefers flat organizations, collaborative teams, self-empowerment, macro-management, individual responsibility to make impactful changes, and the ability to fail. This is the exact opposite of most campaigns and causes. Most campaigns are heavy handed on management and prefer to limit freedom of creativity to all ranks. It is understandable due to the visibility of a campaign and the unforgiving nature of the media. Even granting this point, it is self-evident that technical talent won’t thrive or produce results in such an environment to develop products.

A “willingness to fail” is probably the scariest difference between a tech enabled environment and a more political environment. Tech talent’s view on failure stems from how innovation happens. Thomas Edison said, “I have not failed. I've just found 10,000 ways that won't work.”

Campaigns are risk averse, worrying about their candidate or staff making a gaffe at any moment. This same approach to risk aversion, while sensible for some aspects of the campaign, must be carefully reworked for digital and analytics based operations. Campaigns shouldn’t let perfect be the enemy of good. Failing often, but failing cheaply, is one of the most important lessons campaigns should adopt from the technology world. Otherwise, with a risk-averse oriented strategy, the efficiency gains that come from technology won’t appear. Digital and analytics require constant experimentation, and quite often these experiments fail.

Failing through iteration does not necessarily equate to a blunder. A new mobile app is executed flawlessly but doesn’t gain traction with users. This isn’t a blunder. Something was assumed that ended up being wrong. Even when campaigns attempt to operate “perfectly” and executes as planned, mistakes still happen on a regular basis. Following a blunder, campaigns’ instincts are to hunker down even further, implement more control, and fear experimentation. This reaction works against the iteration that leads to developing game-changing innovation. The ability to accept failure and the humility to try new ideas are ingredients that leads startups to overtake larger incumbents. Campaigns and causes should be constantly iterating on a small scale to explore ideas, anticipating many of them won’t work out.

When Sheryl Sandberg worked at Google, co-Founder Larry Page praised her for making a costly mistake for Google. Page said, “I’m so glad you made this mistake because I want to run a company where we are moving too quickly and doing too much, not being too cautious and doing too little. If we don’t have any of these mistakes, we’re just not taking enough risk.” Empower your team to be bold on digital and analytics by trying new ideas with an understanding that failing often and failing cheaply is okay.

Campaign Levels

No campaign is the same: They vary in size, reach, and resources. We broke out the guide below into National, Statewide, and Local. The buckets are split by the number of people expected to be reached and budget. Within each tier, we’ve outlined a basic digital team structure and highlight high level recommendations or common mistakes based upon discussions with experts. At the end of each section, we have a short list of recommended tools to either buy or build in-house.

National Campaign Characteristics Needed contact: > 15 million Digital budget: > $25 million Overview Do Build a core team Iterate and experiment Build something that scales Don’t Take shortcuts for time Be wary of third party software Silo your data Team and Organization

Product Team Engineers: 20 - 30 Designers: 6 - 12 Product Managers: 5 - 8 Data Scientists/ Analysts: 5 - 15 Digital Marketing Social Media: 5 - 8 Email Marketing: 3- 5 Brand Marketing: 3 - 5 Performance / Direct Marketing: 5 – 10

Most national campaigns and causes follow phases: a primary or regional level and then a general level applicable to everyone. Both stages have different challenges, mostly centered around resources, momentum, and the scope of who needs to be reached for a given desired outcome. The numbers try to reflect this nuance; however, please look refer to the state-based recommendations for regional contests.

Pointers Scaling requires a good foundation

Not everyone should or can be hired immediately. It is important to view your first hires as critical and not throwaway talent. Typically, the first members of the technical team are viewed to have less experience because the risk of failure is greater and money is tighter. First hires set the culture of a team going forward. Compromising on quality at this phase will create a culture of lower quality work that will be difficult to unwind.

Building a good foundation also applies to technical product development. All technical debt - poor hiring decisions that have to be fixed in the future - eventually has to be repaid. Poor decisions on technical infrastructure, coding practices, and design thoroughness will escalate to cost of making future changes. A campaign or cause will have to hire more people than necessary if a good foundation isn’t laid from the beginning. “Interest” on the technical debt compounds as an organization continues to expands, slowing down more and more technical talent.

