AI New Year's Resolutions for Professional Services Firms

It's New Year's Eve. If I could give your firm one gift—beyond time, money, and talent—it would be a set of realistic, achievable resolutions for AI in 2026. Not aspirational stuff like "transform with AI." Specific things that actually matter.

Resolution 1: Measure One Thing Well

Pick one AI use case that's already deployed. Measure its impact. How much time does it actually save? Is that time being converted to billable work, internal projects, or just slack? Is the quality better or the same? What's the real ROI?

Don't try to measure everything. Pick one. Get the data right. Share the results internally. This single resolution, if done well, will inform every AI decision you make in 2026.

Commitment: Allocate 40 hours from someone who knows your business and can pull data. Have numbers by March 1st.

Resolution 2: Document Your AI Governance

You probably have some governance around AI already. It might be informal, it might be scattered, but it exists. Document it. What AI tools are approved? What data can they see? Who approves AI use? What's the audit trail? How do you handle errors?

Once it's documented, share it with your insurance broker and outside counsel. Get their feedback. Fix the gaps. Then enforce it.

Commitment: 20-page governance document by February. Review and update quarterly.

Resolution 3: Start One Serious AI Hiring Initiative

You can't do AI strategy well without people who understand both AI and your business. Hire one person (could be an existing person redeployed) whose job is AI strategy and implementation. Not "manage the AI tool." Strategy and implementation. Someone who knows enough about AI to make good choices and enough about your business to make relevant ones.

Commitment: Hire or redeploy one AI-focused person by March 31st.

Resolution 4: Clean Up One Dataset

Pick the dataset that would be most valuable if it were clean: your CRM data, your project data, your financial data. Whatever. Spend 2–3 weeks cleaning it. Remove duplicates, fix inconsistent entries, fill gaps. Just one dataset, not everything.

This one dataset will become your foundation for better analytics and better AI. And you'll learn hard lessons about data quality that will inform your next choices.

Commitment: One complete dataset audit and cleanup by end of Q1.

Resolution 5: Have the Conversation About AI and Pricing

Don't charge for it yet. But have the conversation internally. If AI makes you 20% faster, what do you do with that? Cut prices? Increase profit? Invest in quality? The answer shapes 2026 strategy. Make the choice deliberately, not accidentally.

Commitment: Internal pricing discussion and decision by February. Document the decision.

Resolution 6: Run One Experiment You'll Learn From

Pick something ambitious but not critical. Deploy an AI system that could fail without destroying anything. A new research tool. An internal assistant. A client communication system. Run it for 6 weeks. Measure. Learn. Kill it or scale it.

The goal isn't success. It's learning. You'll learn more from one serious experiment than from reading ten articles about AI.

Commitment: Identify, launch, and conclude an experiment by end of Q2.

Resolution 7: Get Your Partners on the Same Page

If you're a partnership, you probably have some disagreement about AI strategy. Some partners think it's the future. Some think it's overhyped. Most are undecided. In 2026, clarify the position. Don't need perfect agreement, but need clarity on: Are we investing in AI? How much? What's the tolerance for risk? What's the timeline?

This resolution is harder than technical ones. But it's more important. You can't execute strategy without partnership alignment.

Commitment: Leadership offsite or discussion in January. Document the position.

Resolution 8: Build One Knowledge Base People Actually Use

Not a wiki with 500 pages nobody reads. One focused knowledge base: your most common client questions. Your standard workflows. Your best practices. AI-powered search. Real humans using it because it actually helps them.

Start small. 20–30 core documents. Get it right. Then expand.

Commitment: Minimum viable knowledge base launched and in use by April 30th.

Resolution 9: Stop Doing One Thing to Make Room for AI

If you're going to invest in AI, something else has to give. A meeting. A process. A report. A system. Identify what. Kill it. Use that bandwidth for AI work.

This is harder than it sounds. It requires saying no to something that's currently happening. But it's necessary. You can't add AI without subtracting something else.

Commitment: Identify and eliminate one time commitment by end of Q1.

Resolution 10: Know Your AI Spend

How much is your firm actually spending on AI? Tools. Services. Time. Salary for people working on it. Most firms have no idea. They spend $3K on a tool, $50K on a consultant, 200 hours on experimentation, and someone's full-time salary on AI work. Total: $150K. They think they're spending $3K.

Know your real spend. It's not too much or too little until you know what it is.

Commitment: Full AI spend audit by March. Track it monthly going forward.

Why These Resolutions?

These aren't sexy. They're not "implement agentic AI" or "revolutionize with AI." They're basics: measure, govern, hire, clean up, think strategically, experiment, align, build knowledge, focus, and track spend.

But these basics are what separate firms that do AI well from firms that waste money on it. Every successful AI implementation I've seen started with these fundamentals.

The Resolution Behind the Resolutions

If you could only pick one resolution, pick this: Treat AI strategy like you treat any other business strategy. Think carefully. Make deliberate choices. Measure results. Adjust. Don't get swept up in hype. Be skeptical of easy answers. Invest steadily. Learn consistently.

Do that, and 2026 will be a good year for AI in your firm.