People join campaigns for different reasons: electing a leader they believe in, advancing an agenda, cleaning up government, or experiencing the rush and adrenaline of campaign life. These are some of the reasons we got involved in politics. We certainly didn’t sign up because we wanted to become cyber experts and we’re guessing you didn’t either.

We come from different political parties and don’t agree on much when it comes to public policy, but one thing uniting us is the belief that American voters should decide our elections and no one else. Our increasingly digital way of living and working offers new ways for adversaries to influence our campaigns and elections. While you don’t need to be a cyber expert to run a successful campaign, you do have a responsibility to protect your candidate and organization from adversaries in the digital space. That’s why Defending Digital Democracy, a project of Harvard Kennedy School’s Belfer Center for Science and International Affairs, created this Cybersecurity Campaign Playbook [PDF].

The information assembled here is for any campaign in any party. It was designed to give you simple, actionable information that will make your campaign’s information more secure from adversaries trying to attack your organization—and our democracy.

Good luck.

P.S. Do you see a way to make the Playbook better? Are there new technologies or vulnerabilities we should address? We want your feedback. Please share your ideas, stories, and comments on Twitter @d3p using the hashtag #CyberPlaybook or email us at connect@d3p.org so we can continue to improve this resource as the digital environment changes.

Top Five Checklist

The Playbook Approach

A bipartisan team of experts in cybersecurity, politics and law wrote this Cybersecurity Campaign Playbook to provide simple, actionable ways of countering the growing cyber threat.

Cyber adversaries don’t discriminate. Campaigns at all levels—not just presidential campaigns—have been hacked. You should assume that you are a target. While the recommendations in this playbook apply universally, it is primarily intended for campaigns that do not have the resources to hire full-time, professional cybersecurity staff. We offer basic building blocks to a cybersecurity risk mitigation strategy that people without technical training can implement (although we include some suggestions that will require the help of an IT professional).

These are baseline recommendations, not a comprehensive reference to achieve the highest level of security possible. We encourage all campaigns to enlist professional input from credentialed IT and cybersecurity professionals whenever possible.

Candidates and campaigns face a daunting array of challenges. There are events to organize, volunteers to recruit, funds to raise, and the relentless demands of the modern media cycle. Every staffer must anticipate unfortunate surprises like gaffes or a last-minute attack ad. Cyber attacks now belong on this list as well.

As campaigns have become increasingly digital, adversaries have found new opportunities to meddle, disrupt, and steal. In 2008, Chinese hackers infiltrated the Obama and McCain campaigns, and stole large quantities of information from both. In 2012, the Obama and Romney campaigns each faced hacking attempts against their networks and websites. In 2016, cyber operatives believed to be sponsored by Russia stole and leaked tens of thousands of emails and documents from Democratic campaign staff.

The consequences of a cyber breach can be substantial. News of a breach itself, compounded by a slow-drip release of stolen information, can derail a candidate’s message for months. Attackers overloading a website can lead to lost donations at key moments. The theft of personal donor data can generate significant legal liabilities and make donors reluctant to contribute to a campaign. Destructive attacks aimed at staff computers or critical campaign servers can slow down campaign operations for days or even weeks. Cleaning up the resulting mess will divert precious resources in the heat of a close race, whether it’s for president or city council.

For the foreseeable future, cyber threats will remain a real part of our campaign process. As democracy’s front line, campaign staff must recognize the risk of an attack, develop a strategy to reduce that risk as much as possible, and implement response strategies for that moment when the worst happens. While no campaign can achieve perfect security, taking a few simple steps can make it much harder for malicious actors to do harm. Ironically, the most sophisticated state actors often choose the least sophisticated methods of attack, preying on people and organizations who neglect basic security protocols. That is our primary reason for creating this Cybersecurity Campaign Playbook.

In today’s campaigns, cybersecurity is everyone’s responsibility. Human error has consistently been the root cause of publicized cyber attacks, and it’s up to the candidate and campaign leaders to weave security awareness into the culture of the organization. The decisions humans make are just as important as the software they use. Going forward, the best campaigns will have clear standards for hard work, staying on message, being loyal to the team—and following good security protocol.

Before we get into our recommendations, let’s quickly frame the problem:

• the environment in which your campaign is operating;
• the threats your campaign will likely face; and,
• the importance of cyber risk management.

The Threats Campaigns Face

Unfortunately for campaigns and our country, foreign adversaries may think that harming or helping a particular candidate advances their national interest, whether that means creating chaos and confusion among American voters, or punishing an official who has spoken out against them. This may sound like thriller fiction, but the reality is that a sophisticated foreign intelligence service, cybercriminal or hacktivist with a grudge against a candidate, could decide that you or someone on your campaign is a target.