Plan and Strategize Hiring

Campaign and causes on the national stage often have more resources at their disposal compared to those at state and local levels and therefore can often draw better talent. Despite being comparatively easier, it is important to form a recruiting team to go and out find the best. The speed at which national issues move makes it difficult to remain disciplined. As a team scales, don’t rely on inbound leads and hiring out of desperation for any body in response to rapid growth. Build a network of potential candidates who represent the best and brightest. Top talent is typically in a position they already enjoy and won’t leave everything at the drop of a hat. As an effort builds momentum and public visibility grows, it will become easier to get the best minds in the field. Create a recruiting roadmap with milestones and key hires as guideposts.

Let Experimentation and Data Lead the Way

At this scale, data is plentiful, but it comes with a need to use it wisely. Even a shoestring national cause or campaign can run a simple A/B test on email blasts. Despite this, too often the fear of mistakes or bad publicity prevents campaigns from utilizing this simple and effective testing methodology.

Campaigns at the national level draw so much public attention that little things quickly turn into a mistake waiting to explode on Reddit, Facebook, or Twitter. This unfortunate reality creates fear within the organization; driving prioritization and resourcing haywire. While the media may force this upon a campaign, it’s a resourcing methodology and ROI strategy. It leads to decisions being defined by “saving face” rather than optimizing for a victory.

If the job of an advocacy organization is to achieve a specific goal, experimentation is the only viable path forward. Testing validates the efficacy of almost every tactic. It proves that resources are being well-spent and are impactful. It is not enough to operate only off experience because if a team ends up being wrong, you won’t know until it is too late.

Run experiments on emails, on landing pages, on product features, on messaging. This will save a great deal of money and prove what works. There are a host of third party tools that will make experimentation easy.

Recommended Tools These are the tools we recommend to outsource too: Infrastructure AWS Data and Tracking Google Analytics Engineering Tools Github Email marketing Email Service Provider: Bronto or StrongView Paid marketing Channels: Google SEM, Facebook Ads, Twitter Ads Social media marketing Channels: Facebook, Twitter, Instagram, YouTube, SnapChat Monitoring Tools: Hootsuite Events Eventbrite Ecommerce Stripe, Braintree

Recommended Products to Build Custom web properties with a content management system to allow the web properties to be editable without the need of engineering Internal A/B Testing tool Custom mobile app using “short-cut” frameworks for faster development A consistent and reliable way of tracking users across platforms, from email to phone banking to website visits.

Statewide Campaign characteristics Needed contact: 5 – 15 million voters Digital budget: $5-$25 million

Overview Do Build a core team to lead outsourced talent Outsource key features to have flexibility in execution Focus product development on only essential products Prioritize data integrity and collection to help the entire campaign over user driven features Don’t Be blinded by unnecessary and glitzy technology utilized by national campaigns and causes Run unscientific experiments with low amounts of data Live with your data siloed

Team and Organization Scale to at least the following Product Team Engineers: 2 - 6 Designers: 1 - 3 Product Managers: 1- 2 Data Scientists/ Analysts: 1- 2 Digital Marketing Social Media: 2 - 3 Email Marketing: 1 Brand Marketing: 1 - 2 Performance / Direct Marketing: 1 - 2

Pointers Avoid “Code-Monkey” Syndrome

At this scale, it is common to rely primarily on outside expertise and consultants to drive most campaign decisions. While hiring leading experts may be impossible, don’t hamstring your internal staff by relegating them to serve as “code-monkeys” for external experts. Look for young and hungry talent. Adopting a culture of experimentation with younger, smarter and more capable technologists (the DNA of a startup) will often lead to surprisingly good results. Building an effective organization is about creating an all-star team, not a team of all-stars. Avoid Glitzy Technology

Since the game-changing Presidential election cycles of 2008 and 2012, technology has been sold as cookie-cutter and savior of all woes. This assumes that the culture that produced technology is not important. Technology is developed for a specific purpose that may not be applicable to a state-level campaign or cause. There is often a deep misunderstanding why these tools were first developed and how they were integrated. Considering implementation costs and time needed for deployment, it is best to avoid buying these tools as they often have a marginal impact. For example, Obama’s door collaborative knocking app was built primarily as a tool for their internal field ops, only later it was converted into a public app. The primary users of the app where people within the campaign, not the average consumer (despite how the media spun it).

