Rate the tools' ability to explain and evidence how each output was produced, at the depth an audit requires

The mean rating was 4.6 and the median 5.0 (n = 800 respondents). Ratings of 3 or below, the distressed range, made up 32.75% of the answers. The rating was associated with all four study measures. Higher ratings came with less cancellation (r = −0.65, 95% CI −0.73 to −0.54, p < 0.001, n = 150 engagements), higher customer satisfaction among active engagements (r = 0.67, 95% CI 0.54 to 0.77, p < 0.001, n = 94 engagements), a higher engagement indicator mean computed without the indicator (r = 0.88, 95% CI 0.84 to 0.91, p < 0.001, n = 150 engagements) and lower intent to leave (r = −0.65, cluster bootstrap 95% CI −0.70 to −0.59, p < 0.001 with standard errors clustered by engagement, n = 754 respondents). After Holm correction, ratings differed by engagement status and lifecycle stage (Welch F 45.00, Holm p < 0.001, omega squared 0.407), respondent role (Welch F 6.89, Holm p = 0.009, omega squared 0.013) and vendor type (Welch F 5.11, Holm p = 0.001, omega squared 0.026).

Of the 163 written responses, 145 were negative and 18 positive, and 84 carried the strongest negative score of −10. Respondents described tools that could produce an answer but not the reasoning or the source behind it, so that teams could not answer auditors, compliance, HR or finance and rebuilt the trail by hand. Across engagements, a more negative engagement mean written score came with cancellation, and respondents whose written responses were more negative reported higher intent to leave.

Ratings of 3 or below, the distressed range, accounted for 262 of the 800 answers (32.75%). The engagement mean fell in the distressed range in 33 of 150 engagements.

The primary responses numbered 71, all written in answer to C177, “Add any comments on the agentic AI tools that the ratings in this section do not capture.” The remaining 92 responses, 56.44% of the total, were secondary and came from 4 other questions. That secondary share ranked 19th from the highest among the 71 Part 2 indicators with 30 or more written responses, against a median of 18.92%. Respondents who wrote both a primary and a secondary response numbered 8. Negative responses made up 87.32% of primary responses (62 of 71) and 90.22% of secondary responses (83 of 92).

C029 asked respondents for their top three challenges to job satisfaction. Every one of its seventy-two responses named, among the three, agent decisions that could not be seen into, traced or audited. The seven responses to C055, on how the vendor shaped performance reports, described audit logs that captured only the agent's final output (four responses) and agent actions logged under a single service account so that no decision could be attributed (two responses).

Sixteen responses to C177 described an auditor, examiner, compliance team, HR or finance asking how an output was reached and the team unable to show it: why an account kept a manager role, how a return entered a composite, how a vendor record was decided, why an employee was told ten days' notice and not fifteen. One respondent put the distinction plainly: an examiner does not want a trace, but a reason. Thirteen described outputs that could not be traced to a source document, page or row, or that cited a source that did not contain the content, including a policy clause from a retired version and a news article about a different company with a similar name that was sent to a relationship desk.

Ten responses described a record kept or rebuilt by hand because the tool's trail was not enough: every supplier decision saved with its source documents at the end of each day, prompts and diffs pasted into a ticket, intermediate drafts kept for compliance, a personal note of every lead and its source taking an hour a day, and reasoning rebuilt from ticket history.

The eighteen positive responses came from two sources. The nine responses to C030, on what respondents loved about their job, named traceable answers, audit trails or an opaque agent made auditable. Four came from respondents describing a reconciliation agent whose decisions had been made fully auditable. The positive C177 responses described traces that compliance or audit accepted directly, several by contrast with an earlier engagement where no one could reconstruct what an agent had done.

Of the 163 responses, 145 (88.96%, 95% CI 82.17% to 94.84%) were negative and 84 (51.53%) carried the strongest negative score of −10. Positive responses numbered 18 (11.04%), and no response was neutral. The mean written score was −6.3 (95% CI −7.3 to −5.1, n = 163 responses), and the median was −10.0.

Of the 163 responses, 158 came from respondents who also rated the indicator. Among them, the written score rose with the rating (Spearman rho 0.54, 95% CI 0.39 to 0.65, p < 0.001, n = 158 responses). The written responses corroborated the rating. Of the 140 negative responses with a rating, 65 (46.43%) came from respondents whose rating was in the distressed range.

Respondents who wrote about the indicator and gave a rating numbered 147, with a mean rating of 4.2. The 653 respondents who rated the indicator without writing about it had a mean of 4.6. The difference, −0.4 (95% CI −0.8 to 0.0), included zero. Ratings in the distressed range made up 41.50% of the ratings from respondents who wrote (61 of 147) and 30.78% of the ratings from respondents who did not (201 of 653).

Every figure in the segment tables is at respondent level. Segments with fewer than 30 respondents carry no interval. Segments with fewer than 10 report counts only.

Segment families recorded at engagement level were tested on engagement means. Respondent role and vendor type were tested on respondents. Where an expected count fell below 5, the test of the distressed share used Fisher's exact test for a two-by-two table, marked exact, or a Monte Carlo test, marked simulated. After Holm correction, ratings differed by engagement status and lifecycle stage (Welch F 45.00, Holm p < 0.001, omega squared 0.407). Active engagements in Build-Out rated highest (mean 5.5, n = 168 respondents) and canceled engagements lowest (mean 3.2, n = 302 respondents). Ratings also differed by respondent role (Welch F 6.89, Holm p = 0.009, omega squared 0.013). Portfolio directors rated highest (mean 5.2, n = 29 respondents) and individual contributors lowest (mean 4.4, n = 619 respondents). Ratings also differed by vendor type (Welch F 5.11, Holm p = 0.001, omega squared 0.026). Consulting firms rated highest (mean 5.2, n = 109 respondents) and BPO specialists lowest (mean 4.1, n = 186 respondents). Outsourcing model, institution type, process family and class of work showed no difference.