These are the sorts of threats managers and staffers have to realize are possible.

Risk is best understood in three parts. First, there are vulnerabilities: weaknesses in your campaign that make information susceptible to theft, alteration, or destruction. Vulnerabilities can originate in hardware, software, processes, and in the vigilance level of your staff. Then there are actual threats: the nation-states, hacktivists, and other nonstate groups with the capability to exploit those vulnerabilities. Risk exists where vulnerabilities and threats meet. Lastly, there are consequences—the impact when malicious actors capitalize on unmitigated risk.

There’s little you or your campaign can do to prevent threats themselves—they are the result of larger geopolitical, economic, and social forces. What you can do is substantially reduce the likelihood that your adversaries will succeed by reducing your own vulnerability. Reducing vulnerability reduces risk—it’s up to you to decide which ones are most essential to address based on the possible consequences. For example, you may decide that the most damaging thing a hacker could do is to steal your candidate’s self research report. In response, you devote extra resources for secure cloud-based storage, use two-factor authentication, and restrict access to a small number of people. You may decide to make other documents on the campaign more widely available and less secure, since more people need them to do their jobs and they wouldn’t cause much damage if they were leaked.

There are technical aspects to risk mitigation, but what matters most is that you take a holistic approach. As a campaign leader, you must make fundamental choices, such as who has access to information, what information is kept or discarded, how much time you devote to security training, and how you behave as a role model. As a campaign professional, risk management is your responsibility—both technical and human. It’s up to you to decide what data and systems are most valuable and what resources you commit to protect them.

Our security recommendations are organized according to three principles:

1. Prepare: The success of nearly every one of the Playbook’s recommendations depends on the campaign manager creating a culture of security vigilance that minimizes weak links. That means establishing clear ground rules that are enforced from the top down and are embraced from the bottom up.

2. Protect: Protection is critical. When you discover you have a security problem, it is already too late. Building the strongest defenses that time and money allow is key to reducing risk. Internet and data security works best in layers: there is no single, bulletproof technology or product. A few basic measures used in combination can make a campaign’s digital architecture more difficult to breach and more resilient if compromised.

3. Persist: Campaigns now face adversaries with ever-increasing levels of resources and expertise; even the most vigilant culture and the toughest infrastructure may not prevent a security breach. Campaigns need to develop a plan ahead of time to deal with a breach if one occurs.

Some campaigns have more time and money for cybersecurity than others. That’s why our recommendations offer two tiers of protection: “good” and “enhanced.” The “good” tier represents everything a campaign must do to have a minimum level of security. Using the “good” recommendations in a piecemeal fashion will leave you vulnerable. You should always aspire to do more as time, money, and people allow, which is why we recommend using the “enhanced” level whenever possible. If you have the resources to get reputable, trained IT support, it’s money well spent. Threats are constantly evolving and professional IT services will help get you beyond what this playbook provides and keep you abreast of the latest threats and solutions.

Management
Campaign managers need to take responsibility for their cybersecurity strategy, but most will delegate development and supervision to a deputy or operations director. It’s important that cybersecurity is tightly integrated into HR and IT work, since correctly onboarding staff, provisioning hardware, and controlling permissions will be critical to your strategy. Many small campaigns will rely on volunteer support for IT and cybersecurity. You can use this playbook to guide your discussion with your volunteer support. The key is to carefully vet the volunteers who support you and carefully control access, so that volunteer support doesn’t create new vulnerabilities. You should make sure a campaign staffer is supervising IT work and controlling permission to access different systems. 

When To Start
Whatever support model you have, cybersecurity should start on Day One. What follows is a “top five checklist” of measures that are absolutely vital. Make sure these are in place at the very beginning, even if there are just one or two staff, then complete the other “good” recommendations as soon as possible.    

Cost

A lot of what we recommend here is free or very low cost. In fact, everything on our top five list is free, except getting a cloud-based platform, which will only cost a few dollars per month per employee. High target campaigns will need to budget enough resources for hardware and software to execute a responsible strategy, but this should still be a very small percentage of a multi-million dollar statewide campaign budget. Smaller campaigns will be able to execute the recommendations here for a few hundred to a few thousand dollars depending on how many staff or volunteers work on the campaign.

Any references to vendors and products are intended to help provide examples of common solutions, but do not constitute endorsements. If challenges arise when implementing products or services, we encourage you to reach out directly to the vendors, who can usually provide user-level technical assistance. When it comes to product and service selection, we encourage every campaign to consult with a cybersecurity expert or conduct independent research to find the best product for their needs.

