Home · Business · Sales · Prospecting & outreach
As of 13 August 2026, AI can only partly decide whether to contact a prospect.
Most people should hand this to a purpose-built tool.
Can you do it?
5 minutesto a draft.
30 minutesto something you’d act on.
Cost, all in£0
Skill neededchat-fluent
Who has to check ityou
What the alternative costsThe alternative is a salesperson making the decision from CRM records, account research and previous conversations.
If this goes wrong: you contact someone at the wrong time or with the wrong reason, damage the relationship and lose a potential opportunity.
What to actually do
Use a tool built for this
The route this page recommends
Do it yourself
Second choiceA chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.
How to actually do it
- Open your CRM record for the prospect and write down the contact's role, account, previous interactions, stated needs, timing and any contact restrictions.
- Open the account's current public information and copy only relevant, sourced facts about its situation, separating facts from your assumptions.
- Define the commercial objective and the specific reason this prospect might benefit from your offer.
- Paste the information into the prompt, keeping personal data to the minimum needed for the decision.
- Compare the recommendation's factual claims against the CRM record and source pages, removing any claim the sources do not support.
- Choose contact now, wait or do not contact based on the checked evidence, and record the reason and next review point in your CRM.
- If you choose contact now, use the checked reason to prepare and send an appropriate message through your normal sales process, rather than allowing the chatbot to contact the prospect.
Prompt
Decide whether I should contact this prospect now, wait, or not contact them based only on the information below. My objective: [what I want from the contact] My offer: [what I sell and who it is for] Prospect and account information: [paste verified facts] Prospect's role and likely responsibilities: [paste verified facts] Previous interactions: [paste dates, channels and outcomes] Known needs, priorities or problems: [paste evidence, separating facts from assumptions] Timing, budget or buying process: [paste what is known] Contact preferences, exclusions or consent restrictions: [paste what is known] Relevant market or company changes: [paste sources or facts] Return: 1. One recommendation: contact now, wait, or do not contact. 2. The three strongest reasons, labelling each as fact or inference. 3. Any evidence that argues against contacting them. 4. The most important missing information. 5. A proportionate next action if contact is recommended, without drafting the message. 6. A short list of claims I must verify before acting. Do not invent facts, buying intent, consent, urgency or personal details. Do not treat company size, job title or recent public activity as proof that the prospect wants contact. If the evidence is weak or conflicting, say so plainly. Do not contact anyone or make a decision on my behalf.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
Hand it to a person
The distant thirdA person who owns the outcome does this end to end, worth it when the failure is dear.
What it gets wrong
- AI cannot see the informal relationship history, internal politics or tone that a salesperson may know from working the account.
- AI cannot establish genuine buying intent from a job change, website visit or other weak signal.
- AI cannot know whether contacting this person fits your team's account strategy or the relationship another colleague is managing.
- AI can turn assumptions in your notes into confident-sounding reasons unless you separate facts from inferences.
- The recommendation does not carry responsibility for a damaged relationship or a missed opportunity.
What caps this at PARTLY: judgement under ambiguity, context depth and stakes of error.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 1 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 1 |
| Total | 6 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI tell me which prospects to contact?
- Partly. It can rank or group prospects using the information you provide, but it cannot reliably establish buying intent or understand relationship context that is missing from the data. Check its factual reasons before acting.
- Can AI score my sales leads accurately?
- It can apply a consistent scoring method to your CRM fields and research, but the score is only as sound as those inputs. Treat it as a recommendation, not proof that a prospect is ready to talk.
- Should I let AI contact prospects for me?
- No, not without your review. AI can help prepare a reason for contact, but you should check the recipient, timing, claims and contact restrictions before sending anything.
- What information does AI need to decide whether to contact a prospect?
- Give it your objective, offer, verified account and role information, previous interactions, known needs, timing, buying process and any contact restrictions. Mark assumptions clearly and do not provide more personal data than the decision requires.
Nearby answers
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.