Everyone is buying AI tools. Far fewer can say what they got back. Global research points to one place where AI most often adds revenue, and three workflows where a growing business can start.
McKinsey's 2025 global survey found revenue gains from AI are most often reported in marketing and sales. Its earlier analysis estimated generative AI could add value worth 5% to 15% of total marketing spending. Start with lead qualification and follow-up, content operations, or reporting. Pick one workflow and one number, get your data in one place, redesign the process, keep a person accountable, and measure before and after.
- Revenue gains from AI are most commonly reported in marketing and sales (McKinsey, 2025).
- Generative AI could lift marketing productivity by 5% to 15% of total marketing spend (McKinsey, 2023).
- 88% of organisations use AI, but only about a third are scaling it and 39% report any earnings impact.
- Start with lead follow-up, content operations or reporting, each tied to one clear number.
- In regulated sectors, keep a person checking every AI output; NMC now bans AI-generated promotional campaigns by doctors and hospitals.
What did McKinsey find?
McKinsey's global survey, The state of AI in 2025, collected 1,993 responses from organisations in 105 countries between June and July 2025. It asked where AI is actually delivering value. Revenue gains from AI are most commonly reported in marketing and sales, strategy and corporate finance, and product and service development. Cost benefits show up most in software engineering, manufacturing and IT (McKinsey, 2025).
The same survey shows how far most companies still have to go:
- 88% of organisations use AI in at least one business function, up from 78% a year earlier.
- Only about a third have started scaling AI across the organisation.
- 39% report any impact on earnings from AI, and most of those say it's less than 5%.
- About 6% are "AI high performers", attributing 5% or more of earnings to AI.
McKinsey's earlier economic analysis explains why marketing and sales stand out. Of the 63 generative AI use cases it studied, about 75% of the potential value fell in four areas: customer operations, marketing and sales, software engineering and R&D. It estimated generative AI could lift marketing productivity by a value of 5% to 15% of total marketing spending, and sales productivity by about 3% to 5% of global sales spending (McKinsey, June 2023).
Where generative AI's marketing and sales value comes from
Why marketing and sales?
Marketing and sales are full of work that repeats every day, involves a lot of text and data, and has a clear number attached: leads, response time, conversion, cost per customer. That's exactly where AI saves time and shows results quickly.
Marketing leaders are seeing it already. In Gartner's 2025 CMO survey, 49% reported improved time efficiency from generative AI, 40% improved cost efficiency and 27% more capacity to produce content or handle more business. Only 1% said generative AI wasn't a priority (Gartner, 2025).
3 places AI pays off first
| Workflow | What AI does | What a person still does | Number to watch |
|---|---|---|---|
| Lead qualification and follow-up | Acknowledges enquiries instantly, asks qualifying questions, routes hot leads | Calls, consults, closes | Time to first response; lead-to-appointment rate |
| Content operations | First drafts, variations, resizing, FAQs in your voice | Decides the idea, checks facts and claims, approves | Hours per piece; output per week |
| Reporting and insight | Pulls numbers from ads and CRM, drafts a plain-language summary, flags anomalies | Interprets, decides what to change | Hours spent on reporting; decisions made per month |
1. Lead qualification and follow-up
AI can acknowledge an enquiry the moment it arrives, ask two or three qualifying questions on WhatsApp or chat, and route the lead to the right person with context. Speed is the win: a lead answered in minutes is worth far more than one answered tomorrow. We explain the routine in Your leads go cold while you sleep.
2. Content operations
First drafts of posts, ad variations, emails, product descriptions and FAQs, in your brand's voice, reviewed by a person before they go out. The team spends more time on ideas and less on blank pages. The limit is judgment: AI writes plausible copy, but it doesn't know which claims you can prove, which matters enormously in regulated sectors.
3. Reporting and insight
Pulling numbers from ad platforms and your CRM into one weekly summary, with notes on what changed, takes hours by hand. AI and automation can do it in minutes and flag what needs attention, like a campaign whose cost per lead has doubled. People still decide what to do about it.
A simple business case (illustrative)
A college's admissions team handles 3,000 enquiries a season. Counsellors take an average of 6 hours to make first contact, and 18% of enquiries book a campus visit.
