All resources

· 31 min read

15 Marketing Tasks You Can Hand to an AI Employee in 2026

The 15 marketing tasks you can hand to an AI employee in 2026, the ones that stay human, and the delegation matrix teams use to run Lapis safely.

Lapis, the YC-backed #1 AI ad generator rated 4.9/5 on G2, uses AI agents to create, launch, and optimize ads on ChatGPT, Google, Meta, and LinkedIn.

The 15 marketing tasks you can hand to an AI employee in 2026 fall into five groups: research (competitor ad monitoring and audience personas), planning (campaign briefs), production (ad copy, static creative, resizing and localization, and landing pages), execution (campaign launch, budget pacing, cross-channel allocation, testing, and creative refresh), and measurement (reporting, attribution against confirmed revenue, and ChatGPT and AI search advertising), while positioning, pricing, legal claims, and final approvals stay with people.

Most teams have adopted AI tools without moving the work itself, and the gap is measurable. HubSpot's 2026 State of Marketing report, published April 10, 2026, found that 67% of marketing teams say AI saves them 10 or more hours a week, yet DoubleVerify reported on October 29, 2026 that marketers still spend 9 hours and 58 minutes a week on manual campaign optimization, only 14 minutes less than a year earlier. A tool speeds up a task that a person still owns; an employee takes the task, the workflow around it, and the number it is meant to move.

I run operations at Lapis, where agents run paid growth across Meta, Google, LinkedIn, Reddit, and ChatGPT through OmniSense, which adds a dedicated strategist, and ChatSense, inside the budgets and approval rules each client writes down first. Below are the 15 tasks we see teams hand over, with the human time, evidence, and day-one permission level for each, followed by the delegation matrix and the hours arithmetic we use in onboarding.

Which Marketing Tasks Can You Hand to an AI Employee?

You can hand an AI employee five groups of marketing work in 2026, namely research, planning, production, execution, and measurement, which together cover the 15 recurring tasks that fill most of a paid marketer's week.

The distinction that matters is ownership, not capability. A full-stack AI marketer, as defined in the full-stack AI marketer explainer, is an AI employee that researches, plans, creates, launches, moves budget, measures confirmed revenue, and reports what it will do next, inside permissions a human sets, while an AI tool completes the one task you prompt and waits; the 10 differences between an AI tool and an AI employee draw that line in detail.

So AI can do your marketing for you in the sense that matters most to a lean team: it does the execution half today, and the strategic half only as an advisor. HubSpot's State of Marketing data, published April 10, 2026, shows that 34.1% of marketers use AI extensively for advertising automation and optimization, but HubSpot reported on September 2, 2026 that only 19.2% use AI agents to automate marketing initiatives end to end. The teams in that smaller group wrote down which tasks the AI owns, which it drafts for approval, and which it never touches.

Should You Delegate Tasks, Workflows, or the Whole Job?

Delegate whole workflows rather than isolated tasks, and hand an AI employee the entire job only where the outcome is measurable in your own systems and every action can be capped, logged, and reversed within a day.

Task delegation is where most teams stall, and DoubleVerify's report explains why in its own words: "AI has saved time within individual workflow steps. But overall, the total number of hours marketers spend on manual optimizations hasn't budged." When AI writes the headlines but a person still briefs it, builds the campaign, checks pacing, and assembles the report, the handoffs absorb the time the model saved. McKinsey wrote on June 22, 2026 that always-on AI orchestration can cut execution work from 60 to 70 percent of a marketer's time to as little as 10 to 15 percent, and shrink campaign cycles from six to ten weeks to same-day execution.

Three tests decide where each task belongs. The first asks whether the result is measurable in a system you own, such as your store, CRM, or billing platform; the second asks whether any action can be reversed within a day; the third asks whether a written rule can govern it, such as a budget cap or an approved claims list. A task that passes all three acts alone, a task that passes two needs approval, and a task that fails the reversibility test stays human, which is why budget pacing moves on day one and pricing never does.

The 15 Marketing Tasks You Can Hand to an AI Employee

The 15 tasks below run in the order a campaign does, from watching competitors to reporting confirmed revenue, and each states the human time it absorbs, what the AI employee does, the dated evidence, and the day-one permission level.