The Vulnerable Campaign Environment

Today’s campaigns are uniquely soft targets. They’re inherently temporary and transient. They don’t have the time or money to develop long-term, well-tested security strategies. Large numbers of new staff are often onboarded quickly without much time for training. They may bring their own hardware from home and the malware lurking on it.  Events move quickly, the stakes are high, and people feel that they don’t have time to care about cybersecurity. There are a lot of opportunities for something to go wrong.

At the same time, campaigns rely more and more on proprietary information about voters, donors, and public opinion. They also store sensitive documents like opposition research, vulnerability studies, personnel vetting documents, first-draft policy papers, and emails on various servers. The risks of a potential attack are increasing and so are the consequences. .

Cybersecurity is fundamentally a human problem, not a technical one. The best technological solutions in the world will have no effect if they are not implemented properly, or if they are not continuously updated as technology evolves. Successful cybersecurity practices depend on creating a culture of security awareness.

GOOD: What You Need to Do

1. Establish a strong information security culture that emphasizes security as a standard for a winning campaign. Just as campaign staffers are instructed not to take an illegal donation, employees should know to avoid clicking on links or opening attachments in emails from unknown senders.
   • Onboarding: Provide basic information security training when you onboard new staff. You can distribute the Staff Handout at your training.
   • Trainings: Make security part of all your ongoing staff trainings, such as senior staff retreats or GOTV trainings. Provide additional training for those in sensitive roles, such as the candidate, press staff, senior staff, and anyone with system administrator privileges on your network. Managers should require that the most important people in the campaign—including the candidate—have their security settings checked by whoever runs IT (that may be the manager herself).
   • Set the example: Senior campaign staff and the candidate must take a visible leadership role, advocating for cybersecurity during trainings. Senior staff should provide periodic reinforcement of cybersecurity’s importance to junior staff in meetings and on calls. Don’t just have technical experts conduct trainings. The campaign manager or operations director can be a more powerful messenger precisely because they’re seen as less “technical.”

2. Train and educate your staff to be on the lookout for phishing.  Phishing attacks against campaigns are on the rise and continue to be a primary method used by malicious actors.  Train your staff to be suspicious of any email asking for information, or claiming they need to click a link to reset their credentials. Sophisticated phishing may attacks may come from spoofed or compromised senders that appear legitimate. Encourage staffers to share anything suspicious with you or your IT staff.  The more people share, the more confident you can be that they’re being vigilant and the more intelligence you will have. The overall rule of thumb should be to “think before you click”, but we’ve included three key points you can remind you staff about when you train and re-train.  Training should emphasize best practices such as ‘hovering’ over a link to identify the actual url, expanding email details to confirm the email address of the sender, and using a different communications channel such as a quick phone call to confirm authenticity of the sender and email contents 2FA is another important way to prevent a spear-phishing attack from leading to an account compromise—just having your username and password will not be enough to access your account. As part of the campaign’s strong security culture, senior staff should recognize and praise anyone who reports suspicious behavior on their system or admits to clicking a potentially malicious link.

   • Phishing can happen on the phone, too!  Staff should never share information, wire money, or give anything else away on the phone if they aren’t certain who the caller is.  Make staff aware of the threat and tell train them to listen to how the caller greets them and to ask questions that outsiders may not be able to answer.  You can easily test your staff on phishing calls—and your friends will enjoy doing it!

3. Conduct a thorough vetting of staff, volunteers, and interns—anyone requesting access to campaign information—to avoid giving credentials to someone who wants to steal or sabotage your systems. Establish a definition for sensitive information and rules for its use. For example, you could choose to classify all polls, research materials, strategy memos, and related emails as “sensitive.” Prohibit the transfer of sensitive information on communication channels that aren’t managed and secured by the campaign. You can require that it be transferred only through encrypted messaging (see Step 2).
4. Confirm that consultants and vendors with access to sensitive information have secure email and storage (see Step 2). When in doubt, require vendors and consultants to use an account on your cloud-based office suite (See Step 2).
5. Control access to important online services, such as the official campaign social media accounts, to prevent use by unauthorized individuals. Make sure that those who leave the campaign can no longer access campaign-related accounts. You can do this easily by using a social media account management tool that acts as a gateway to all your accounts. If someone leaves the campaign, you should immediately disable their account.

ENHANCED: Take the Next Step

1. Software products such as Phishme and KnowBe4 can train your staff by sending them fake phishing emails. This is a safe, quick, and effective way to learn who is at risk of clicking a link, so you can give them counseling and extra training. Many of these products also filter some phishing attempts out of your email.
2. If you have the resources, hire a dedicated IT professional to manage your campaign’s systems and an IT security expert to help protect, maintain, and monitor your campaign’s digital infrastructure. He or she can provide regular security training and testing of your people and systems, while customizing security solutions.
3. Contract with a cybersecurity firm to provide security solutions, review your defenses, and/or monitor your systems for a breach. Know which firm you want to contact if you are breached and need urgent incident response support. This is an alternative to hiring a full-time IT security expert. Do your research and go with a highly reputable, U.S.-based firm—not all cybersecurity firms provide the same level of service.

