40 Questions Business Development Managers Should Ask Before Trusting AI Prospect Research
Your instinct about which deals matter comes from years of reading people and markets. When Apollo.io flags a prospect or ChatGPT writes your outreach sequence, you need to know whether the AI has actually found a real opening or simply assembled plausible patterns. These questions help you separate useful AI outputs from confident nonsense.
These are suggestions. Use the ones that fit your situation.
1When Apollo.io identified this prospect, did it find evidence of a business problem they actually have, or just keywords that suggest they might?
2Does the AI research explain why this person moved to their current company, or just confirm they work there now?
3Has the AI noted any recent company restructuring that might affect who owns the decision you're chasing?
4Does the prospect research include evidence of budget allocation or buying behaviour, or only job title and company size?
5When you read what the AI found about this person's background, does it suggest they would care about your solution, or does it just prove they exist?
6Has the AI identified any mutual connections or shared experiences that might create actual credibility with this prospect?
7Does the research show whether this prospect has bought similar solutions before, or is that information missing?
8Has the AI research captured any public statements or interviews from this person that reveal how they think about problems in your space?
9Does the output explain the relationship between the person you want to reach and the actual budget holder, or does it assume they are the same?
10When the AI compiled this research, did it distinguish between a prospect who fits your ICP and a prospect you actually have a reason to call?
Outreach and Messaging
11When ChatGPT wrote this outreach sequence, can you identify a specific insight about this prospect's situation, or does the copy work for almost anyone in their industry?
12Does the AI-generated email reference something real about this person or company, or only general pain points that apply to hundreds of prospects?
13Would you be willing to attach your name to this opening message as written, or would you edit it to sound more like you?
14Does the proposed sequence account for the fact that this prospect might already be aware of your solution, or does it assume they have never heard of you?
15When the AI drafted your call script, did it include a reason for the prospect to talk to you right now, or just a description of what you sell?
16Has the AI suggested a specific business outcome you can promise, or only generic benefits that your competitors also claim?
17Does the outreach message match the voice and approach that has actually converted deals for you in the past?
18Did the AI consider your prospect's current priorities based on recent company news, or does the sequence assume their situation is static?
19Does the proposed messaging explain why you are calling this person specifically, rather than someone else at their company?
20When you read this AI-written copy, does it create genuine curiosity, or does it feel like a template that has been personalised at surface level only?
Opportunity Assessment
21When Salesforce Einstein flagged this deal as high probability, did it identify the actual compelling event that might make them buy, or just correlate it with deals that closed before?
22Does the AI assessment explain who in this organisation might lose credibility if they don't solve this problem, or does it only quantify the business impact?
23Has the AI analysis confirmed that your prospect has a process for buying something like this, or is it assuming they will create a buying process for you?
24When you compare the AI's opportunity score to your own sense of the deal, where do you disagree, and what does that disagreement tell you?
25Does the AI assessment include evidence that the prospect is actually ready to move forward, or only that they fit your customer profile?
26Has the system identified any technical or political obstacles that might block this deal, or is it projecting only the upside?
27Does the AI analysis distinguish between a prospect who is exploring options and a prospect who has already decided to move, or does it treat both the same?
28When the AI estimated deal size, did it account for how much budget this prospect would actually allocate, or did it assume your standard contract value applies?
29Has the opportunity assessment included any red flags about this prospect's organisation or history, or only the factors that suggest they should buy?
30Does the AI prediction account for your own track record with prospects that initially scored the same way, or is it applying a generic probability model?
Relationship and Negotiation
31When LinkedIn Sales Navigator suggested an outreach angle based on shared interests, have you actually verified that this person cares about that topic, or only that they followed a relevant account?
32Does the AI research reveal anything about how this prospect prefers to be approached, or only the professional channels where they are visible?
33Has the system identified any previous failed attempts to reach this prospect or their company that you should know about?
34When the AI suggests a partnership strategy, does it account for the reality of how decisions actually get made at this company, or is it building strategy on incomplete information?
35Does the AI analysis include what this prospect might ask you in return, or only what you want from them?
36Has the system identified whether this prospect operates as a relationship builder or a transactional operator, and adjusted your approach accordingly?
37When Perplexity compiled information about this person's company, did it surface any recent departures or internal conflicts that might affect your negotiation position?
38Does the AI suggest a price or terms position based on this specific prospect's leverage, or just your standard model?
39Has the system identified any secondary relationships at this company who might influence the decision in your favour?
40When you evaluate the AI's assessment of this relationship, can you distinguish between what the system knows and what it is guessing about how this person actually operates?
How to use these questions
If the AI research reads like it could describe five different prospects in the same industry, you do not have enough insight to call yet. Your deal instinct should kick in here and tell you to spend more time before you dial.
Test your AI tools against deals you have already closed. Go back and run the same prospect through Apollo.io or LinkedIn Sales Navigator and see whether the system would have actually surfaced the real reason they bought from you. This tells you what each tool is actually good at and where it fails.
When an AI sequence feels impersonal, your prospect will feel it too. Before you send anything, ask yourself whether a prospect who already knows your brand would believe you actually researched them or whether they would recognise the template underneath.
Your judgement about which opportunities are real matters more now, not less. The AI can find prospects and write copy faster than you can. Your value is in knowing which prospects are actually worth your time and which deals have real momentum. Do not let the tools replace that instinct.
Create a working rule: if the AI found something, you verify it independently before you act. Use ChatGPT to draft your research question, then use Perplexity to cross-check what it found. This extra step catches the confident mistakes that AI tools make when they confuse pattern matching with actual insight.