Prompt & Tell: Share Your Best RevvyAI Prompt to Earn a Reward!

Lauren Triance-Haldane

RevvyAI is now generally available and helping revenue teams move faster by turning questions into answers (and actions) in a single prompt. 

Whether you’re using it for pipeline insights, account research, campaign performance, or outreach, we want to see what’s working for you. 

Drop your best RevvyAI prompt and a short sentence on why you’ve found it useful, effective, or repeatable in the comments below 👇 before 11:59 EST on Friday, August 21! All commenters will earn a $25 reward and be eligible for a complimentary Breakthrough pass.

Share with us: 

• A prompt you rely on regularly that has delivered impactful insights for your team 
• A prompt that saved you time 
• Something that helped you uncover a new insight or workflow 

 What’s in it for You? 

• Everyone who shares will receive a $25 reward via email within 24 hours 
• You’ll be eligible for a complimentary Breakthrough pass (Nov 2–5, Fontainebleau Las Vegas) 
• Sharing is caring and you’ll get the sweet satisfaction of helping others! 

If you haven’t played around much with RevvyAI, here are some prompts to help you get started:  

• “Show me hot accounts not yet touched by sales” 
• “Rank my campaigns by new 6QAs generated” 
• “What should I know before my next call with [account]?” 

Then build on them — refine, iterate, and see what else RevvyAI can surface.  

Your prompt might be someone else’s breakthrough — share it below by end of day August 21, 2026 to earn your reward. 
 
Read the full contest rules and regulations. 

Claire Couch

Excited to see the submissions of prompts @Lauren Triance-Haldane!

Vinicius Scott

Top 100 pages by account engagement, for any segment and any date range

I run this once per segment when I pull monthly content performance. It returns a verbatim table I can drop straight into analysis with no reformatting or cleanup to create content engagement reports

SEGMENT_ID: {{SEGMENT_ID}}

SEGMENT_NAME: {{SEGMENT_NAME}}

SEGMENT_URL: {{SEGMENT_URL}}

DATE_START: {{DATE_START}}

DATE_END: {{DATE_END}}


You are pulling data for the segment specified above.


TASK

Return the Top 100 website pages visited by accounts in SEGMENT_ID for the date range DATE_START through DATE_END (inclusive on both ends).


OUTPUT FORMAT

A single markdown table with exactly three columns and exactly 100 data rows (no fewer), ordered by account count descending:


Position | URL | Account Count

1 | [full URL] | [integer]

2 | [full URL] | [integer]

... | ... | ...

100 | [full URL] | [integer]


RULES: read carefully, these are non-negotiable


1. Use the exact DATE_START and DATE_END values from the edit block above. Do not substitute a default range (30-day, 90-day, "last month," etc.). Both endpoints are inclusive. If the UI rounds or adjusts the range, confirm the exact start and end dates you applied in the confirmation line below.

2. Use the exact SEGMENT_ID from the edit block above. Apply only the filters that are part of that segment's definition (as described in SEGMENT_NAME). Do not widen, narrow, or modify the segment scope. If you cannot resolve the segment, stop and report the error instead of substituting a different segment.

3. Return full URLs exactly as they appear in 6sense. Do NOT shorten, normalize, strip query strings, collapse trailing slashes, or merge variants (e.g., keep /foo/ and /foo as separate rows if 6sense lists them separately). Keep regional subdomains intact (xx.example.com, yy.example.com, etc.).

4. Do NOT deduplicate, consolidate, or group URLs. One row per URL string as 6sense returns it.

5. Do NOT classify, categorize, label, or annotate URLs. No campaign names, no "blog" / "editorial" / "generic" tags. URLs only.

6. Account count is the unique account count for that URL in the date range. Integers only. Do not round, bucket, or display ranges.

7. Tie handling: when two or more URLs share the same account count, order them alphabetically by URL within that tie group, and continue position numbering sequentially (1, 2, 3, …) with no skipped positions and no shared positions.

8. If the segment has fewer than 100 pages with engagement in the date range, return all available rows and add a one-line note below the table stating the actual count (e.g., "Note: only 87 pages had engagement in this date range").

9. If the segment has more than 100 pages, return exactly the top 100 by account count. Do not truncate the table mid-row, do not insert "..." rows, do not summarize the long tail.

10. Do not include any prose commentary, takeaways, summaries, or analysis. The table and the confirmation line are the entire output.

11. Do not exclude bot traffic, internal IPs, or any URLs unless 6sense applies those filters by default at the segment level. If it does, list the filters applied in the confirmation line.