Not all methods of communication are equally secure, so use the most secure method possible. Campaign leadership should set a standard that encourages in-person conversations whenever possible, and discourages needless or superfluous emails. Whether it is phone calls, texting, or emailing, different products and services offer different levels of protection, so do your research before you choose which systems your campaign is going to use.

GOOD: What You Need to Do

1. Use a cloud-based office suite that provides secure email communication, document creation, chat, and file sharing, such as GSuite or Microsoft365. For GSuite Enterprise, choose “Advanced Protection” and for Office365, use E5.  Both will have built in settings that are more secure.For example, GSuite includes Google Drive for file sharing, Gmail for email hosting, Google Hangouts for chat, and Google Docs for word processing, spreadsheets, and presentations. Microsoft365 offers OneDrive/SharePoint for file sharing, Outlook/Exchange for email, Microsoft Teams for chat, and Microsoft Office for word processing, spreadsheets, and presentations. Cloud-based systems managed by major firms will be better protected than any servers you could set up in your campaign. There are free versions of both products, but the paid versions give you many more administrative capabilities. Google’s Advanced Protection Program provides extra security against targeted online attacks like phishing (this is available for their consumer Google/Gmail accounts). Through Google’s Protect Your Election effort, they also offer a free service to protect your website against disabling attacks.
2. Use the most secure systems possible for communication.
   • Use encrypted messaging services such as Signal, Wickr, especially for messages, document sharing and phone calls. Many campaigns require that sensitive information only be transmitted by encrypted messaging, although you can use it for all communication if you want (this is especially smart for high-risk individuals like the candidate). Signal and Wickr allow you to auto-delete messages, which reduces risk.
   • Switch off archiving for messaging services, such as Google Chat and Slack, so that old chats can’t be stolen later. This requires going into “settings” and adjusting “retention policy” timelines. Some services require you to do this for every single chat conversation. We recommend retaining chat messages for one week or less.
3. Defend your email
   • Turn on Auto-delete in your email application for old emails to reduce the number of emails that could potentially be stolen. This usually requires going in and changing “retention policy” to shorter time periods in “settings.” To ensure emails do not just sit in a “deleted items” folder, adjust settings to auto purge “deleted items” folder after a certain time period. We recommend retaining emails for one month or less.
4. Secure personal accounts
   • Campaign business should never go on personal accounts. However, adversaries will target personal accounts for hacking, so have your staff use strong passwords and two-factor for their personal accounts as well (this is included in our Staff Handout).

One of the most challenging aspects of security is keeping unauthorized people out. This means preventing adversaries from gaining access to your data and preventing people within your campaign from having access to information they do not need. While some of the recommendations below may seem cumbersome, hackers depend on those who value convenience over security.

GOOD: What You Need to Do

1. Require two-factor authentication (2FA) on all systems and applications. Avoid texting (SMS) for two-factor authentication, because attackers can easily clone a phone number and get access to texts. There are several 2FA apps that work just as well as texting, such as Google Authenticator, Microsoft Authenticator, and Duo Mobile. You can also use a physical FIDO (“fast identity online”) key that is inserted into your USB drive such as Yubikey or Feitian. As malicious actors evolve their methods, we are seeing advanced attack techniques used to phish second factor codes sent to users. We strongly encourage the use of physical security keys for two-factor authentication, as they are not vulnerable to phishing. The website “TwoFactorAuth.org” is a helpful guide to services that do and do not offer 2FA.
2. Passwords.
   • Ensure no systems are using default usernames or passwords. This is a commonly overlooked error. Make sure each system and user has their own unique username and password.
   • Require strong passwords. As we noted earlier, “make passwords that are long and strong.” Current computing capabilities can crack a seven-character password in milliseconds. A 20-, or even 30-character password will take much longer for a hacker to crack. Choose a string of words that you can easily remember.
   • Use a different password for different accounts so a hacker can’t break into multiple accounts if a single password is stolen.
   • If someone reaches out requesting a password or password reset, require the request to be made in person or over a video chat to ensure it is the actual campaign staff member or volunteer. Only share passwords in person or over short-lived encrypted messages. Never share passwords over email or store/distribute using a helpdesk system.
3. Use password managers such as LastPass, 1Password, or Dashlane to help you manage a lot of long, strong passwords easily. But ensure that your management system has a long, strong password and two-factor authentication. We don’t currently recommend password managers built into browsers, which are often less secure than these standalone managers.
4. Create separate accounts for administrators and users, and severely restrict access to administrator accounts. Administrators should also have two separate campaign accounts—one used only for their admin duties and one that is their standard user account for all other campaign business. This will reduce the likelihood that an adversary will be able to compromise an administrator account, which would provide access to the entire network.
5. Conduct periodic reviews of who has access to different devices and networks. Immediately block access of people who leave the campaign. Immediately change passwords if suspicious activity is observed.
6. Use encrypted messaging for sensitive information. The federal government has confirmed the presence of “stingray” machines that can intercept cellular data, especially in the Washington, DC area.  This is yet another reason to keep communications on encrypted messaging apps, which will be unreadable even if your signal is intercepted.
7. Monitor all campaign, staff, and related social media accounts. It is important to identify and respond quickly if you have been hacked.In line with the “Prepare, Protect, Persist” framework, campaign leadership should create a response plan for this scenario. This response plan should should include finding appropriate points of contact within each social media company to get in touch with in the event that campaign accounts are hacked.

