Safety Norms: Search Engine Marketing and Paid Search

1Purpose, scope, and safety principles

These norms apply to anyone who plans, builds, launches, edits, or monitors paid search campaigns. That includes in-house marketers, agency staff, freelancers, analysts, and developers who touch ad accounts, landing pages, or conversion tracking. The scope covers keyword and bid strategy, ad copy, sitelinks and other extensions, landing page alignment, conversion measurement, audience lists, budget control, and the people and data connected to them.

The governing principle is that every paid search element is a public statement made with someone else's money and someone else's data. Ads can be captured, reviewed by platform teams, reported by competitors, and examined by regulators. Treat each campaign as if it will be read in full by a skeptical customer, a journalist, and a compliance reviewer.

• Mandatory: every factual claim in an ad must be supported by linked evidence before launch.
• Mandatory: ads must send users to a landing page that matches the promise made in the ad.
• Recommended: write a one-page campaign brief that states the objective, audience, claims, and budget before building anything.

2Roles, competence, and accountability

Name one accountable account owner for each ad account. The account owner approves budget changes, billing changes, new claims in regulated categories, and new tracking. Campaign managers draft and optimize campaigns. Legal or compliance reviewers approve claims, trademark use, and restricted-category advertising. Privacy or data protection leads approve tags, consent configuration, and audience uploads. Developers own landing pages and tag implementation. Where one person holds several roles, a second person must still review the high-risk changes.

Access should follow the least-privilege principle, meaning each person receives only the permissions their role needs. Competence is a precondition for access. Staff should understand the platform's advertising policies, the privacy rules that apply to their markets, and the organization's approval path before they receive admin rights. Local rules on who may sign contracts, process personal data, or provide advice in regulated fields are jurisdiction dependent and should be confirmed with the organization's legal function.

• Mandatory: only trained staff with documented authorization may change billing, payment methods, or account users.
• Mandatory: agencies and contractors must sign access terms that assign responsibility for data handling, confidentiality, and policy compliance. Counsel should confirm the wording.
• Recommended: complete platform policy and privacy training before receiving admin or editor access, and keep a record of completion.

3Course-specific hazards and risk assessment

Paid search hazards fall into several groups. Financial hazards include unauthorized spending, bids set far above the intended cost per click, broad match keywords that pull irrelevant traffic, and duplicate conversion tags that inflate reported returns and lead to poor budget decisions. Content hazards include false or unsubstantiated claims, outdated prices, comparative claims about competitors that cannot be proven, and ads in restricted categories such as health, financial products, gambling, alcohol, employment, or housing. Platforms and regulators may limit or prohibit these categories, and some require special certification or disclosure. Confirm current requirements for each category before launch.

Privacy hazards arise when tags collect personal data without a lawful basis or without consent where consent is required, when personal information is passed into URL parameters, and when customer lists are uploaded through unsecured spreadsheets or shared inboxes. Security hazards include phishing of ad account logins, shared passwords, and browser extensions that can read page content and account sessions. Intellectual property hazards include using logos, photographs, or copyrighted text without permission, and bidding on or using trademarks in ways that the owner or platform prohibits. Reputational hazards arise when ads appear beside offensive content, when a campaign is perceived as deceptive, or when a single misstep is amplified on social media.

Rate each campaign before launch using a simple likelihood and impact matrix scored as low, medium, or high. A new campaign in a restricted category, a campaign that bids on a competitor's trademark, or a campaign using a new tracking pixel is high risk until legal or privacy review is complete. A new landing page or new audience source is usually medium. Minor copy edits within already approved claims and bid adjustments within approved ranges are usually low. Record the rating and the reviewer's name in the campaign file.

• High risk: any regulated claim, sensitive audience targeting, new tracking technology, third-party trademark bidding, or spend above the approved budget limit.
• Medium risk: new landing page, new extension content, new customer list source, or a new market.
• Low risk: routine copy edits within approved claims and bid changes within set limits.

4Preventive controls and safe work practices

Preventive controls work before money is spent or personal data is collected. Maintain a claim register that links each factual statement in an ad, extension, or landing page to evidence such as a product specification sheet, approved pricing record, certified test result, or legal statement. When the evidence expires or changes, the claim must be withdrawn or updated before the ad continues to run. Campaign launches should follow two-person review: one person builds and a second person checks claims, targeting, final URLs, tracking, budget, and naming. Platform preview and policy check tools are useful, but automated approval is not legal clearance and must not replace human review.

Safe operating practices also cover change control. Record every bid strategy change, budget change, tracking edit, and audience upload with the person responsible, the reason, and the expected effect. Make changes in a test or staging environment where possible, and verify conversion tracking with test conversions before counting results in reports. Avoid personal email addresses, personal messaging apps, and personal devices for account access unless the organization has approved them in writing. Where possible, use managed devices with current security updates.

Consent and tracking deserve particular care. Tags, pixels, and cookies must be configured in line with the consent management setup required in each market. Confirm with privacy specialists which consent and disclosure rules apply, because they differ by jurisdiction and by the type of data involved. Never place sensitive information, such as health conditions, financial hardship, religious belief, or sexual orientation, in URLs, custom audience names, or free text fields. Customer list uploads should be used only for purposes the customer was told about and that the organization has a documented basis to pursue.

