Claude prompts for sales, by role

Here is a job description {company} posted this week:
{paste}

I sell {product} to {buyer}. Read the posting for what
it implies about their priorities and constraints, not
just the listed skills.

Give me three opening lines. Each must reference
something specific in the posting. No compliments,
no "I noticed you're hiring". Under 20 words each.

Job posts describe a problem someone got budget to solve. That is a better trigger than funding news, which every tool surfaces at once.

Here are the four emails in my current sequence:
{paste}

Rewrite each one so the subject is the prospect's
situation, not my product. Rules:
- First sentence must be about them
- One idea per email
- One ask, at the end
- No em dashes, no "hope this finds you well"
- Keep each under 90 words

After the rewrites, tell me which of the four you
would delete entirely and why.

Asking which email to delete is what gets you an honest answer. Models will happily improve four emails that should have been two.

Our ICP: {criteria}
Disqualifiers: {criteria}

Here are 25 accounts with firmographics:
{paste}

Score each 1 to 5 against the ICP with one line of
reasoning. Anything scoring 3 or below, tell me the
single fact that would have to be different for it
to be a 5. Do not pad the list to fill my quota.

The last sentence stops the model rating everything a 4 to be helpful.

I emailed {contact} twice about {topic}. No reply.
Here are both emails:
{paste}

Write a third that does not mention the earlier ones,
does not ask if they saw my last email, and gives them
a reason to reply that did not exist two weeks ago.
If you cannot find one in what I gave you, say so
instead of inventing urgency.

Most bump emails fail because they add nothing. Making the model admit when there is nothing new is more useful than a fourth touch.

Notes from my last two calls with {account}:
{paste}

Prepare me for tomorrow. I want:
1. What they said they cared about, in their words
2. What I promised and whether I delivered it
3. Anything unresolved
4. The two questions most likely to open the call
5. One thing I have been avoiding asking them

Quote the notes directly. Do not smooth over
contradictions between the two calls.

Point 5 surfaces the deal risk. Everything above it is a summary a rep could have written.

Here is everything I have on {account}:
{paste}

Argue that this deal will not close this quarter.
Be specific and use only what is in the notes.
Then tell me which single piece of missing
information would settle the argument, and give me
the exact question to ask to get it.

Inverting the prompt gets past the model's default optimism. Ask it to prove the deal is dead, not to assess it.

Transcript from discovery with {account}:
{paste}

Build a mutual action plan: every step from here to
signature, with the owner on their side, the owner on
ours, and a date. Only include steps implied by what
was actually said. Mark anything you inferred with
[assumed] so I can check it with them.

The [assumed] tag is the difference between a plan you can send and a plan you have to rewrite.

Here are 20 questions from {account}'s security
review and our security documentation:
{paste}

Draft an answer to each using only the documentation.
Where the documentation does not cover a question,
write NEEDS INPUT and name who should answer it.
Never approximate a compliance claim.

The hard constraint on compliance claims matters more than the drafting speed. Get this one wrong and it is a legal problem, not a productivity one.

Transcript of {rep}'s discovery call:
{paste}

Give me: talk-to-listen ratio, the three highest
value questions they asked, the two moments they
should have asked a follow-up and moved on instead,
and one habit worth working on this month.
Quote the exact lines. Be specific enough that the
rep can hear themselves in it.

Generic coaching feedback gets nodded at and ignored. Quoted lines get argued with, which is where the coaching happens.

Closed-lost reasons and notes from the last 20
deals:
{paste}

What is the actual pattern? Ignore the reason field,
which reps fill in to close the record, and read the
notes. Tell me what stage things really died at and
what was true in the ones we lost that was not true
in the ones we won.
If the sample is too small to conclude anything, say
that.

The closed-lost picklist is the least reliable field in any CRM. The notes are where the truth is.

Territory: {definition}. Rep: {name}, ramping.
Here is the account list with firmographics and
history:
{paste}

Write a one page brief: the ten accounts to work
first with a reason each, the three that look good
but are not, and the two things about this territory
a new rep would take six months to learn.

The 'looks good but is not' section is the part that saves a quarter.

Account: {name}. Renewal in {days} days.
Usage summary, support tickets, and CRM notes:
{paste}

Tell me honestly whether this renews. Separate what
the data says from what the relationship says, and
say which one you trust more here and why.
Then give me the agenda for the call, opening with
the hardest topic rather than the easiest.

Opening with the hardest topic is the instruction that changes the agenda. Left to default, the model writes a friendly check-in.

Here are 6 months of tickets and usage for
{account}:
{paste}

Where are they working around a limitation that
{product feature} would solve? Cite the specific
tickets. If the honest answer is that they are fine
as they are, say that instead of building a case.

Expansion prompts produce fiction unless you give the model permission to say no.

Usage, support, and outcome data for {account}
this quarter:
{paste}
Their stated goal at the start of the quarter:
{goal}

Write the QBR narrative in three parts: what they
set out to do, what actually happened including
where we fell short, and what changes next quarter.
Lead with the shortfall. No slide filler.

Leading with the shortfall is what makes the rest of the QBR credible.