With AI acknowledging every enquiry instantly, answering fee and course questions on WhatsApp, and routing serious applicants to counsellors, first contact drops to under 15 minutes. If visits rise even from 18% to 22%, that's 120 more families on campus from the same ad spend. That's the kind of number to test against: specific, measured before and after, and tied to revenue.
How to make AI pay off
- Pick one workflow and one number. For example: cut average lead response time from hours to minutes.
- Get the data in one place first. AI needs clean inputs. Gartner found only 30% of marketing teams are ready to scale AI, mostly because processes and data aren't mature (Gartner, 2026). See why only 30% of teams can scale AI.
- Redesign the workflow, don't just add a tool. McKinsey's high performers were nearly three times as likely to have fundamentally redesigned their workflows (McKinsey, 2025).
- Keep a person accountable. AI drafts and sorts; people approve and decide. McKinsey's high performers also more often defined how humans validate AI outputs.
- Measure before and after, so you know whether it worked, then move to the next workflow.
Guardrails for regulated sectors
If you market healthcare, education, real estate or legal services, AI needs firmer rules. For example, the National Medical Commission's October 2026 guidelines prohibit AI-generated promotional campaigns for commercial interest by doctors and hospitals, require AI-made content to be labelled, and ban AI-manipulated patient images, voices or testimonials (NMC, Oct 2026). We cover the details in NMC's new ad guidelines. Education and property ads have their own rules too: see admission ads and RERA ads.
A sensible default: AI can draft, sort and summarise, but nothing customer-facing goes out without a person checking it against the rules for your sector.
What to avoid
- Ten tools, no owner. Spread budget and nobody learns any tool properly.
- AI on top of a broken process. It just produces mistakes faster.
- Automation in someone else's accounts. If your chatbot, prompts and data live with a vendor or agency, the capability isn't yours. More in 1 in 5 CMOs say GenAI cut their agency reliance.
AI pays off where the work repeats and the number is clear. In most businesses, that's marketing and sales.
How Ad Asthra helps AI pay off
We set AI up inside your own accounts and workflows, and train your team to run it.
Frequently asked questions
Where does AI deliver the most value in business?
McKinsey's 2025 survey found revenue gains from AI are most often reported in marketing and sales, strategy and corporate finance, and product development, while cost savings show up most in software engineering, manufacturing and IT.
How much can generative AI improve marketing productivity?
McKinsey estimated in 2023 that generative AI could deliver value worth 5% to 15% of total marketing spending, and about 3% to 5% of global sales spending.
What's the best first AI project for marketing?
Lead qualification and follow-up is usually best: the steps are simple, speed matters, and you can measure response time and appointments before and after. Content operations and reporting are good next steps.
Why do most companies not see ROI from AI?
Most haven't scaled beyond pilots. McKinsey found only about a third are scaling AI, and high performers are far more likely to have redesigned their workflows rather than just adding tools.
Can AI write our marketing content?
AI can write good first drafts, variations and FAQs. A person should decide the idea, check facts and claims, and approve anything customer-facing, especially in healthcare, education, real estate and legal marketing.
Is AI allowed in healthcare marketing in India?
NMC's October 2026 guidelines prohibit AI-generated promotional campaigns for commercial interest by doctors and hospitals, require AI-made content to be labelled, and ban AI-manipulated patient images, voices or testimonials.
How does Ad Asthra help businesses use AI in marketing?
Our AI Suite brings platform data into one place with dashboards and automated insights. We set up AI-assisted workflows in your accounts, train your team through Enablement Programs, and, through BOT, hand the whole system over to you.
Sources
- The state of AI in 2025: Agents, innovation, and transformation, McKinsey & Company, 2025
- The economic potential of generative AI: The next productivity frontier, McKinsey & Company, 14 June 2023
- Gartner 2025 CMO Spend Survey Reveals Marketing Budgets Have Flatlined at 7.7% of Overall Company Revenue, Gartner, 12 May 2025
- Gartner 2026 CMO Spend Survey Finds CMOs Allocate 15.3% of Marketing Budgets to AI, But Only 30% Are Ready to Scale AI Capabilities, Gartner, 11 May 2026
- Public Notice: Guidelines on Ethical Advertising and Public Communication by Hospitals/Medical Institutions and Registered Medical Practitioners, National Medical Commission, 6 October 2026
Last updated
Curious where AI could save your team the most time? Let's map it out.
Message us on WhatsApp or book a call