The greatest feature about technology is scalability; but this attribute cuts both ways. The ROI of a technology tool is significantly less for state and local focused campaigns or causes. “Old-school” techniques have a threshold for effectiveness and efficiency. Since 2008, a lot of advocates feel like these traditional techniques now have a zero ROI, but the incremental creation of value between traditional and modern tools at the state and local level is questionable. The goal is to solve a problem, not appear in a tech article. Sometimes technology is not the answer.

Be Led by Data But Don’t Be Overly-Confident

Robust data analysis has swept the field as a “must have” for campaigns. This sentiment is valid in that data analysis is incredibly valuable; but scope and applicability are essential. At the state and local levels, the amount of data is likely not enough to be statistically significant (aka “provable”) for very narrow and more actionable analysis; however, it is important to still be led by data. Data can tell you a story about what is going on and how you can improve but must be broad in scope. For example, analyzing how many people are likely to drive to work in a metropolitan region is achievable. Answering how many financial analysts who drive a blue coupe are likely to drive to work in that city is a much, much harder question. Generally speaking, the narrower the question the more data is needed. Recommended Tools Infrastructure Outsourced hosting company Data and Tracking Google Analytics, Mixpanel Engineering Tools Github Email marketing Email Service Provider: Sailthru Paid marketing Channels: Google SEM, Facebook Ads, Twitter Ads, Social media marketing Channels: Facebook, Twitter, Instagram, YouTube Monitoring Tools: Hootsuite Events Eventbrite Ecommerce Shopify Mobile app and web app Wordpress, third-party web CMS platforms, out-of-box mobile app solution Recommended Products to Build None: Plenty of off-the-shelf products available Local Campaign Characteristics Needed contact: Less than 5 million voters Budget: Less than $5 million

Overview Do Hire very smart and scrappy talent Focus on doing the basics really well Use third party tools as much as possible Don’t Build your own technology unless absolutely necessary Reinvent the wheel; rely on real-time learning Waste money on a lot of online paid advertising Team and Organization Scale to at least the following Product Team Engineers: 0 - 1 Designers: 1 Product Managers: 0 Data Scientists/ Analysts: 1 Digital Marketing Social Media: 1 Email Marketing: 1 Brand Marketing: 0 Performance / Direct Marketing: 1

Pointers Digital Marketing is Relevant: Forget Product Development

Regardless of the scale, everything a campaign does can be leveraged to expand its reach with digital products. At a minimum, a candidate should create a basic social profile, such as Twitter, Facebook and Instagram. Smaller campaigns and causes don’t need to worry about filters or camera angles or other niceties (hyper polish may actually be viewed as disingenuous); the point is simply to populate these social media accounts with material and help followers engage in a conversation. Organizations should aim to post new material everyday and develop a relationship with their followers, with the founder or core leader of the organization largely responsible for brand and messaging on social.

Anything focused at the local level does not need product development. A basic digital team consists of digital marketing and a design resource to help with graphic and visual design. If another technology is needed, it should be purchased from a vendor. Building custom technology is inefficient and unnecessary.

Digital Isn’t Your Golden Ticket

Like statewide campaigns, smaller efforts can get wrapped up in the buzz of what happens with technology at the national level. Digital strategies deliver the highest ROI at scale. The size of a local race is a Catch 22: it makes digital less relevant from an ROI perspective, but it makes in-person contact possible. Also, targeting in a hyper-local way is a weakness of technology. It is easier to find 1,000 people nationwide interested in a given subject, much harder to find and reach a 1,000 people within a single city limit. Don’t get distracted by grand visions of technological magic when door knocking and phone calling every possible voter may just do the trick. While the modern world can feel like a vast digital ocean, people still prefer personal contact. Recommended Tools These are the tools we recommend to outsource too Infrastructure Hosting company Data and Tracking Google Analytics, Mixpanel Engineering Tools Github Email marketing Email Service Provider: Sailthru Paid marketing Channels: Google SEM, Facebook Ads, Twitter Ads, Social media marketing Channels: Facebook, Twitter, Instagram, YouTube Monitoring Tools: Hootsuite Events Eventbrite Ecommerce Shopify Mobile app and web app Wordpress, out-of-box mobile app solution

Recommended Products to build None

Case Study Romney 2012 vs. Obama 2012 President Obama’s reelection campaign built on the foundation of his 2008 race. The scope of the Obama campaign vision was nothing short of breathtaking: they built the world’s most sophisticated digital and analytics campaign every assembled. Perhaps more extraordinary -- they built the operation almost entirely internally with minimal reliance upon outside products. While some have credited this as a startup approach, in fact it’s actually very unlike a startup, as few startups have the resources to become experts in every subject and to develop every solution from the ground up. It was a risky strategy that had it’s share of failures along the way. Only a 20-month campaign would have enough time to build these tools internally.

