<h2>Executive Summary</h2><ul><li><p>Most firms are <strong>funding AI within existing MarTech budgets</strong>. Highly regulated sectors are the exception, where AI spend sits closer to IT infrastructure.</p></li><li><p>Martech stacks are <strong>consolidating from many point tools into fewer AI-native platforms</strong>. Vendor count is falling, yet complexity persists.</p></li><li><p>Firms are responding to AI through <strong>upskilling, governance charters, training and role redesign</strong>, rather than headcount reduction. </p></li><li><p><strong>The biggest investments are in unifying internal data</strong> across partners, platforms and functions.</p></li><li><p>As data becomes marketing’s core asset, the <strong>CMO’s remit is widening into facilities, operations and client experience</strong>.</p></li><li><p>In more highly regulated sectors, <strong>compliance issues are slowing AI adoption</strong>. Here, firms tend to favour proprietary infrastructure and human oversight.</p></li><li><p>AI has delivered speed and scale in content production but <strong>not (yet) the expected step-change in ROIs</strong>.</p></li></ul>.<p>Marketing teams are adding AI and analytics tools faster than they can measure their value. At the same time, CMOs are being asked to take responsibility for growth, customer experience and brand trust, even when they do not control all the systems or teams involved. At a recent open-house discussion, members of the India CMO Forum compared where AI spending is going, whether their MarTech stacks are becoming simpler and how far the CMO’s responsibilities now extend beyond marketing.</p><h2><strong>The Uneven Arithmetic of AI Investment</strong></h2><p>Most organisations represented had not created a separate AI budget. They were funding it through existing MarTech, marketing or digitalisation allocations. Some firms were increasing their MarTech budgets to absorb AI costs and measuring the returns through indicators such as cost per piece of content or total output. Others were approving AI spending case by case, without a fixed annual allocation. Regulated and data-sensitive firms were the main exception. These organisations were more likely to hold AI budgets centrally and to require each proposed use-case to meet a formal business need. The discussion indicated that budgeting discipline depends heavily on data sensitivity and regulatory exposure.</p><p><strong>From Tool Sprawl to Tool Consolidation</strong><br>MarTech stacks have traditionally consisted of several specialised tools, each covering a narrow task. This is starting to change with AI-native platforms that can now combine functions such as copy editing, research and content generation that previously required separate licences. Some organisations are also bringing social listening and reputation monitoring in-house instead of relying on several external agencies for separate reports. However, fewer vendors does not necessarily mean simpler operations. Rather, the responsibility for maintaining data quality, testing model output and managing reliability has moved inside the organisation.</p><h2><strong>Building New Capabilities</strong></h2><p>None of the organisations represented at the meeting say they are reducing headcounts because of AI. Most were instead investing in training, governance and changes to existing roles. One approach discussed was an AI Governance Charter, backed by the CEO, with regular training for the leadership team and dedicated training budgets across levels of the organisation. Another organisation reported that almost its entire workforce has received AI training, supported by weekly sessions and governance councils at both the organisational and business-line levels. The immediate effect has been higher output expectations from broadly unchanged teams. Role descriptions are also changing. Copywriters, for example, are spending less time producing first drafts and more time on content strategy, editing and quality control.</p><h2><strong>Data as the Precondition for Intelligence</strong></h2><p>AI tools only become useful when the organisation can combine and trust its internal data. One firm replaced an external PR measurement retainer with an internal tool built in three weeks. The tool tracks media mentions and share of voice, then feeds the information into sales and business-development systems. This has helped the company identify early signals of global capability centres (GCCs) exploring entry into Indian markets. The same organisation is combining data on space, HVAC and Wi-Fi use across client facilities to improve personalisation. In both examples, the AI application depends on connecting data that had previously been held by different teams, partners and platforms.</p><h2><strong>Where the Mandate Now Extends Beyond Marketing</strong></h2><p>Marketing is being drawn into areas such as customer experience, facilities and operations because it increasingly holds the data used across these functions. Marketing performance is no longer measured only through awareness and demand generation. It now includes customer experience and retention. This may place the CMO in a strong position to connect customer data across the organisation, even when formal responsibility for the underlying activity sits elsewhere.</p><h2><strong>Regulation as a Constraint</strong></h2><p>In highly regulated sectors, customer data cannot easily be sent to public large language models (LLMs). Firms are therefore building proprietary systems using open-source frameworks. The expected move towards consent-based data architecture under the Digital Personal Data Protection (DPDP) Act strengthens the case for this approach. Much of the AI budget in these organisations goes into processing infrastructure and model development before any customer-facing application is launched. Human review also remains necessary because confidence in model output is not yet high enough to remove supervision. For these firms, the pace of adoption is determined primarily by governance, security and data sensitivity. Class-leading organisations treat data architecture as part of marketing operations rather than as a separate IT issue. Tools, budgets and governance structures will continue to change, but the basic requirement will remain the same: firms need control over their data before they can automate decisions with confidence</p>