1. Competitor Ad Monitoring

An AI employee checks the Meta Ad Library, the Google Ads Transparency Center, and LinkedIn's ad library for every named rival on a schedule and logs each new angle, offer, and landing page. By hand, five rivals across three libraries means 15 checks a week, which is why stretched teams drop the task first. HubSpot reported on September 2, 2026 that 33.2% of marketers already use AI extensively for market research and competitor analysis, and Lapis runs the job through Brand Radar, using the method in how to analyze competitor ads with AI. Grant it at the acts-alone level with read-only access.

2. Audience Research and Personas

An AI employee reads your site, reviews, support tickets, and competitor messaging, then builds personas with the pains, objections, and triggers each angle needs, refreshing them whenever inputs change rather than after days of interviews once a year. Salesforce's tenth State of Marketing report, published February 19, 2026, found that 78% of marketers need more personalized content than they can produce and 75% are turning to AI to close the gap. Lapis learns your audiences when you paste in your website, and the segment method is in how to create ads for different audience segments. Research runs alone, and a new segment needs approval before it receives spend.

3. Campaign Briefs

An AI employee drafts the brief from the goal, budget, audience, offer, and past results, which fixes the step teams most often compress into a Slack message when a launch date is close. DoubleVerify found on October 29, 2026 that 37% of marketers used AI to summarize media briefs in 2025, up from 28% a year earlier, and HubSpot launched Campaign Agent on September 16, 2026 to turn plain-language goals into campaign plans. Feed the draft the 12 inputs in the AI ad creative brief template. The AI drafts alone, and a named person approves the goal, budget, and offer before anything is built.

4. Ad Copy for Every Platform

An AI employee writes to each platform's format, from Meta primary text and Google responsive search headlines to the 50-character format in how to write ChatGPT ad copy, keeps a library of approved claims, and retires losing lines, sparing a copywriter hours of reformatting before any test runs. Meta's researchers reported in a paper revised December 15, 2025 that AdLlama, a model trained on ad performance, lifted click-through rates 6.7% across nearly 35,000 advertisers. Copy runs alone inside the approved claims library, and any new claim, statistic, or comparison needs approval.

5. Static Ad Creative

An AI employee pairs each on-brand visual with a distinct angle rather than a cosmetic variation, so creative ships daily instead of at the pace of one designer's calendar. Meta said on October 21, 2026 that nearly 10 million advertisers used its generative AI creative tools in the third quarter, and a March 2025 study in the International Journal of Research in Marketing found an AI-generated banner beat a stock photo on click-through by more than 50%, at 0.80% against 0.53%. Lapis creates static image ads and copy, and teams pair it with a video tool for footage, as ChatGPT vs Gemini vs Lapis for making ads explains. Creative runs alone inside the brand kit, and new angles need approval.

6. Resizing and Localization

Resizing and localization quietly cap testing volume, because one concept in three aspect ratios and four languages is 12 assets before a single variation is tested, and five concepts a week becomes 60. Alphabet said on October 28, 2026 that Gemini helped advertisers generate more than 1.5 billion campaign assets in the third quarter, a volume that exists only because machines now handle the multiplication, as creating ad creatives with AI at scale and multilingual ad campaigns with AI describe. Resizing runs alone, and translated regulated claims get a native-speaker review.

7. Landing Pages

Paid traffic too often lands on a homepage built for every visitor at once, because a page per ad used to need a designer, a developer, and a copywriter. An AI employee builds a dedicated page for each ad and tests headlines and forms against real conversions. Unbounce's 2024 Conversion Benchmark Report, summarized in a guide updated July 25, 2025, put the median landing page conversion rate at 6.6% across 41,000 pages, the baseline a matched page has to beat. Lapis builds these pages through RapidDomain, and the evidence is in whether custom landing pages increase conversion rates. Pages run alone from approved templates, and new offers or prices need approval.

