PROMPT This AI Challenge

"Pipeline Saves Lives: A CRO’s Guide to Real AI" featuring Jason Rushforth

Every episode of the PROMPT This podcast includes an AI Challenge for the audience.  Follow the instructions below to complete this episode's challenge.

Start With the Problem, Then Earn the Tool

Most organizations approach AI backwards. They start with tools, demos, and vendor pitches. Then they go looking for a problem to justify the purchase.

That’s how you end up with shelfware.

The disciplined approach is different. You start with a constraint inside the business. You define its impact. Then you force any AI solution to prove it can fix it.

This challenge builds that muscle.

Step 1: Identify a Real Business Constraint

This is not limited to sales. Every function has friction.

Across a business, the starting points typically look like:

• Marketing: Campaigns generate activity but not qualified demand

• Customer Success: Renewals are reactive instead of proactive

• Finance: Forecasting cycles are slow and heavily manual

• Operations: Planning and resource allocation take too long

• Product: Feedback loops from customers are fragmented

Now translate that into a single sentence:

“The biggest constraint to performance in this part of the business is: ______”

If the problem isn’t specific, the solution won’t be either.

Step 2: Define the Impact in Business Terms

Before evaluating any AI solution, quantify the problem.

• What is the time cost?

• What is the revenue or margin impact?

• What decisions are delayed or degraded?

Set a clear threshold:

“A solution must improve this by ___% or reduce effort by ___ hours per week.”

This becomes your filter. Without it, everything looks useful.

Step 3: Match the Problem to an AI Use Case

Only after defining the problem do you map it to a category of AI.

Across functions, most high-value AI use cases fall into a few buckets:

• Decision support: Better forecasting, prioritization, or planning

• Workflow automation: Removing manual steps

• Execution assistance: Improving how people do their jobs in real time

• Data consolidation: Bringing fragmented information into one view

Now make it concrete with sales as the example.

Step 4: Sales Examples (From Abstract to Practical)

Here are some more in-depth examples for your sales department.

Constraint #1: Reps struggle in live conversations

Pain & impact: Reps hesitate or respond poorly in key moments, leading to lost deals and inconsistent win rates.

• Category: Execution assistance

• Approach: Use AI to generate responses and guide messaging
  
  

Check out this solution: ChatGPT (customized for objection handling and responses)

Constraint #2: Poor qualification leads to bad pipeline

Pain & impact: Unqualified deals enter the pipeline, inflating forecasts and wasting sales cycles that never convert.

• Category: Workflow + execution assistance

• Approach: Guide reps through key questions and capture answers automatically
  
  

Check out this solution: Winn.ai

Constraint #3: Pipeline creation is inconsistent

Pain & impact: Reps spend time on the wrong accounts, resulting in low pipeline coverage and missed revenue targets.

• Category: Decision support + execution

• Approach: Identify who to contact and orchestrate outreach
  
  

Check out this solution: Reggie.ai

Constraint #4: Inbound demand is inefficient

Pain & impact: High-value leads are delayed or mishandled, reducing conversion rates and increasing customer acquisition cost.

• Category: Workflow automation

• Approach: Automate qualification and routing of leads
  
  

Check out this solution:  Qualified

Constraint #5: Territory planning is slow and political

Pain & impact: Weeks are lost to manual planning and internal debates, delaying execution and creating uneven coverage across accounts.

• Category: Decision support

• Approach: Model scenarios and rebalance coverage quickly
  
  

Check out this solution:  BoogieBoard

Step 5: Run a Focused Pilot

No broad rollout. No long-term commitment.

• Select a single team or workflow

• Use real data

• Measure before and after
  
  

Track:

• Time saved

• Output improved (pipeline, conversion, retention)

• Decision speed

If the tool doesn’t move the metric you defined earlier, it doesn’t move forward.

Step 6: Decide With Discipline

At the end of the pilot:

• Scale it if it delivers measurable impact

• Kill it if it doesn’t

Avoid partial adoption. Avoid indefinite testing. This is where most organizations lose clarity.

What This Challenge Teaches

• AI is not a starting point. It’s a response to a problem

• The same evaluation discipline applies across every function

• The best use cases are tied directly to measurable outcomes

• Fewer tools, applied well, outperform broad experimentation
  
  

Final Thought

AI does not create clarity. It exposes whether you already have it.

If you can clearly define a constraint, measure its impact, and test a solution against it, AI becomes a powerful lever across the business.

If you can’t, you’ll accumulate tools and still have the same problems.