• Mandatory: do not publish a price, discount, guarantee, or availability statement unless the supporting evidence is current and linked in the claim register.
• Mandatory: do not bid on a competitor's trademark or use third-party brand names in ad text unless counsel has confirmed the use is permitted in the relevant jurisdiction and on the relevant platform.
• Mandatory: do not target or exclude audiences using characteristics the platform prohibits, and do not upload customer lists without documented consent or another lawful basis.
• Mandatory: set a budget cap and an alert threshold on every account and campaign, and do not remove caps without account owner approval.
• Mandatory: enable two-step verification and use a password manager for all ad accounts. Never share credentials by email, chat, or spreadsheet.
• Mandatory: do not install browser extensions with access to ad accounts unless security has approved them.
• Recommended: add negative keywords for known irrelevant terms and review search term reports on a scheduled basis.
• Recommended: keep landing pages accessible, readable, and quick to load, and make sure the promise in the ad appears on the page.

5Equipment, environment, information, and protective safeguards

Paid search work is screen and keyboard intensive, so set up workstations with adjustable seating, screens at a comfortable eye level, and regular breaks away from the screen. These are recommended practices, and local workplace health and safety rules apply. Keep the workspace free of trip hazards and cables that could be pulled from power strips. Lock the screen when stepping away, even briefly, because ad accounts and customer data can be viewed by anyone at an unlocked workstation.

Information safeguards matter as much as physical ones. Exported search term reports, customer lists, and conversion data often contain personal data or commercially sensitive information. Store exports only in approved, access-controlled locations, share them through managed links with expiry settings where available, and delete temporary files when the task ends. Protect against shoulder surfing in shared spaces and avoid discussing customer-level details in public areas. Use licensed images, fonts, and text in creative assets, and keep proof of licenses with the campaign file.

• Mandatory: store exported customer data only in approved, access-controlled locations, and delete temporary exports when the task ends.
• Mandatory: mask or aggregate personal data in reports shared beyond the core team.
• Mandatory: do not capture screenshots that show other people's personal data, account credentials, or payment details.
• Mandatory: use only images, music, fonts, and copy that the organization has licensed or created, and keep the license record with the asset.
• Recommended: use a separate browser profile dedicated to ad account work, without unrelated extensions.
• Recommended: maintain a list of approved third-party tools, with a note on where each tool processes data and what contract terms apply.

6Incident reporting, emergency response, and escalation

Define incidents in advance so staff recognize them quickly. Relevant incidents include suspected account compromise, unauthorized spending, a platform suspension or disapproval for a restricted category, a misleading ad reported by a customer or competitor, a privacy exposure through tags or uploads, a trademark or copyright complaint, and an inquiry from a regulator or consumer group. Report every suspected incident to the account owner and to the privacy or legal lead promptly, and record when it was discovered, who found it, and what action was taken first. Do not wait for certainty before reporting.

Preserve evidence before making large changes where it is safe to do so. Capture screenshots of ads, URLs, search term reports, and change logs, because deleting an ad or account history can make a complaint harder to resolve. Legal notification duties after a data exposure are jurisdiction dependent, so counsel must decide on any external notice. Emergencies involving threats of harm to people, harassment, or immediate safety concerns go to local emergency services and the organization's security function, not to the platform support queue.

• Account compromise: change the password from a clean device, end active sessions, remove unknown users, and report through the platform's official support channels.
• Runaway spending: pause affected campaigns or reduce budgets to zero, then investigate. Do not resume without account owner approval.
• Misleading ad: pause the ad or asset, preserve evidence, and escalate to legal for review before any replacement copy runs.
• Data exposure: stop the tag or upload, contain the data, and follow the organization's breach process. Counsel decides on external notification duties.
• Incident reviews: record root cause, corrective action, responsible owner, and evidence that the action was completed.

7Records, training, inspection, and review

Keep records that show decisions, not just outcomes. Useful records include the claim register, campaign risk ratings, approvals, change logs, consent configuration records, customer list sourcing notes, license records, access reviews, training completion, and incident reports. Retention periods depend on jurisdiction, contracts, and internal policy, so confirm them with the records management function rather than deciding them locally.

Training should be role based and refreshed whenever platform advertising policies, privacy rules, or internal procedures change. Inspections should compare live ads against the claim register, check that tags match the approved configuration, confirm that access lists match current staff, and review negative keyword and search term reports for drift. Track policy disapproval rates, tracking discrepancies, and complaint volumes as quality indicators, and review trends over time rather than reacting to single events. Use findings to improve templates, checklists, and training content.

• Mandatory: complete role-based training on platform policies, privacy, and advertising claims before admin access, and refresh it when policies change.
• Mandatory: review user access for every ad account at least once every quarter and remove access that is no longer needed.
• Recommended: run a periodic audit that compares active ads, extensions, and landing pages against the claim register.
• Recommended: hold a short post-incident review and feed the lessons into the campaign launch checklist.

8Practical checklist and conclusion

Safe paid search depends on truthful ads, lawful data use, respect for trademarks and creative work, disciplined account security, and fast response when something goes wrong. Adapt these norms into locally approved procedures, and have legal, privacy, and security specialists review them against current laws, regulator guidance, and platform terms. Nothing in this document authorizes an individual to use personal data, make regulated claims, or bypass platform controls without the required approval.

Implementation checklist

• Confirm every ad claim against linked evidence before launch.
• Check trademark, copyright, and license status for brand names, images, and copy.
• Verify consent and a documented lawful basis before adding tags or uploading audiences.
• Obtain account owner approval for budget caps, billing changes, and new users.
• Enable two-step verification and use a password manager for every ad account.
• Log every bid, budget, tracking, and audience change with the reason and the person responsible.
• Remove sensitive personal data from URLs, reports, screenshots, and shared files.
• Report suspected incidents to the account owner and the privacy or legal lead without delay.
• Pause spending and preserve evidence before making large changes after an incident.
• Review access, training records, and this norm set on a regular schedule.