8. Campaign Build and Launch

Digiday described the manual version on May 13, 2026, when TikTok opened its ads platform to outside agents, as the laborious, click-heavy work of setting up creatives, adjusting bids, shifting budgets, and tweaking targeting. DoubleVerify's October 29, 2026 report found that 35% of marketers used AI to activate campaigns in 2025, while Meta's ads MCP server creates every new campaign paused and Google's Ask Advisor, introduced May 20, 2026, acts only with the advertiser's approval. Lapis builds and launches across up to three agreed channels on OmniSense, following the path in how to turn a product URL into a complete AI ad campaign. Let it build alone, launch with approval for 30 days, and then launch alone inside approved caps.

9. Budget Pacing

Overspend and underspend both happen overnight, so pacing suits an employee that checks spend against plan several times a day, adjusts daily budgets inside a cap, and pauses delivery when a stop rule trips. DoubleVerify reported on October 29, 2026 that 57% of marketers spend between 8 and 16 hours a week on manual optimizations, and that 31% used AI for mid-flight optimization in 2025, with 36% planning to in 2026. Every Lapis plan includes budget guardrails and stop rules, and the levers are in the AI ad optimization playbook. Pacing runs alone inside a daily change cap and a monthly ceiling, the structure Meta's rules API enforces when it denies agent budget increases above a set percentage such as 20%.

10. Cross-Channel Budget Allocation

No platform AI moves money to a rival, and in DoubleVerify's survey of 2,020 marketers, published October 29, 2026, 41% named budget reallocation complexity as a top challenge of advertising across platforms. An AI employee compares marginal cost per confirmed customer by channel and moves budget to where the next dollar earns most. If LinkedIn costs $210 per qualified demo at the margin and Google costs $140, moving $7,000 a week to Google buys 50 demos instead of about 33, a gain of 17 at the same spend. Lapis moves budget into winners inside your guardrails, using the logic in ChatGPT Ads vs Google Ads vs Meta Ads. Weekly shifts up to 15% of budget run alone, and larger moves need approval.

11. A/B and Angle Testing

An AI employee runs distinct angles as experiments with a hypothesis, a budget, and a stopping rule, while human teams test less than they intend because each test needs new creative, a new page, and someone to read the result. Interconnections' analysis of 368 Meta creatives across 15 DTC brands, published August 12, 2026, found that the top 10% of creatives carried 68% of spend, so a team that tests few concepts rarely finds the one that carries the account. The logic is in AI ad angles vs variations and A/B testing AI ads on a small budget. Testing runs alone inside a fixed test budget and approved angles.

12. Creative Fatigue Detection and Refresh

Fatigue is invisible in a weekly report and obvious in daily data, and Meta's analytics team showed on May 10, 2023 that by four repeated exposures to the same creative the likelihood of a conversion falls by about 45%, while adding new creative to high-fatigue ad sets lifted conversion rates 8% on average across about 26,000 split tests. An AI employee tracks cost per acquisition by creative and rotates in the next tested angle when the curve turns; Interconnections found in its August 12, 2026 benchmark that cost per acquisition rose 19.6% by the 500,000 to 1 million impression mark while click-through fell only 5.5%, so cost should trigger the refresh, as how to fix creative fatigue explains. Rotating approved creative runs alone.

13. Daily and Weekly Reporting

Reporting has the most measurable hours: Emplifi reported on October 30, 2026 that 62% of marketers spend four or more hours a week on reporting and analytics, and NinjaCat found on April 15, 2026 that 77% of 532 marketing leaders say data preparation alone takes 3 to 10 hours per report. An AI employee sends a daily note with spend against plan, confirmed results against target, the changes it made and why, and one decision it needs. Every Lapis plan includes custom reporting on an agreed cadence, and the method is in analyzing Meta and Google ads data with ChatGPT. Reporting runs alone, because a report changes nothing until a person or a rule acts on it.

14. Attribution Reconciliation Against Confirmed Revenue

Each platform counts every conversion it touched inside its own window, and the totals overlap: the Outcomes Marketing Council, convened by Affinity Solutions, reported on May 18, 2026 that 91% of 210 senior marketers believe platform-reported results are overstated. An AI employee pulls confirmed orders or qualified leads from your store, CRM, or billing system and counts each conversion once. If Meta claims 260 purchases, Google claims 190, and Shopify records 340 paid orders, at least 110 were counted twice, and optimizing to platform claims overfunds whichever channel claims most. Lapis runs this through Wayfinder, in every plan, with the discipline in the ChatGPT ads ROI measurement framework. Reconciliation runs alone, and a human owns the conversion definition.

