Generic prompts produce generic output. This should be obvious, but I watch revenue leaders make the same mistake every week.
They type something like "write me a cold outreach email for a hospital CFO" into Claude or ChatGPT, get a bland three-paragraph template that could have been written by any marketing intern in 2019, and conclude that AI is not useful for their work.
The problem was never the technology. The problem was the instruction.
I have spent the last two years building AI-enabled sales workflows for healthcare technology companies and PE-backed portfolio companies generating between $10M and $200M in annual revenue. The difference between teams that generate measurable revenue impact from AI and teams that generate impressive demos but no pipeline always comes down to the same thing: prompt architecture.
"Not clever tricks. Not one weird hack. Structured frameworks that force AI to operate with the same rigour you would expect from a senior strategist on your team."
Why Most Prompting Advice Fails Revenue Teams
Most prompting guides are written for generalists. They teach you how to get better creative writing, cleaner code, or more organised meeting notes. Useful, but not what a VP of Sales needs at 7am on a Monday when pipeline coverage is at 2.1x and the board meeting is in six weeks.
Revenue work has specific constraints that general-purpose prompting ignores:
Deals have history.
A cold outreach email to an account you lost eight months ago requires completely different framing than a first touch to a net-new prospect. Most prompts treat every interaction as if it is happening in a vacuum.
Buying committees are political.
The language that resonates with a CMO is structurally different from the language that moves a CFO. Not just in tone — in what counts as evidence, what triggers concern, and what kind of specificity signals competence.
Sales cycles have momentum.
A discovery call follow-up three days after a strong meeting carries different energy than one sent after two weeks of silence. Context is not a nice-to-have. It is the difference between advancing a deal and losing it.
The prompting frameworks that move revenue are designed to account for these constraints. They encode the judgment of experienced operators into the instruction layer, so the output reflects actual sales intelligence rather than surface-level pattern matching.
Framework 1: The CRTO Prompt Structure
CRTO stands for Commercial, Regulatory, Technical, and Organisational. It is a framework I developed for healthcare technology GTM, but it applies to any complex B2B sale where the buying process involves multiple stakeholders and multiple decision criteria.
When you need AI to help you prepare for a deal, structure your prompt around these four dimensions:
- Commercial. What is the financial case? What does the prospect's current cost structure look like? What is the ROI model? What budget cycle are they in?
- Regulatory. What compliance requirements constrain the decision? What regulatory changes are creating urgency or hesitation?
- Technical. What systems does this integrate with? What is the technical risk? Who on the buying committee cares most about implementation complexity?
- Organisational. Who are the decision-makers? What is the internal political landscape? Who benefits from the status quo?
"I am preparing for a discovery call with the VP of Revenue Cycle at a 600-bed health system. They currently use Epic for core revenue cycle management but have expressed frustration with denial rates… Help me build a discovery call framework that addresses all four CRTO dimensions and identifies the three most likely objections I will face."
That prompt produces output that a senior sales leader would actually use. The generic version produces output that belongs in a training manual for first-year reps.
Framework 2: The Buyer-Back Prompt
Most sales content is written from the seller's perspective. The Buyer-Back framework reverses this by prompting AI to think from the buyer's position first, then work backward to the seller's message.
The Structure
"You are [title] at [company type]. You have [specific problem]. You have tried [what they have already attempted]. You are skeptical of [common vendor claims]. You care most about [their actual priority]. Now, as this person, evaluate the following value proposition and tell me where it would make you lean forward and where it would make you check out."
One of my portfolio company clients ran their entire outbound sequence through the Buyer-Back framework before launch. They identified two claims in their messaging that would trigger immediate skepticism from their target buyer — hospital CIOs. They revised those claims before a single email went out. Their reply rate was 11% in the first month, roughly triple the industry average for cold outreach to health system executives.
Framework 3: The Evidence Ladder
This framework is purpose-built for creating proof-based sales collateral. In complex B2B sales, buyers do not trust claims. They trust evidence, and they trust it in a specific hierarchy.
- Tier 1: Named customer results with specific metrics.
- Tier 2: Industry benchmarks and third-party validation.
- Tier 3: Logical arguments grounded in the buyer's known operational reality.
- Tier 4: Analogies from adjacent industries where the same principle has been proven.
The instruction: "Do not fabricate case studies or invent statistics. Where we lack Tier 1 evidence, acknowledge the gap and recommend what we should measure to fill it."
This framework prevents the most common AI failure in sales content: confident fabrication. AI models will happily invent case studies, statistics, and customer quotes if you do not explicitly instruct them not to. In B2B sales, a single fabricated proof point can destroy a deal and a reputation.
Framework 4: The Compression Prompt
Revenue leaders are not short on information. They are short on time to process it. The Compression Prompt takes a large body of context — a 40-page RFP, a 90-minute call transcript, a competitor's 20-slide pitch deck — and reduces it to the three to five things that actually matter for the next decision.
The Structure
"Read the following [document type]. Extract only: (1) The three strongest signals that this prospect is ready to buy. (2) The two biggest risks to this deal advancing. (3) The single question I should ask in the next meeting that would clarify the most uncertainty. Ignore everything that does not directly affect whether this deal closes in the next 90 days."
I use this framework after every advisory client meeting. It takes a 60-minute conversation and turns it into a 200-word action brief in under two minutes — almost always sharper than the notes I would have written manually, because the framework forces ruthless prioritisation.
How to Implement These Across a Sales Team
Individual adoption is easy. Team adoption is where most AI enablement initiatives stall. The implementation pattern that works:
- Week 1. Select three reps who are already curious about AI. Give them the CRTO and Buyer-Back frameworks. Have them use both for every deal prep and outreach draft for five consecutive business days.
- Week 2. Review the outputs together. Identify which prompts produced genuinely useful results and which needed refinement. Adjust based on your specific market and buyer personas.
- Week 3. Expand to the full team. Distribute refined frameworks as shared prompt libraries. Pair each with a specific workflow moment: CRTO for deal prep, Buyer-Back for messaging review, Evidence Ladder for proposals, Compression for call follow-up.
- Week 4+. Measure. Track reply rates, meetings booked, qualified pipeline, average deal velocity. If the frameworks are working, those numbers move.
Frequently Asked Questions
What is the best AI prompting framework for B2B sales?
There is no single best framework. CRTO works well for complex enterprise sales. Buyer-Back is strongest for testing messaging. Evidence Ladder is built for proof-based collateral. The most effective teams use all four at different points in the sales process.
Can AI prompting frameworks replace sales training?
No. Frameworks encode judgment, but they do not teach it. What they do is amplify existing skill. A rep with three years of experience and a strong prompt framework will outperform a rep with ten years of experience and no framework.
How do AI prompting frameworks fit into an AI enablement strategy?
Prompting frameworks are the tactical layer of a broader AI enablement strategy. Frameworks without strategy produce isolated pockets of efficiency. Strategy without frameworks produces slide decks that never reach the field.
What is an AI enablement strategy for sales teams?
A structured plan for identifying which sales workflows benefit most from AI assistance, selecting and configuring the right tools, training on effective usage, and measuring commercial outcomes. The goal is not "use AI more." The goal is "generate more qualified pipeline and close more deals."
Work with Brett
I run a four-week GTM Sprint that includes building and deploying these frameworks inside your revenue team. If your team is experimenting with AI but not yet seeing it in the pipeline numbers, let's talk.