Skip to Content
BayWise LensReferenceConfidence Levels

Confidence Levels

Every finding in Lens carries a confidence level — Low, Medium, or High — that reflects how strongly the available evidence supports the finding and its money figure.

Confidence is not a measure of whether the finding is real. It is a measure of how precisely the finding has been quantified. A Low confidence finding can still represent a significant, real gap — it means that the data to size it precisely was not available.


The three levels

High

The key inputs to the finding’s calculation are verified from hard data — DMS exports, call logs, financial statements, or other structured records. The money figure is directly computable from real numbers.

What it means for the Dealer Principal: The figure shown is the calculated result from their own data. It stands up to scrutiny and to an auditor.

Example: The booking call abandonment rate is 22%, verified from three months of phone system logs. The average RO value is AED 1,450, verified from the DMS export. The monthly RO volume is 820, verified from the appointment schedule export. The finding’s annual figure is directly calculated from these verified inputs.


Medium

Some inputs are verified from hard data; others are estimates derived from consultant observations, interview notes, or regional medians. The money figure is a blend of hard figures and reasoned estimates.

What it means for the Dealer Principal: The direction and order of magnitude of the finding are reliable. The precise figure carries some uncertainty.

Example: The MPI completion rate is estimated at 40–45% based on advisor interviews and a sample of 20 inspection reports, but the dealership does not produce a formal MPI tracking report. The average additional work per completed MPI is AED 380, sourced from the DMS for completed jobs with additional work. The finding is Medium confidence because the completion rate input is an estimate, not a measured figure.


Low

The finding is based primarily on estimates, observations, and interview signals. Structured data was not available or was not provided. The money figure is an order-of-magnitude estimate.

What it means for the Dealer Principal: The finding direction is reliable — the gap exists — but the specific figure requires further measurement to confirm. The finding is a recommendation to investigate and measure, not a certified figure.

Example: The follow-up and feedback section showed no evidence of a post-service outreach programme, based on the service manager’s description during the interview. No call logs, CRM data, or customer satisfaction records were available. The retention impact is estimated from regional medians for dealerships with no outreach process. The finding is Low confidence.


How confidence affects findings

Confidence is displayed on the expanded finding card alongside addressability. It is visible in both the Working view and the Principal view.

Lower confidence does not suppress a finding — it contextualises it. When you present a Low confidence finding to the Dealer Principal, the natural follow-on is: “This is our estimate based on what we could observe. The first 30-day action is to instrument this properly so we have hard numbers for the next engagement.”

Low confidence findings often highlight a data maturity gap as much as a performance gap. Recommending measurement as the first action is appropriate and credible.


Improving confidence

Confidence improves when better data is available. If you receive additional documents during the engagement:

  1. Upload them to the relevant section
  2. The section summary updates
  3. The finding recalculates with the new inputs
  4. The confidence level updates to reflect the improved evidence base