This photo is from a wonderful bookstore I visited earlier this year. I felt that it was the perfect image for this work. Wealth is everywhere, but its value depends on finding the right thing, in the right place, with enough context to trust.
The martech world may have finally reached “peak SaaS.” Looking to the outlook for chief martech in 2026, Scott Brinker points out that while the underlying market has been volatile, the number of commercial martech products has barely increased. What’s even more interesting is that martech claims to be moving from human-operated apps to infrastructure that can be safely used by AI agents. In that world, the question is no longer just whether a platform has a good interface. What matters is that AI systems have access to the right data, permissions, workflows, and guardrails at the right time.
This change should feel very familiar to law firm marketing and business development teams.
For many years, legal CRMs have been treated as systems of record, but more often as systems of partial records, wishful remembrances, and occasional clean-up panics. Businesses have invested in CRM platforms, marketing automation, experience databases, email tools, event systems, and now AI. But the difficult problem wasn’t the existence of another tool. Apparently the meeting wasn’t enough. Software dissemination was also necessary.
The difficult issue is trust.
Can a company trust relationship data? Can a company trust that contact companies, job titles, locations, business relationships, and relationship history are accurate? Can a company trust that the list is clean? Can a company trust that business development activities reflect the company’s actual knowledge of the client? Can it trust that the right people are reviewing the right exceptions before questionable data becomes institutional truth?
AI makes these questions more urgent, not less.
And this is not just a concern for the legal industry. B2B marketing leaders are recognizing that AI cannot repair fragmented, inconsistent, incomplete, or outdated data. It amplifies it. Poor segmentation leads to inefficient scaling. Weak signals create a false sense of confidence. A broken system becomes a faster broken system. The stakes are even higher in the legal field, as data is not just commercial. It involves relationships and reputation, and is often sensitive.
Companies that succeed in AI-enabled business development will be those that treat governance as an operational layer beneath CRM, rather than a clean-up project that happens once every few years.
Human users often look at messy CRM data and make decisions. AI systems may not. It is possible to operate confidently on outdated, duplicated, incomplete, or incorrectly connected information. It has the potential to generate sophisticated recommendations from weak context. We may suggest outreach based on data that appears structured but is actually unmanaged. It’s not artificial intelligence. It’s automatically bewildered by better grammar.
This is why governance is no longer a side process. Governance is infrastructure.
In traditional CRM environments, governance often meant regular cleanup. So whether it’s a practice group project, a pre-event push, a one-time data quality effort, or a heroic spreadsheet circulated by people who deserve better. In an AI-enabled environment, that model breaks down. Businesses don’t need to perform data hygiene every once in a while. There needs to be ongoing governance under the systems that lawyers, marketers, business development teams, and ultimately AI agents rely on.
This means data quality, permissions, exceptions, evidence, workflow, and company-specific rules need to exist as a living layer beneath the CRM. It is not a policy document. It’s not a quarterly guilt exercise. infrastructure.
This is a process that many companies are beginning to take. The point is not to replace the judgment of experienced marketing and business development professionals, nor to pretend that AI can magically and uniquely understand a company’s relationships, priorities, and sensitivities.
The more valuable work is the more basic. This means creating a managed data layer that will make future AI use cases safer, more useful, and more reliable.
This means thinking of your CRM as an infrastructure rather than a static database. Relationship data must be maintained over long periods of time. Problems need to be identified on an ongoing basis. Company-specific rules must be applied consistently. Exceptions should be migrated through structured workflows rather than informal workarounds. Evidence matters. Responsibility is important.
This is a lot of what we focus on at Cirrom, but it’s more important than any one platform. Companies that succeed in AI-enabled business development will be those that treat governance as an operational layer beneath CRM, rather than a clean-up project that happens once every few years.
The same principle applies to AI. AI can generate suggestions. I can summarize it. Can be drafted. You can guess it. However, in a legal CRM, the AI does not automatically recognize company-specific criteria, relationship history, customer sensitivity, or which records are actually important.
These outputs must be grounded in reliable data, clearly defined rules, evidence, workflows, and accountability before they can be executed with confidence.
That distinction is important. The future of legal CRM is not “AI not governance.” It’s AI built on governance.
But when governance becomes infrastructure, CRM modernization can become more strategic. CRM becomes a trusted context layer for business development. Increases the credibility of your marketing list. Easily maintain customer and prospect data. The exception would be workflow, not folklore. AI becomes more operationally capable than just a shop trick.
The legal industry tends to approach new technologies with a good mix of interest, caution, committee, and exhaustion. That caution is not necessarily wrong. Law firms deal with sensitive relationships, regulated communications, confidential matters, and reputational risks. But caution doesn’t mean waiting until AI is magically safe.
That won’t happen.
Companies that gain value from AI in business development are not just those who adopt the latest tools or create the smartest prompts. These companies will be the ones that invest in the infrastructure underlying AI: managed data, clear rules, strong workflows, trusted context, and systems that understand how the company actually operates.
AI will not fix legal CRM.
This will reveal whether a legitimate CRM is trustworthy or not.
