The AI Questions Every PI Managing Partner Should Ask Before Approving the Budget

Your marketing director wants to invest in AI. Your intake manager is asking about AI-powered chatbots. A vendor just pitched “AI-driven lead scoring.” And you're sitting across the table from all of it, trying to figure out which conversations are worth having and which are just technology theater.

This guide is written specifically for managing partners at personal injury firms. Not for marketing directors who already understand the landscape. For the partner who needs to ask the right questions before approving budget — and who doesn't want to get sold something that sounds impressive but doesn't move the firm forward.

Question 1: “What specifically does this AI do — and what does it not do?”

This is the most important question you can ask, and most vendors won't answer it directly. “AI-powered” is a marketing term. It covers everything from a basic rules engine to genuine machine learning that improves with data. You need to know which one you're buying.

The follow-up is equally important: what does it not do? Honest vendors tell you the boundaries. Vendors overselling their technology either can't answer or pivot to another feature. The boundary between genuine AI capability and marketing language is where the real due diligence happens.

For PI-specific context: the most credible AI applications in your market right now are in performance anomaly detection (catching CPL spikes before they become expensive), predictive signed-case forecasting, and budget reallocation recommendations. These are data-intensive tasks where AI genuinely outperforms manual analysis. AI chatbots replacing intake specialists is a much more contested claim — and the conversion rate data on fully automated intake is not flattering.

Question 2: “How does this connect to cost per case — and how long until I see results?”

The only number that matters in PI marketing is cost per signed case. Not cost per lead, not click-through rate, not “AI confidence scores.” Before approving any AI investment, ask the vendor to draw a direct line from their product to your cost per case.

If the answer is vague — “it improves your overall marketing efficiency” — push harder. The question is: which specific mechanism reduces cost per case, by how much, and in what timeframe?

The honest timeline for AI-driven marketing improvements at a PI firm is 60–90 days. The first 30 days are integration and baseline calibration. The next 30 days surface the first optimization opportunities — typically a vendor underperforming that you hadn't caught, or a budget allocation that's demonstrably off. By day 90, you should have clear cost-per-case data that either justifies continued investment or tells you it's not working.

Question 3: “Is our data good enough for this to work?”

AI is only as good as the data feeding it. And most PI firms, when they honestly audit their data, find more gaps than they expected.

The specific question is whether your lead records carry source tags that follow them through intake to signed cases. If a lead from Google Ads comes in, gets processed by intake, and becomes a signed case — can you connect those three events in one record? If the answer is “not consistently” or “we'd have to check” or “we use spreadsheets to reconcile” — that gap exists before you add AI, and the AI will surface it immediately.

Credible AI vendors will do a data audit before they promise outcomes. If a vendor is willing to promise ROI without first understanding your data infrastructure, treat that as a red flag. The firms that get value from AI investments fast are the ones that know their data state going in.

Question 4: “Who manages this — and what does it cost my team?”

AI platforms require an owner. Not necessarily a dedicated hire, but someone who reviews alerts, acts on recommendations, and brings the data to partner meetings. For most PI firms, that's the marketing director. The question is whether the platform is designed to fit into their existing rhythm or to add to it.

The right answer: a well-implemented AI platform reduces your marketing director's reporting burden significantly. The firms seeing 15 hours per week of manual reporting drop to 15 minutes aren't getting that from a more capable marketing director — they're getting it from connected data infrastructure that automates what used to be manual.

Ask the vendor how many hours per week the platform requires to maintain after the first 90 days. A credible answer is 30–60 minutes of active review per day. If they can't answer, or if the answer is “it depends on how much you want to get out of it,” that tells you the platform is not as automated as the demo suggested.

Question 5: “What does success look like at 90 days, and how do we measure it?”

Managing partners approve investments based on expected outcomes. The problem with most technology investments is that “success” is never defined at the start, which makes it impossible to evaluate at the end.

Before approving any AI investment, define the success criteria explicitly and put them in the contract if possible. Specific metrics for a revenue intelligence platform:

• Cost per signed case visibility across all active lead sources within 60 days of implementation
• At least one vendor optimization decision (scale or cut) based on platform data within 90 days
• Weekly reporting time reduced by at least 50% from current baseline within 60 days
• Managing partner able to answer “what is our cost per case from Google Ads this month?” without asking marketing to pull a report

These are concrete, measurable, and directly tied to firm outcomes. Any revenue intelligence vendor who resists defining success this specifically is telling you something important.

Question 6: “How is this different from what we're already getting from our vendors?”

Lead vendors provide reports. Google Ads provides performance data. Facebook gives you reach and click metrics. Why pay for another layer of analytics on top of that?

The answer comes down to incentives and data scope. Vendor-reported performance data is not independent. Every vendor has an incentive to present their numbers favorably. They report on leads delivered, not on signed cases produced. They don't have visibility into your settlement data.

An independent attribution layer — one that pulls spend data from vendors and connects it to signed case data from your CRM — is the only way to produce cost per case numbers that are free from vendor influence. That independent calculation is what you bring to a vendor negotiation. “Your cost per case, according to our data, is $4,200. The market rate for this type of lead is $2,800. Here's the data.” That conversation is very different from accepting whatever the vendor tells you.

Question 7: “What happens if we don't do this?”

This is the question managing partners rarely ask, but it's the most useful one. The cost of inaction is real — it just shows up in ways that are harder to see than a line-item expense.

The firm spending $300,000 per month on lead generation without independent cost-per-case attribution is making budget decisions based on vendor-reported data and gut feel. If 20% of that budget — $60,000 per month — is going to sources delivering cases at twice the market rate, that's $720,000 per year in recoverable waste. Not incremental improvement. Recoverable waste that shows up as margin if you measure it.

The firms that will own their markets in three years are the ones building attribution infrastructure now — when it creates a compounding advantage over competitors still operating on spreadsheets and vendor reports. The first-mover advantage in PI marketing data is real and growing.

How to Frame the Investment Decision

Revenue intelligence is not a marketing expense. It's a cost recovery investment. The frame that resonates with most managing partners is this: you're already spending the marketing budget. The question is whether you know which of that spending is working.

For a firm spending $200,000 per month on lead generation, recovering 15% in misallocated spend through better attribution produces $30,000 per month in redirected budget. That's $360,000 per year in efficiency gains — from the same total spend, just pointed at sources that produce cases at better economics. Most revenue intelligence platforms cost a small fraction of that recapture figure.

The question isn't whether you can afford the platform. The question is whether you can afford to keep making $200,000 monthly decisions without the data to make them well.

If you're ready to see what cost-per-case visibility looks like across your vendor portfolio, explore the Marketing ROI platform or schedule a callwith a team member who works exclusively with PI firms at your spend level. The conversation starts with your current data state — not a product demo.