The Romney campaign, on the other hand, was in a very different situation; they had an entirely uphill climb. Nearly every aspect of their job was more difficult: 1) They had to start from the ground up, 2) Romney was running in a competitive primary cycle, not securing the nomination until April 2012, 3) It was more difficult for the Romney team to recruit the best and brightest in the field from the tech community, 4) Obama had assembled a massive 13 million person e-mail database from the 2008 campaign 5) Obama as the incumbent President, coupled with an early adoption of digital technology had enormous advantages on social media with Facebook likes and Twitter followers dwarfing the Romney’s initial numbers 18 to 1 and 45 to 1 respectively (11 months before the election).

Despite the negative press about the Romney digital operation, nearly all of which was post-mortem, what is most surprising is the relative parity that both operations were able to achieve. As TechCrunch would argue before the election: “[Romney’s team] won digital parity with the Obama juggernaut by partnering with the brightest Silicon Valley firms. For instance, though Obama has four times the number of Facebook fans, both Romney and Obama have similar levels of users actively commenting and sharing content online.”

Romney 2012 The Romney campaign recognized that the best way to compete from a deficit was to focus on what they could do well in-house and use the best available off the shelf tools to complement their efforts. Also, this allowed them to compete with the Obama operation with a much smaller team at far less cost. They used Rally.org for fundraising, Tout.com for short form videos, Eventbrite for events and handed Square readers to RNC volunteers to solicit field donations, turning them “into an army of walking cashiers.”

All that being said, the Romney campaign did face some serious challenges from which they could not recover.

A. No initial digital strategy, ramping up too late. In April-May, as Romney was wrapping up the nomination the digital team had only 14 staff, compared to around 250 on Obama’s effort. The digital team would scale up significantly over the next 60 days to around 140, but still far short of Obama’s effort.

As Zac Moffatt recalled his assignment, “Can we do 80% of what the Obama campaign is doing, in 20% of the time, at 10% of the cost?”

B. Relative lack of Analytics, outsourced to Optimizely The Romney digital operation had a 20 person analytics team for email, website, online advertising, Facebook postings. Despite the fact that Optimizely was founded by an Obama 2008 campaign alumnae, they used Optimizely to perform A/B testing of their website and email solicitations.

C. ORCA The Romney campaign created an app to deploy volunteers on election day, to provide the campaign with analytics on polling places activity and help the campaign direct GOTV efforts in battleground states (see Ars Technica’s definitive account here). On Election Day, however, the program was so overloaded with incoming data that it quickly exceeded capacity and crashed. In other cases login information didn’t work to begin with. At another point, the Comcast connection at Romney headquarters shut down because Comcast thought the traffic was a Denial of Service Attack.

The Romney campaign had invested heavily in the Orca program, the brainchild of Romney’s Director of Voter Contact Dan Centinello and campaign Political Director Rich Beeson. As Romney’s Communications Director Gail Gitcho explained “The Obama campaign likes to brag about their ground operation, but it’s nothing compared to this.”

While Orca was stress-tested using automated testing tools, there was no attempt to see how Orca would perform over the Internet, and no field testing for the users that would have to use the app on Election Day. One of the authors of this piece (Derek) was in New Hampshire on Election Day with Lawyers for Romney, when he asked to see and test Orca in the preceding 48 hours before the election he was told that it was still not ready to be tested yet -- in fact the Orca application wasn’t turned on until 6:00 AM on Election Day.

These users, many of whom were 50, 60, 80 year old veteran volunteers, were sent a 60+ page PDF instruction book with the voter rolls at 10:00PM the night before the election. Many typed the URL of the website in incorrectly. For others their passwords didn’t work and attempts to reset them failed. Hundreds of volunteers in Colorado and North Carolina were given the wrong passwords altogether, which weren’t reset until 6:00 PM on Election Day.