ENHANCED: Take the Next Step

1. Create user profiles for different types of campaign staff that automatically grant the necessary level of access. Different types of employees—interns, field staff, campaign leadership—require access to different resources. Having predetermined profiles makes it easier to ensure that people are getting access only to what they need.

Every physical device in your campaign—from a cell phone, tablet, or laptop to a router, printer, or camera—represents a potential attack path into your network. A good cybersecurity plan will attempt to control access to, into, and on all devices. You can control access to devices by making sure they are always properly handled and accounted for. You control access into devices via two-factor authentication and strong passwords. You control the content on devices via encryption and the policies guiding how you store data (i.e., storing information in the cloud instead of on machines).

GOOD: What You Need to Do

1. Always use the most updated operating system (OS) available, since system updates regularly include patches for the latest vulnerabilities. If possible, set device settings to auto-install these updates. Make it someone’s job to check on a regular basis that everyone is current.
2. Use an automatic cloud-based backup service to mitigate the impact of data loss if a device is lost or stolen. Examples include Backblaze and CrashPlan.
3. Physical access to the device
   • From the start, campaign leadership should create an environment in which people take physical security of their devices seriously—losing a device could give an adversary access to critical information that can be used to hurt the campaign.
   • Although many campaigns cannot afford to buy new devices, it’s always best to purchase new equipment (especially computers and phones) if you can. At a minimum, you should provide new devices for personnel who work with sensitive data.
   • If staff are using their own computers and phones, establish a “Bring Your Own Device” (BYOD) policy that implements strong security practices (see endpoint protection below).
   • Campaign members should NOT use personal email accounts or devices that have not been secured per the BYOD policy for campaign business, including email and social media. Any important information that resides outside devices or systems controlled by the campaign is vulnerable to attack. Leadership should constantly reinforce that campaign data needs to stay off personal email and unsecured computers.
   • Report lost devices immediately. Require default settings that allow for remote wiping on all devices.
   • Win or lose, have a plan in place for what happens to all data, accounts, and devices when the campaign ends. This includes thinking how to safeguard or erase data at the hardware level. Reformatting a hard drive is not enough to protect your data. Residual data on reformatted or disposed drives can be obtained using commercially available forensics software. Extremely sensitive data should be degaussed or melted  The immediate aftermath of a campaign is an especially vulnerable period.
4. Digital access into devices
   • Change default passwords and settings on all devices. Many devices come from the factory with a default password that is really easy to guess. Also, disable the guest account if a device comes with one.
   • Implement auto-lock for phones and computers after two minutes and require a password or fingerprint ID to unlock.
5. Content on devices
   • Require encryption on all devices (computers and phones) to ensure that the loss of a device does not mean the compromise of its content. Examples include FileVault for Mac and BitLocker for Windows. Some devices like the iPhone do this by default, but not all do. Require all consultants to keep data on their machines encrypted as well.
   • Install endpoint protection software on all devices. Some examples include Trend Micro, Sophos, and Windows Defender. There are special endpoint security apps for phones and tablets. Lookout is an example.
   • Limit what apps can access on all devices. This means limiting what apps are installed and limiting permissions for those apps (e.g., limiting access to contact lists or location and GPS information, turning off ‘always active’ mode).

ENHANCED: Take the Next Step

1. Use mobile device management (MDM) software, which monitors activity to ensure all devices comply with the mobile phone and user device security policies you have established for your campaign. Examples include VMware AirWatch, Microsoft InTune, and JAMF. GSuite and Microsoft Office 365 also include an MDM service.
2. Use advanced threat protection services that monitor and alert for malicious activity, such as CrowdStrike Falcon or Mandiant FireEye. CrowdStrike sometimes offers Falcon breach prevention service pro bono through the Crowdstrike Foundation, depending on the needs of your campaign and campaign finance rules.