15. ChatGPT Ads and AI Search Visibility

OpenAI said on October 21, 2026 that ChatGPT serves 1.2 billion weekly users, that more than 100,000 advertisers have run ChatGPT Ads, and that ads run in 89 countries on the Free and Go plans, and Salesforce found on February 19, 2026 that 88% of marketers have begun optimizing for AI-generated answers. An AI employee writes creative tuned to how people ask ChatGPT, optimizes daily, and tracks which prompts turn into confirmed customers. Lapis runs this through ChatSense in all 89 countries, and Enterprise scope can add Google AI Overviews and AI Mode placements; see 10 ways to advertise on ChatGPT and, for organic visibility, getting recommended by ChatGPT. Launch needs approval, with a $25 minimum daily budget in the US, and daily optimization then runs alone.

Which Marketing Tasks Should Stay Human?

Positioning, offers and pricing, regulated claims, final approval of new creative directions, creator relationships, and big brand bets should stay human, because each one defines the company rather than executing a plan, and each one fails the reversibility test.

DecisionWhy it stays humanWhat the AI employee does instead
PositioningA wrong turn takes quarters to undoTests angles inside the position
Offers and pricingThey change unit economicsProposes offer tests for finance
Regulated claimsLiability sits with the companyUses approved claims and flags new ones
New creative directionsA new tone is a brand decisionDrafts it and stages it paused
Creator relationshipsTrust between people does not transferMeasures which creator assets win
Big brand betsThey trade short-term return for positionSizes the bet and measures the halo
Conversion definitions and ceilingsAn AI that sets its own target grades its own homeworkWorks to the definition you write

The legal line is the firmest, because penalties land on the company rather than the software: the Federal Trade Commission finalized its rule on consumer reviews and testimonials on August 14, 2024, banning fake ones, including AI-generated testimonials, and its December 2025 warning letters cite civil penalties of up to $53,088 per violation. Taste is the other line, since IAB and Sonata Insights reported on January 15, 2026 that 82% of ad executives believe Gen Z and Millennial consumers feel positive about AI-generated ads while only 45% of those consumers do. Gartner said on May 11, 2026 that CMOs must clarify "where AI supports judgment, and where humans remain accountable," and the matching partner questions are in 25 questions to ask an ad agency.

The AI Delegation Matrix: Which Tasks Act Alone, Need Approval, or Stay Human?

The AI Delegation Matrix assigns every part of the 15 tasks to one of three autonomy levels, acts alone, needs approval, or human only, and pairs each task with the guardrail that makes the assignment safe enough to paste into a vendor contract unchanged.

#TaskActs aloneNeeds approvalHuman onlyGuardrail
1Competitor monitoringLibrary checks and alertsNoneResponding to a rival's moveRead-only access
2Audience personasResearch and persona draftsNew segments before spendWho the brand is forNo restricted categories
3Campaign briefsFirst draftsGoal, budget, and offerQuarterly strategyTemplate with required inputs
4Ad copyCopy inside approved claimsNew claims and comparisonsRegulated claimsClaims library and banned words
5Static creativeVisuals inside the brand kitNew angles before spendBrand identity changesLocked logo, colors, and type
6Resizing and localizationAll resizingTranslated regulated claimsMarket entryNative-speaker review
7Landing pagesPages from approved templatesNew offers or prices on a pagePricing policyPage change log
8Build and launchBuilding paused campaignsLaunch for the first 30 daysEntering a new channelNew campaigns start paused
9Budget pacingDaily changes inside the capDaily increases above 20%The monthly ceilingChange cap plus absolute ceiling
10Cross-channel allocationWeekly shifts up to 15%Larger shifts and new channelsAnnual channel strategyChannel floors during tests
11A/B and angle testingTests inside the test budgetTests that need a new offerTopics never to testFixed budget and stopping rule
12Creative fatigue refreshPausing and rotating approved creativeNoneNoneTrigger on cost, not frequency
13ReportingAll reports on cadenceNoneBoard narrativesFixed format with one decision request
14AttributionMatching and deduplicationAttribution window changesThe conversion definitionSystem of record agreed with finance
15ChatGPT ads and AI searchDaily optimizationLaunch and budget increasesTopics to avoid$25 US daily minimum; exclusions

Use the matrix as a ladder rather than a fixed chart. Start a new AI employee one column to the right of where the table places each task, promote a task after 30 days in which every logged change has a written reason and no guardrail was breached, and demote it immediately after any breach, recording each move where finance, legal, and the owner of the number can see it. The five-stage end-to-end automation workflow sequences the same handoffs.