Essentially nothing about Orca worked properly. ORCA was an outsourced project built by Microsoft and an undisclosed app creator, rather than built internally. It was run entirely by the Political department, rather than by the digital or analytics side. As several have told us in various interviews, no one on the digital side would have allowed a program of this magnitude to be field tested on Election Day.

Overall there was a lack of organization for this project. No one had the technical chops to own the timeline, development milestones, or to call the Microsoft managers out for missing checkpoints. Even a dry run of the tool should have included demand levels for Election Day. Moreover, any application that required volunteer involvement should have included robust training for those volunteers with the application in hand.

ORCA demonstrates what happens when the digital and analytics teams aren’t empowered to own digital/analytics operations. Digital advisors must be in a position to run everything that is digital because they understand the necessity of field testing, simulated demand spikes on websites, and hitting milestones on product development, etc. A digital team would never have allowed ORCA to happen, but the fundamental mistake of the Romney campaign was to have an excellent digital team available but empowering the political office to run a substantial digital project on their own.

Obama Campaign 2012: As noted, unlike the Romney campaign, Obama-Biden 2012 had a four-year head start and nearly unlimited resources. They put their advantages to good use.

A. Data driven Obama’s 2012 campaign manager, Jim Messina said, “We were going to put any analytics team inside of us to study us the entire time to make sure we were being smart about things.” This is akin to an Accenture or a McKinsey-based approach. They would use analytics to improve field teams, websites, e-mails and other communications. There was almost no aspect of the campaign that wasn’t evaluated analytically based upon data.

“The core of the campaign was not flashy or even particularly innovative except in the willingness of senior staff to listen to numbers people rather than consultants acting on old-fashioned political intuition.”

According to CMAG/Kantar, the Obama campaign’s cost per ad was lower ($594) than the Romney campaign ($666) or any other major buyer in the campaign cycle. That 10% advantage was enormous with 550,000 advertisements.

B. Digital Grassroots As David Plouffe describes the 2008 campaign, “Technology, like the grassroots focus, would be at the core of our campaign from the start. In order to build a grassroots movement, it was clear that the only way to get to scale quickly enough was to use the power of the Internet to sign people up and ask them to get involved. . . So many people are living their lives through technology – how can we expect their interaction with politics to be the one exception?”

Obama assembled a massive email list, but in order to bulk up their phone number list, they set up a program to notify people by text as to their VP pick. This “be the first to know” program, encouraged supporters to sign up with their cellphone to receive a text message as soon as the news broke. This program was extremely successful in increasing the phone list.

By 2012, Obama assembled a list of 4.4 million donors and a 16 million person email list, in comparison to 1.1 million donors and a 2-3 million record email list for Romney.

Conclusion:

In 1999 high stakes politics changed forever when 2,000 volunteers delivered petitions signed by thousands of Americans to members of the House of Representatives in 219 districts across America, in a campaign to stop the impeachment of President Clinton. What made this campaign novel was its use of a website leveraging an email universe. Moveon.org’s campaign was so successful that they remain a potent political force today. In 2012, a more advanced online campaign derailed legislation that was likely to pass, SOPA/PIPA. With over 10 million phone calls to Congress and shutting down popular websites in protest, the effort killed the momentum on Capitol Hill. And in 2013-2014 the first online petition directly resulted in legislation passing Congress.

Online movements can also generate a tremendous amount of money, as was seen with Ron Paul’s successful money bombs where he raised unprecedented grassroots dollars in 24 hours. President Obama’s 2008 and 2012 campaigns were able to leverage digital fundraising to create new levels of low dollar supporters donating $5, $10 or $50, often as first-time donors. Previous outreach to such low dollar supporters would have been cost-prohibitive and relied heavily on direct mail or phone call solicitations. With digital technology the Obama campaign was able to raise large amounts of money with relatively low per-acquisition costs.

The 2016 political and cause cycle may spend well over $5 billion. With the realization of declining returns for 30 and 60-second radio and television ad buys, digital is likely to continue to grow as a percentage of total allocations. This is ultimately a good thing for campaigns, as well-run digital efforts have feedback loops to identify what is working, constantly evolving to improve and reach people where they are.