CONFIRMATION LINE (required, above the table)

"Date range applied: [actual start date] to [actual end date] ([N] days). Segment: [SEGMENT_ID] ([SEGMENT_NAME]). Filters applied: [list any, or 'none beyond segment definition']."


Begin.

Lily Adkison

I've been impressed with the keyword analysis capabilities in RevvyAI, and love that it actually can take actions on my behalf while running analysis.

As part of our keyword update for this year, I ran "review the full keyword library and evaluate effectiveness, provide recommendations for improving keyword and targeting" and was impressed with the thread it delivered.

Krishna Kumar

I use this prompt every two weeks or so to compare campaign performance:

Here's a prompt optimized for Revy AI that gives it enough context to normalize the results rather than simply comparing the raw metrics.

Prompt:

I want to compare the performance of two advertising campaigns over the last 14 days, but I need the comparison to be normalized for spend rather than using the actual campaign budgets.

Campaigns:

Campaign 1: ABM campaign

Campaign 2: Broad (general audience) campaign

These campaigns have different total budgets, pacing, and daily maximum spend limits, so a direct comparison of total clicks, impressions, engaged accounts, etc. is not meaningful.

Please analyze both campaigns as if each campaign had spent exactly $1,000 over the same 14-day period.

Specifically:

Calculate or estimate the normalized performance of each campaign at a fixed spend of $1,000.

• Show the following metrics both in their actual values and normalized to $1,000 spend (where possible):ImpressionsClicksCTREngaged accountsAccount CTR (ACTR)Account View-Through Rate (AVTR)VisitsCost per click (CPC)Cost per engaged accountCost per visitCost per 1,000 impressions (CPM)Any available pipeline or opportunity metricsAny available conversions
• Calculate efficiency metrics, including:Engaged accounts per $1,000Clicks per $1,000Visits per $1,000Impressions per $1,000Pipeline generated per $1,000 (if available)Opportunities per $1,000 (if available)
• Compare the two campaigns and identify:Which campaign is more efficient at generating awareness.Which campaign is more efficient at driving engagement.Which campaign delivers the greatest value per advertising dollar.Which campaign appears to scale better if additional budget were allocated.
• Highlight any statistically significant differences or caveats, such as:Small sample sizesDelivery limitationsAudience saturationBudget constraintsFrequency differencesAny assumptions made when normalizing spend.

End with a recommendation on which campaign would receive additional budget based on performance per dollar spent, not total volume.

Please present the output as:

A comparison table showing actual vs. normalized ($1,000 spend) metrics.

A concise executive summary.

Key insights and recommendations.

This prompt makes it clear that you're looking for a spend-normalized efficiency analysis, which is the fairest way to compare an ABM campaign against a broader campaign with different budgets and pacing.

Lauren Triance-Haldane

Thank you so much for sharing your go-to prompts with us @Vinicius Scott @Lily Adkison and @Krishna Kumar! 👏

Please watch your inbox for an email from @Nicole Lesher on our team to claim your reward for participating.

Val

Create a personalized outbound outreach message for this prospect.
Prospect Information:[Paste LinkedIn profile]

Product Information:Name: [Product Name]

Description:[Product description, key capabilities, target users, differentiators, common challenges solved]

Instructions:- Use specific details from the prospect's profile to personalize the opening.
- Connect their role, responsibilities, initiatives, or recent activity to problems the product solves.
-Avoid generic compliments and obvious LinkedIn observations.
- Focus on likely business challenges and outcomes relevant to their role.
- Write in a natural, conversational, non-salesy tone.
- Keep the message under 120 words.
- End with a soft call to action.
- Do not use buzzwords or marketing language.
- Sound like a peer reaching out, not a salesperson.

Output:
1. Email subject line
2. Email body
3. LinkedIn message version
4. Why this personalization should resonate

Carl Landers

We find it very useful when we (unfortunately) lose a deal to immediately identify lookalike prospects for which we can deploy all the learnings we have from the recent deal. These two prompts have been helpful:

1. Unfortunately we lost a deal at {insert account name} that we should have won. Please identify the next best 5 {insert industry} accounts our sales team should pursue based on similar attributes, intent signals and likelihood to win.

2. Now with that list in hand, please prepare a summary brief for each of these five accounts outlining the compelling events, known pain points, topics being researched and what our entry approach should be for each. Analyze the company's position and create C-suite relevant value hypotheses for ABM positioning. Focus on value-based selling: every insight should connect back to C-suite relevant business outcomes. Make each value hypothesis specific with a value graph: Capability > Improved Metric > Business Outcome > C-suite Value.