Networks are the system of physical hardware, digital software, and their connections. They represent another target-rich environment for attack. Network security comprises everything from how devices communicate with one another to using cloud services for data storage.

GOOD: What You Need to Do

1. Embrace the cloud. Store data on cloud services, not on personal computers or servers. Anything stored on a personal device faces higher risk than the cloud.
   • No one should have access to all files on the network; accounts with comprehensive administrator access should not be used for day-to-day work. Divide your file storage into department folders and grant access accordingly.
   • Ensure access to shared content is by invitation only. Some file management services also allow for implementing expiration dates on invitations and access.
   • Periodically audit what is being shared and with whom.
2. Have a separate “guest” wifi network for visitors and volunteers that limits their access to campaign resources. Try to purchase routers that offer a “guest profile” that will automatically segment your network.
3. When traveling, or before you set up your campaign office, avoid public wifi services as much as possible and use trusted wifi networks wherever possible. If you need mobile wifi, then try to provide campaign staffers with mobile wifi hotspots for tethering. Public wifi is often free and easy to connect with, but attackers can also use it to penetrate your hardware.
   • Where possible, staffers should use a VPN (virtual private network). VPNs help protect against intruders when on public wifi. Examples of VPN services include ExpressVPN or TunnelBear. Not all VPNs are created equal. Beware of free services: many are looking to take your data!
4. Secure your browser. PC Magazine ranked Chrome and Firefox as the two safest browsers in 2017. Regardless of what browser you use, keep it up to date.

ENHANCED: Take the Next Step

1. You can take more advanced steps to protect your network, but they should be implemented by an IT professional. We would suggest you ask them to include the following:
   • Set up a hardware firewall.
   • Encrypt your wifi connection using the WPA2 or 802.1x security protocols (do not use WEP).
   • Configure cloud-based web proxies to block access to suspicious sites from any campaign-owned device, no matter where it is. Service provider examples include Zscaler, Cisco Umbrella and McAfee Web Gateway Cloud Service.
     Have your activity logs stored on a cloud service provider such as LogEntries or SumoLogic.
   • Segment your cloud-based storage so that not everything is stored in the same place. Opposition research, strategy memos, and personnel files should be kept in different folders, and access to those folders should be restricted to the people who really need them. Consider a different storage system entirely for your campaign’s most sensitive information. Restrict access so that only key personnel can access it, and only when using specific devices. (For example, if you use Microsoft365 for your office suite and document storage, but your most sensitive documents on a Dropbox or Box account.) If a member of the campaign becomes compromised, this kind of segmentation can limit the damage.
2. Train staff not to connect their devices to unknown ports or devices.  Don’t use public chargers at airports or events. Don’t accept free phone chargers or batteries at events (that free USB drive may be loaded with malware!).  

Information operations have been in the news a lot recently, especially campaigns run by foreign intelligence services to influence opinion in the United States.  It will be up to elected leaders and policymakers to decide how to confront information operations moving forward and there’s little we can do as campaign staff to impact whether they happen or not, but there are a few things we can do to manage them if they’re happening.  Campaigns have and will continue to be targets of these operations and need to be prepared.  Defending ways your campaign communicates with the public is an important part of this. Below are some ways to better protect against information operations, identify when they are happening to your campaign or candidate, and how to respond quickly when they do occur.

