How to Set Go/No-Go Criteria Before a Pilot Starts

Setting clear go/no-go criteria is a foundational step for any technology pilot, especially when evaluating new sales AI tools. These criteria are specific, measurable benchmarks that determine whether a pilot moves forward to full adoption or is stopped. Without them, pilot programs often drift, consuming resources without a clear path to a decision.

The purpose of defining these criteria upfront is to remove subjectivity from the evaluation process. It forces teams to agree on what success looks like before any money or time is invested. This proactive approach ensures that decisions are based on data and predefined objectives, not on enthusiasm or sunk costs.

Why Go/No-Go Criteria are Non-Negotiable

Many organizations initiate pilots with vague goals like “improve efficiency” or “boost productivity.” These are aspirations, not measurable criteria. Without concrete go/no-go metrics, pilots can become endless experiments. They consume budget and team attention without ever reaching a definitive conclusion.

A pilot without clear go/no-go criteria is an experiment without a hypothesis; it can generate data, but it cannot prove or disprove anything definitively.

This lack of definition leads to several common problems:

• Sunk Cost Fallacy: Teams continue investing in a failing pilot because of the resources already spent, rather than cutting losses.
• Stakeholder Disagreement: Without agreed-upon metrics, different departments will have different ideas of what constitutes success, leading to conflict.
• Lack of Accountability: It becomes difficult to hold vendors or internal teams accountable for results if the targets were never clearly set.
• Delayed Decision-Making: Indecision prolongs the pilot phase, delaying the realization of benefits from successful tools or the reallocation of resources from unsuccessful ones.

A well-defined set of criteria provides a framework for objective evaluation. It makes the decision to scale or stop a pilot straightforward and data-driven. This discipline is critical for effective AI pilot governance.

Types of Go/No-Go Criteria

Go/no-go criteria typically fall into several categories. A balanced set of criteria will include elements from each.

1. Performance Metrics

These are the most common and often the most critical. They measure the direct impact of the new technology on key business outcomes.

• Sales Cycle Reduction: For an AI tool promising to accelerate deal progression, a target might be a 10% reduction in average sales cycle length for pilot participants.
• Conversion Rates: If the tool aims to improve lead quality or outreach effectiveness, a target could be a 5% increase in MQL-to-SQL conversion or meeting-booked rate.
• Revenue Impact: For tools directly influencing pipeline or closed-won deals, a target could be a specific uplift in pipeline generated or average deal size.
• Efficiency Gains: For automation tools, this might be measured in hours saved per week for SDRs or AEs, or a reduction in manual data entry errors.

2. User Adoption and Satisfaction

Even if a tool performs well on paper, it will fail if users do not adopt it. These criteria measure how well the tool integrates into daily workflows and user acceptance.

• Usage Rate: A minimum percentage of pilot users actively logging into and using the tool daily or weekly. For example, 80% daily active users.
• Feature Adoption: Specific key features of the tool must be used by a certain percentage of the pilot group.
• Satisfaction Scores: A minimum Net Promoter Score (NPS) or internal satisfaction survey score from pilot participants.
• Training Completion: A high percentage of pilot users completing required training modules.

3. Technical and Integration Feasibility

These criteria ensure the solution is technically sound and can integrate smoothly with existing systems.

• System Uptime: The tool must maintain a minimum uptime percentage during the pilot, e.g., 99.9%.
• Data Sync Accuracy: Data flowing between the new tool and your CRM must be accurate, with a defined error rate threshold (e.g., less than 0.5% discrepancies).
• Security Compliance: The solution must meet all internal security and data privacy standards.
• Scalability Test: The pilot should demonstrate the tool’s ability to handle expected future load without performance degradation.

4. Cost and ROI

Ultimately, the pilot must demonstrate a clear return on investment. This often ties back to performance metrics but frames them in financial terms.

• Positive ROI Projection: A clear projection that the tool will deliver a positive ROI within a specified timeframe (e.g., 12 months) if scaled. This requires careful calculation, as discussed in How to calculate the real ROI of a sales AI tool before you buy it.
• Cost-Benefit Ratio: The benefits derived must outweigh the operational costs of the tool.
• Budget Adherence: The pilot must stay within its allocated budget.

Structuring Your Go/No-Go Criteria

When setting criteria, follow the SMART framework: Specific, Measurable, Achievable, Relevant, and Time-bound.

Here is a typical structure for defining criteria:

It is crucial to establish a baseline before the pilot begins. Without knowing current performance, you cannot accurately measure the impact of the new tool. This often involves pulling historical data from your CRM or other sales tools.

The Role of the Pilot Steering Committee

The pilot steering committee is responsible for defining, approving, and monitoring these go/no-go criteria. This committee, comprising cross-functional leaders, ensures that the criteria are comprehensive and reflect the organization’s strategic goals.

During the pilot, the committee regularly reviews progress against these criteria. They analyze data, gather feedback, and make interim adjustments if absolutely necessary (though criteria should be largely stable). At the pilot’s conclusion, the committee uses the predefined criteria to make the final go/no-go decision. This structured approach prevents a “failed pilot” from being seen as a complete loss; instead, it becomes a learning opportunity, as explored in Does a failed pilot mean the category doesn’t work?.

Common Pitfalls to Avoid

Even with the best intentions, organizations can stumble when setting go/no-go criteria.

• Too Many Criteria: Overloading the pilot with too many metrics can make it difficult to focus and interpret results. Prioritize 3-5 critical criteria.
• Vague Criteria: Avoid subjective language. “Significantly improve” is not measurable. “Increase conversion rate by 10%” is.
• Lack of Baseline Data: Without understanding current performance, you cannot measure improvement. Ensure you have solid baseline metrics before starting.
• Ignoring User Feedback: Even if performance metrics are met, a tool that users hate will not scale. Incorporate adoption and satisfaction criteria.
• Moving the Goalposts: Changing criteria mid-pilot undermines objectivity and trust. Any changes must be rare, justified, and formally approved by the steering committee.
• Focusing Only on “Go”: It is equally important to define what constitutes a “no-go.” What specific thresholds, if breached, will automatically terminate the pilot?

Documenting Your Criteria

All go/no-go criteria, along with baselines, measurement methods, and responsible parties, should be thoroughly documented. This document serves as the single source of truth for the pilot. It should be shared with all stakeholders, including the vendor. Clear documentation also aids in future evaluations and helps streamline subsequent pilots, as detailed in How to document a pilot so the next one is faster.

The documentation should include:

• Pilot Objectives: A high-level statement of what the pilot aims to achieve.
• Scope: What teams, processes, and systems are included/excluded.
• Pilot Duration: Start and end dates.
• Go/No-Go Criteria Table: As outlined above, with specific metrics, targets, and thresholds.
• Measurement Plan: How and when data will be collected, who is responsible.
• Decision-Making Process: Who makes the final go/no-go decision and based on what process.

By rigorously defining these criteria, organizations can approach sales tech pilots with clarity and confidence. This structured approach maximizes the chances of successful adoption for valuable tools and minimizes wasted effort on those that do not deliver.