<h2>Executive Summary</h2><ul><li><p>Most firms are <strong>funding AI within existing MarTech budgets</strong>. Highly regulated sectors are the exception, where AI spend sits closer to IT infrastructure.</p></li><li><p>Martech stacks are <strong>consolidating from many point tools into fewer AI-native platforms</strong>. Vendor count is falling, yet complexity persists.</p></li><li><p>Firms are responding to AI through <strong>upskilling, governance charters, training and role redesign</strong>, rather than headcount reduction. </p></li><li><p><strong>The biggest investments are in unifying internal data</strong> across partners, platforms and functions.</p></li><li><p>As data becomes marketing’s core asset, the <strong>CMO’s remit is widening into facilities, operations and client experience</strong>.</p></li><li><p>In more highly regulated sectors, <strong>compliance issues are slowing AI adoption</strong>. Here, firms tend to favour proprietary infrastructure and human oversight.</p></li><li><p>AI has delivered speed and scale in content production but <strong>not (yet) the expected step-change in ROIs</strong>.</p></li></ul>.<p>Marketing teams are adding AI and analytics tools faster than they can measure their value. At the same time, CMOs are being asked to take responsibility for growth, customer experience and brand trust, even when they do not control all the systems or teams involved. At a recent open-house discussion, members of the India CMO Forum compared where AI spending is going, whether their MarTech stacks are becoming simpler and how far the CMO’s responsibilities now extend beyond marketing.</p><h2><strong>The Uneven Arithmetic of AI Investment</strong></h2><p>Most organisations represented had not created a separate AI budget. They were funding it through existing MarTech, marketing or digitalisation allocations. Some firms were increasing their MarTech budgets to absorb AI costs and measuring the returns through indicators such as cost per piece of content or total output. Others were approving AI spending case by case, without a fixed annual allocation. Regulated and data-sensitive firms were the main exception. These organisations were more likely to hold AI budgets centrally and to require each proposed use-case to meet a formal business need. The discussion indicated that budgeting discipline depends heavily on data sensitivity and regulatory exposure.</p><p><strong>From Tool Sprawl to Tool Consolidation</strong><br>MarTech stacks have traditionally consisted of several specialised tools, each covering a narrow task. This is starting to change with AI-native platforms that can now combine functions such as copy editing, research and content generation that previously required separate licences. Some organisations are also bringing social listening and reputation monitoring in-house instead of relying on several external agencies for separate reports. However, fewer vendors does not necessarily mean simpler operations. Rather, the responsibility for maintaining data quality, testing model output and managing reliability has moved inside the organisation.</p><h2><strong>Building New Capabilities</strong></h2><p>None of the organisations represented at the meeting say they are reducing headcounts because of AI. Most were instead investing in training, governance and changes to existing roles. One approach discussed was an AI Governance Charter, backed by the CEO, with regular training for the leadership team and dedicated training budgets across levels of the organisation. Another organisation reported that almost its entire workforce has received AI training, supported by weekly sessions and governance councils at both the organisational and business-line levels. The immediate effect has been higher output expectations from broadly unchanged teams. Role descriptions are also changing. Copywriters, for example, are spending less time producing first drafts and more time on content strategy, editing and quality control.</p><h2><strong>Data as the Precondition for Intelligence</strong></h2><p>AI tools only become useful when the organisation can combine and trust its internal data. One firm replaced an external PR measurement retainer with an internal tool built in three weeks. The tool tracks media mentions and share of voice, then feeds the information into sales and business-development systems. This has helped the company identify early signals of global capability centres (GCCs) exploring entry into Indian markets. The same organisation is combining data on space, HVAC and Wi-Fi use across client facilities to improve personalisation. In both examples, the AI application depends on connecting data that had previously been held by different teams, partners and platforms.</p><h2><strong>Where the Mandate Now Extends Beyond Marketing</strong></h2><p>Marketing is being drawn into areas such as customer experience, facilities and operations because it increasingly holds the data used across these functions. Marketing performance is no longer measured only through awareness and demand generation. It now includes customer experience and retention. This may place the CMO in a strong position to connect customer data across the organisation, even when formal responsibility for the underlying activity sits elsewhere.</p><h2><strong>Regulation as a Constraint</strong></h2><p>In highly regulated sectors, customer data cannot easily be sent to public large language models (LLMs). Firms are therefore building proprietary systems using open-source frameworks. The expected move towards consent-based data architecture under the Digital Personal Data Protection (DPDP) Act strengthens the case for this approach. Much of the AI budget in these organisations goes into processing infrastructure and model development before any customer-facing application is launched. Human review also remains necessary because confidence in model output is not yet high enough to remove supervision. For these firms, the pace of adoption is determined primarily by governance, security and data sensitivity. Class-leading organisations treat data architecture as part of marketing operations rather than as a separate IT issue. Tools, budgets and governance structures will continue to change, but the basic requirement will remain the same: firms need control over their data before they can automate decisions with confidence</p>