GOOD: What You Need to Do

1. Remember: information operations are a communications problem, not a technical one.  Adversaries can make their information operations more potent by stealing your data, but once information is out in the environment, you need a communications strategy to manage it.  Think in advance how to handle fake or slanted news—will you ignore it?  Re-tweet it and reinforce that it’s false?  How will you make this decision?  These are among the most difficult decisions any campaign has to make, but what matters most is think about these questions with your team in advance, so you and your team have guidance about how to respond, if you respond at all.  
2. Know what’s going on. Encourage activists to share posts, sites, or news stories they find suspicious.  If you want, you can deputize some interns or volunteers to focus on this specifically, conducting searches to find out what content is out there.  One ongoing challenge is that it’s impossible to see everything that voters may be getting on their Facebook feeds.  The platform has made it harder to post political advertising and has increased staff to monitor news content, but you cannot search all content.  The best way to solve this right now is to deputize a team of volunteers, who represent different geographies and demographic groups in your state/district, so you can catch as much as possible.  
3. Establish contact with key social media platforms and notify them if you find fake or misleading information. Most social media platforms will now remove “fake” or misleading content and imposter profiles.  Ask your relevant campaign committee or state party for the best contact at social media platforms and establish contact early in the campaign so you can reach out quickly if something goes wrong.
   • Facebook
   • Twitter
   • Google/YouTube
4. Monitor for imposter sites. To-date, there are no public reports of imposters trying to steal money or activist data through fake websites, but it’s such an easy vector of attack, you should be on the lookout.  Make sure to purchase any web addresses you may want to use (or could be used against you). If you want, you can retain a reputation management service that will monitor the web for you. Some can do this at a fairly modest price.  
5. Protect Against a Distributed Denial of Service Attack (known as DDoS). A DDoS attack is when an adversary takes control of a lot of machines, and uses them to “ping” your website all at once, causing it to crash.  Most of what we focus on in this guide is how to keep people away from your campaign data, but, in the case of a DDoS, you want to keep your website open and available all the time for donors and activists.  DDoS has not yet become a common threat to campaigns, but it could be used to block you from fundraising or simply cause a really frustrating disruption to your campaign.  There are two free tools you can use to protect your site, Google’s Project Shield and Cloudflare.

It’s just as important to plan for responding to an attack as it is to develop a security strategy to prevent one. How you respond often has more to do with the ultimate outcome of an incident than what was compromised. You should budget some time at strategic retreats or longer senior staff meetings to discuss what will happen if something does go wrong. Here’s a checklist of the steps you should take:

LEGAL

✔ Identify outside counsel you will retain in the event of a cyber incident, and discuss the response process with them at the outset of the campaign. In most cases, this will be the same person who represents your campaign on other matters, but ideally you would have someone who specializes in incident response on call, either pro bono or for a $0 retainer.

✔ Ask your lawyer to explain your legal obligations if data is stolen and what compliance measures you will need to have in place.

✔ Understand your vendors’ legal obligations to notify you or others if they are hacked. Wherever possible, include strict notification requirements in your vendor contracts, since third parties are a frequent source of breaches.

✔ If you believe you’ve been breached, a best practice is for your lawyer to oversee your response under attorney-client privilege.

✔ Talk to your lawyer about the best way to work with law enforcement if a breach occurs.  Every campaign will approach this differently.  

TECHNICAL:

✔ Determine ahead of time whom you will call for technical assistance if you think you’ve been hacked. Your state caucus or national party committee can usually provide referrals.  

✔ Choose someone on the campaign who will interface with technical experts in the event of a breach. This is ideally the same person who is already coordinating IT for the campaign. Managing an incident response can be overwhelming, so you want someone focused on the technical aspects who knows what they are doing. That way you can focus on communicating with stakeholders and the press.  

OPERATIONS:

✔ Decide in advance who will be on your Incident Response Team (IRT) and who will participate in incident response meetings. It’s important to include someone from your IT, legal, operations, and communications teams. If you’re a small campaign and don’t have full-time communications, IT, or operations support, plan to include any key staff who oversee campaign operations.   

✔ Determine the chain of command for decision-making in the event of a breach, especially regarding communications. In many cases, this will be the campaign manager, but some managers may choose to delegate responsibility to someone else.  

✔ Identify what app or technology you will use to communicate if you think your email has been breached (Signal and Wickr are two common options). Communication during a breach is essential, but you don’t want your adversaries to know what you’re saying—or even that you are responding to their actions.    

COMMUNICATIONS:

✔ Conduct scenario planning. For many campaigns, this can be part of an existing strategy retreat. For bigger campaigns at higher risk, it may be necessary to have a dedicated meeting. Your scenario planning should include:

✔ Identifying key internal and external stakeholders, like your staff, volunteers, donors, and supporters. Know whom you need to contact if an incident occurs and rank them in order of priority. Develop a contact list and designate who will reach out to them.

✔ Brainstorm the most damaging scenarios and consider how your stakeholders and messaging may change for each one.  Different scenarios could include:

• Rumors that your campaign has been hacked;
• Credit card and contact information for your donors is stolen;
• Ransomware and an extortion attempt are lodged against your campaign;
• Your systems are wiped and shut down;
• Someone’s emails are stolen;
• Your adversary steals your administrator’s credentials and every file on your campaign drive.
•  A malicious actor alters statements on your website or public accounts.

✔Be careful what you say in the present about cyber security policy or cyber incidents. Some victims of cyber crimes have previously made grandiose pronouncements about their own security measures, or have criticized others who have been attacked. The press will hold you accountable for what you said in the past if you fall victim.