How Many Hours Does an AI Employee Give Back Each Week?

An AI employee returns about 14 hours a week to a paid marketer from optimization and reporting alone, and about 63 hours a week to a three-person team once execution work falls to McKinsey's benchmark, as the arithmetic below shows.

The bottom-up estimate uses the two tasks with the best data. DoubleVerify measured 9 hours and 58 minutes a week of manual optimization on October 29, 2026, about 9.97 hours, and Emplifi found on October 30, 2026 that 62% of marketers spend at least four hours a week on reporting, so a single paid marketer spends about 14 hours, or 35% of a 40-hour week, on tasks 8 through 14 alone. The top-down estimate uses McKinsey's June 22, 2026 finding that execution takes 60 to 70 percent of marketers' time and can fall to 10 to 15 percent: for a three-person team with 120 weekly hours, the midpoints of 65% and 12.5% return 63 hours, within a range of 54 to 72.

Hours returned per week
  = team hours x (execution share today - execution share after)
  = 120 x (0.65 - 0.125)
  = 63 hours

Loaded hourly cost
  = median salary / 0.70 / 2,080 hours
  = $78,760 / 0.70 / 2,080
  = $54.09

Annual value of hours returned
  = 63 x $54.09 x 52 weeks
  = $177,199

The salary input is the Bureau of Labor Statistics median of $78,760 for market research analysts and marketing specialists in May 2025, and the 0.70 divisor reflects the BLS finding, released September 9, 2026, that wages were 70.0% of private-industry compensation in June 2026, so the 63 hours are worth about $177,200 a year, more than one and a half fully loaded specialists. The hours come back only if the work leaves the calendar, which is the lesson of DoubleVerify's flat 14-minute improvement, so spend them on the human column of the matrix and set each channel's target first with AI ad performance forecasting.

Will an AI Employee Replace Your Marketing Team?

An AI employee replaces marketing execution rather than marketing teams: the US Bureau of Labor Statistics projected on August 27, 2026 that marketing manager jobs will grow 7% from 2025 to 2035, more than double the 3% average for all occupations.

BLS projects marketing manager employment rising from 421,600 to 450,800 and market research analysts and marketing specialists from 952,700 to 1,019,000, while advertising and promotions managers, the role closest to buying media, fall 4%, from 33,200 to 31,900. McKinsey's state of AI survey, published August 25, 2026, found that only 14% of respondents at organizations using AI said it reduced their workforce over the past year, less than half the 32% who expected cuts a year earlier. Gartner reported on May 11, 2026 that marketing leaders expect AI automation of marketing work to rise from 16% in 2026 to 36% by 2028, and the roles that grow are the ones directing AI employees, as AI for performance marketers and will AI replace media buyers by 2030 explain.

How Do You Hand Marketing Tasks to an AI Employee Without Losing Control?

Hand marketing tasks to an AI employee the way you would onboard a senior hire with a company card: set permissions by platform, cap budgets, define approval thresholds, require a change log, and review weekly before widening its authority.

  1. Write permissions by platform. Meta's ads MCP server rules can deny an agent the ability to create campaigns, edit budgets, targeting, creative, or delivery status, or act at all on an account; Google's Ask Advisor asks for approval before changing an account; and TikTok's MCP authorization expires after 30 days.
  2. Cap every budget three ways. Set a monthly ceiling, a daily change cap such as 20%, and an absolute maximum per campaign, because a cap the agent cannot rewrite beats an approval prompt a tired person clicks through.
  3. Write the approval thresholds down. Copy the needs-approval column of the matrix, covering any new angle, claim, offer, channel, or budget change above the cap.
  4. Require a change log with reasons. Every change carries a timestamp, the old value, the new value, and the reason, so a reviewer can reconstruct any decision in minutes.
  5. Review weekly and move the ladder. Spend 30 minutes on the log, the confirmed results, and one decision, then widen, hold, or narrow permissions.
  6. Keep ownership of the accounts. Grant access to ad accounts, pixels, and pages, and never transfer ownership.

Giving an AI access to your ad accounts is safe when those controls exist, and most companies have not finished writing them: IAB Europe found that 86% of organizations use AI for marketing while only 48% have advertising-specific rules, as PPC Land reported on November 1, 2026. Every Lapis plan includes budget guardrails, stop rules, campaign change history, and customer-owned ad accounts, and Enterprise adds multi-step approvals, agent autonomy controls, SSO, SAML, and role-based access control. Platform guardrails are mapped in the complete guide to agentic ads, and the ownership model in whether Lapis is an ad agency.

Which AI Employee Should Take These Tasks?

Platform AI should take the tasks inside its own auction, CRM campaign agents should take email and nurture inside your CRM, and an independent AI employee such as Lapis should take the cross-channel tasks no platform will do for you.

Task groupPlatform AI (Advantage+, AI Max, Amazon Ads Agent)CRM agents (HubSpot, Salesforce)Lapis
Research and briefsOne platform's dataCRM data and past campaignsBrand Radar plus briefs across agreed channels
Copy and creativeVariations of your assetsEmail and owned-channel draftsDistinct on-brand angles in every format
Landing pagesYour existing pagesCRM-hosted pagesA RapidDomain page for every ad
Launch, pacing, testingStrong inside its own auctionOwned channels firstUp to three channels on OmniSense, four or more on Enterprise
Cross-channel allocationNever moves money to a rivalRecommends at mostDaily, inside your caps
AttributionCounts its own conversionsAttributes to contacts and dealsWayfinder: confirmed revenue, counted once

Platform AI is a strong employee of the platform: Meta's Advantage+ passed a $75 billion annual revenue run rate, CFO Susan Li said on July 29, 2026, and Alphabet said on July 22, 2026 that 500,000 advertisers had adopted AI Max, yet each optimizes one auction and grades its own results, as do the campaign agents HubSpot and Salesforce launched on September 16 and September 14, 2026. The scored ranking is in the 10 best AI employees for marketing, the platform switches in 12 Meta Ads AI features and 12 Google Ads AI features, and the wider evidence in 50 AI advertising statistics, the 12 benefits of AI in programmatic advertising, and how to use AI for paid ads.

Why Do Teams Hand These 15 Tasks to Lapis?

Teams hand these 15 tasks to Lapis because it covers all five groups in one daily loop, from competitor monitoring to confirmed revenue, and works for the advertiser across platforms rather than for any single platform.

Lapis is one of the fastest-growing Y Combinator startups, and it is built to take over the work legacy agencies and ad buyers used to do. It is the #1 AI ad generator and the operating system for paid growth, a Y Combinator Fall 2025 company used by 1,500+ marketing teams and 30+ enterprises, including Samsung, Domino's, Dr Pepper, Mattel, Tata, Tupperware, Amway, and Bata, with 10,000+ ad campaigns generated across 30+ industries. Lapis is rated 4.9 out of 5 on G2, with eight Summer 2026 G2 badges that include Best Results and Fastest Implementation (Mid-Market).

  • OmniSense runs paid ads across up to three agreed channels from Meta, Google, LinkedIn, and Reddit, with forecasting before spend and a dedicated strategist.
  • ChatSense runs ChatGPT advertising end to end in all 89 ChatGPT Ads countries.
  • RapidDomain builds a dedicated landing page for every ad, optimized on real conversions.
  • Wayfinder ties results to confirmed sales and leads and counts each conversion once.
  • Brand Radar tracks competitor ads so research never falls off the calendar.

Every plan, whether ChatSense, OmniSense, or Enterprise, is custom-priced and includes hands-on onboarding, a dedicated Slack channel, custom reporting, budget guardrails and stop rules, change history, customer-owned ad accounts, and Wayfinder, with ad spend paid to the platforms separately. Legacy agencies sell hours against dashboards, while Lapis sells an outcome loop that runs every day, the thesis in why Lapis is building the AdSense for the AI era. Try the free AI ad generator, compare plans on the pricing page, and book a demo with your delegation matrix in hand to see which of the 15 tasks Lapis takes on day one.