✔ Similarly, avoid providing details about the scope of the event in the early phases of the incident (and if you can avoid discussing the scope altogether, even better). Details available at the outset will change as you investigate. A common mistake is to say something that later turns out not to be true (e.g., “they didn’t steal very much,” or “no personal information was taken”). Saying only what you know for sure is the safest course. Statements should focus on the actions you are taking to make the situation right for the affected stakeholders.

✔ Develop some boilerplate language iin advance, so that you can draft statements or talking points quickly if an incident occurs. At a minimum, create a simple Q & A document that you can rapidly revise if you actually need to use it. Creating a Q & A document in advance will help you to think as much about what you won’t say as what you will say. For example, the first question will often be, “What happened?” However, you may not be able to answer that for days or weeks. The fact that you don’t know what kind of breach will take place can actually help you write better boilerplate answers in advance.

• Questions to include in your  Q & A document are:What happened?
• How did it happen?
• Who did it?
• What was stolen or damaged?
• Was anyone’s personal information stolen?  What are you doing to protect them?
• How did the hackers do it?
• Are the hackers out of your system?
• How long were they in your system?
• What security measures did you have in place?  Why weren’t they effective?
• Shouldn’t you have known this would happen?  Why weren’t your systems better secured?
• Are you working with law enforcement? Has law enforcement contacted you?
• In a ransomware breach, you’ll be asked: Did you pay the ransom and why or why not?

✔ Stay in touch with your key stakeholders and keep them as informed as you can. You probably won’t be able to say much, but contacting them regularly with what you do know, having a clear statement about your intentions, and providing details about what you are doing to manage the situation are key. Avoid setting an expectation of too frequent updates, because often you won’t have new information and your stakeholders will become frustrated if you continue to return to them without new information. Only speak proactively to the media if you have new information to provide.

This project was made possible by dozens of people who generously volunteered their time.  
Special thanks are due to Debora Plunkett for leading the project and Harrison Monsky for writing the document. We are also indebted to the people listed below who invested countless hours in reviewing drafts and providing input.

Defending Digital Democracy
Eric Rosenbach, Co-Director, Harvard Kennedy School Belfer Center
Robby Mook, Belfer Center Fellow
Matt Rhoades, Belfer Center Fellow
Heather Adkins, Director, Information Security and Privacy, Google
Dmitri Alperovitch, Co-Founder and CTO, CrowdStrike
Josh Burek, Director of Global Communications and Strategy, Harvard Kennedy School Belfer Center
Chris Collins, Co-Founder, First Atlantic Capital
Caitlin Conley, student, Harvard Kennedy School
Mari Dugas, Project Coordinator, Defending Digital Democracy, Harvard Kennedy School Belfer Center
Josh Feinblum, student, Massachusetts Institute of Technology
Siobhan Gorman, Director, Brunswick Group
Stuart Holliday, CEO, Meridian International Center
Dai Lin, student, Harvard Kennedy School
Kent Lucken, Managing Director, Citibank
Katherine Mansted, student, Harvard Kennedy School
Nicco Mele, Director Harvard Kennedy School, Shorenstein Center
Debora Plunkett, former Director of Information Assurance, National Security Agency
Jim Routh, Chief Security Officer, Aetna
Suzanne E. Spaulding, Senior Adviser for Homeland Security, Center for Strategic and International Studies
Matthew Spector, student, Harvard Kennedy School
Alex Stamos, Chief Security Officer, Facebook
Phil Venables, Partner and Chief Operational Risk Officer, Goldman Sachs

Additional Authors and Contributors
Ryan Borkenhagen, IT Director, Democratic Senatorial Campaign Committee
Michael Chenderlin, Chief Digital Officer, Definers Public Affairs
Robert Cohen, Cyber Threat Analyst, K2 Intelligence
Julia Cotrone, Special Assistant, Definers Public Affairs
John Flynn, Chief Information Security Officer, Uber
Daniel Griggs, Founder and CEO, cmdSecurity Inc.
Eben Kaplan, Principal Consultant, CrowdStrike
Greg Kesner, Principal, GDK Consulting
Ryan McGeehan, Member, R10N Security
Jude Meche, Chief Technology Officer, Democratic Senatorial Campaign Committee
Eric Metzger, Founding Partner and Managing Director, cmdSecurity Inc.
Zac Moffatt, CEO, Targeted Victory
Harrison Monsky, student, Harvard Law School
Colin Reed, Senior Vice President, Definers Public Affairs
Jeff Stambolsky, Security Response Analyst, CrowdStrike
Frank White, Independent Communications Consultant
Sally White, student, Harvard University
Rob Witoff, Senior Security Manager, Google

Belfer Center Web and Design Team
Arielle Dworkin, Digital Communications Manager, Harvard Kennedy School Belfer Center
Andrew Facini, Publications and Design Coordinator, Harvard Kennedy School Belfer Center