Sources

  • HubSpot, "2026 state of marketing: Data from 1,500+ global marketers," April 10, 2026
  • HubSpot, "AI marketing predictions that will shape 2026," September 2, 2026
  • DoubleVerify, "Beyond the Hype: The Real State of AI in Digital Advertising," October 29, 2026
  • PPC Land, "Marketers still spend 10 hours a week on manual optimization, DV finds," November 1, 2026
  • Salesforce, "75% of Marketers Have Adopted AI, Yet Still Use It to Send Generic Campaigns," February 19, 2026
  • McKinsey, "The future of marketing in the age of AI," June 22, 2026
  • McKinsey, "The state of AI in 2026: On the road to ROI," August 25, 2026
  • Gartner, "Gartner Survey Reveals Marketing Leaders Expect AI Automation of Marketing Work to Double by 2028," May 11, 2026
  • U.S. Bureau of Labor Statistics, "Advertising, Promotions, and Marketing Managers," Occupational Outlook Handbook, August 27, 2026
  • U.S. Bureau of Labor Statistics, "Market Research Analysts," Occupational Outlook Handbook, August 27, 2026
  • U.S. Bureau of Labor Statistics, "Employer Costs for Employee Compensation, June 2026," September 9, 2026
  • Emplifi via Vision Monday, "Marketing Productivity Gets a Boost From AI, but ROI Lags, Emplifi Reports," October 30, 2026
  • NinjaCat, "AI Maturity Across The Analyze-Optimize-Act Cycle," April 15, 2026
  • Affinity Solutions via Marketing Dive, "Nearly 91% of marketers believe their platform results are overstated," May 18, 2026
  • Meta Analytics, "Creative Fatigue: How advertisers can improve performance by managing repeated exposures," May 10, 2023
  • Interconnections, "Meta Creative Fatigue Benchmark 2026," August 12, 2026
  • Unbounce, "What's a good conversion rate? (Based on 41,000 landing pages)," July 25, 2025
  • Meta, "Third Quarter 2026 Results Conference Call," October 21, 2026
  • Meta, "Second Quarter 2026 Results Conference Call," July 29, 2026
  • Meta for Developers, "Rules best practices," ads MCP server documentation, August 12, 2026
  • Alphabet, "2026 Q2 Earnings Call," July 22, 2026
  • Alphabet, "2026 Q3 Earnings Call," October 28, 2026
  • Google, "Meet Ask Advisor, your new AI-powered collaborator," May 20, 2026
  • Digiday, "TikTok launches MCP server to let AI agents run campaigns," May 13, 2026
  • TikTok for Business, "How to connect Claude to TikTok for Business MCP Server," developer documentation
  • HubSpot, "Fall 2026 Spotlight," September 16, 2026
  • Salesforce, "Introducing The Campaign Agent That Turns Goals Into Growth," September 14, 2026
  • OpenAI, "ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform," October 21, 2026
  • Daniel R. Jiang et al., "Improving Generative Ad Text on Facebook using Reinforcement Learning," arXiv, December 15, 2025
  • Jochen Hartmann, Yannick Exner, and Samuel Domdey, "The power of generative marketing," International Journal of Research in Marketing, March 2025
  • Federal Trade Commission, "Trade Regulation Rule on the Use of Consumer Reviews and Testimonials," August 14, 2024
  • Federal Trade Commission, Consumer Review Rule warning letter template, December 2025
  • IAB, "The AI Ad Gap Widens," January 15, 2026

Related Lapis Resources

Built by Lapis

The #1 AI ad generator, built into the operating system for paid growth.

Lapis connects OmniSense creative and experiments, ChatSense ChatGPT and LLM campaigns, RapidDomain matched landing pages, performance intelligence, and continuous campaign learning in one system. Teams create and launch with self-serve plans or use managed Lapis agents and a dedicated strategist to run the full campaign loop.

Lapis is rated 4.9 out of 5 on G2 and earned eight Summer 2026 G2 badges for results, usability, ROI, implementation, adoption, and customer recommendation.

Continue exploring

Frequently Asked Questions