Written sentiment differed by engagement status and lifecycle stage across the 90 engagements with a written response (Kruskal-Wallis H 14.30, df 3, p = 0.003). Of the groups with 10 or more responses, responses from canceled engagements were the most negative (mean −8.2, 68 responses) and responses from active engagements in Transition/Cut-Over the least negative (mean −3.9, 44 responses).

The role test compared individual contributors and engagement managers, the roles with 30 or more responses. Written sentiment did not differ (Kruskal-Wallis H 1.11, p = 0.292, n = 143 respondents): the mean written score was −6.7 for individual contributors (118 responses) and −4.6 for engagement managers (35 responses). Portfolio directors wrote 10 responses, reported as counts only.

Cancellation rate is the share of engagements canceled by the study date, measured at engagement level (150 engagements, 56 canceled). Canceled engagements were rated after cancellation. The association is therefore a descriptive contrast and not a forecast of cancellation.

Across the study set, higher ratings came with less cancellation (r = −0.65, 95% CI −0.73 to −0.54, Holm p < 0.001, n = 150 engagements). The association held after Holm correction in fourteen of the 20 segment levels tested.

Customer satisfaction is the percentage recorded against each engagement. The primary estimate uses the 94 active engagements. A descriptive contrast uses all 150 engagements.

Across the active engagements, higher ratings came with higher customer satisfaction (r = 0.67, 95% CI 0.54 to 0.77, Holm p < 0.001, n = 94 engagements). The association held after Holm correction in fifteen of the 16 segment levels tested.

The engagement indicator mean is the mean of an engagement's ratings across the rated indicators, measured at engagement level (150 engagements). The indicator is one of the rated indicators. Each comparison therefore uses the mean recomputed without it.

Higher ratings came with a higher engagement indicator mean (r = 0.88, 95% CI 0.84 to 0.91, Holm p < 0.001, n = 150 engagements). The association held after Holm correction in twenty-four of the 24 segment levels tested.

Intent to leave is the answer to C019, “How likely are you to leave your current job in the next 12 months?”, rated 1 to 10. It is measured at respondent level on up to 798 respondents, with the engagement-level form reported alongside.

Respondents who gave higher ratings reported lower intent to leave (r = −0.65, 95% CI −0.69 to −0.61, Holm p < 0.001, n = 754 respondents). The association held after Holm correction in thirty-seven of the 37 segment levels tested.

Two comparisons set the engagements with at least one negative response against all other engagements. Of the 63 other engagements, 60 had no written response addressing the indicator. These two comparisons therefore do not isolate written sentiment. Of the 87 engagements with a negative response, 36 (41.38%) were canceled, against 20 of the 63 other engagements (31.75%). The difference did not reach significance (Fisher exact p = 0.238). Customer satisfaction among active engagements did not differ between the two groups (mean 63.29% on 51 engagements against 65.05% on 43, Welch t −0.74, p = 0.464).

Two correlations test written sentiment directly. Across the 90 engagements with a written response, a more negative engagement mean written score came with cancellation (point-biserial r −0.34, p < 0.001). Across 140 respondents, a more negative respondent mean written score came with higher intent to leave (Spearman rho −0.40, p < 0.001). Of the four comparisons, two reached significance at 0.05.

The four associations were compared in pairs on their engagement-level forms, each oriented so that a higher value is the better outcome, across the 150 engagements with all four measures. After Holm correction, five of the six pairs differed. The association with the engagement indicator mean was the strongest and the association with cancellation the weakest. The associations with customer satisfaction and intent to leave did not differ (Holm p = 0.779).

Answers clustered within engagements (ICC 0.58). Clustering reduced the 800 respondent answers to the equivalent of about 229 independent answers. Respondent-level intervals therefore used the cluster bootstrap. Respondent-level tests of intent to leave used standard errors that allow for clustering by engagement.

No association was equivalent to zero. Each of the four differed from zero. The detection floor shown is the study floor at 150 engagements. Intent to leave was tested on its effective sample size.

Each change was computed before rounding and may differ by 0.01 from the difference between the two printed values.

All 22 checks held. A check holds when its test falls on the same side of 0.05 as the primary test. Where the primary test falls below 0.05, the check must also keep the sign of the primary correlation. The thin-base check removes the engagements in any level of outsourcing model or process family with fewer than 30 engagements. Institution type is not used for that check, because every one of its levels has fewer than 30 engagements.

Each of the four measures was computed from data the indicator does not contain. Each result is therefore a finding and not an arithmetic consequence. The indicator is one of the rated indicators, and the engagement indicator mean was recomputed without it. The correction changed r by 0.00, from 0.88 to 0.88.

Each deviation was computed before rounding and may differ by 0.01 from the difference between the two printed values.

The comparison covered 150 engagements. The indicator followed the study's typical pattern on all four measures, and no deviation held after Holm correction.

Among the 71 Part 2 indicators with 30 or more written responses, the indicator's share of negative responses, 88.96%, ranked 15th from the highest, against a median of 80.82%. Its mean written score, −6.3, ranked 11th from the most negative, against a median of −4.7.

The indicator most often present in the same responses was C176, “Rate the quality of the tools' logs, traces, and usage telemetry for detecting and diagnosing failures.”, in 33 of the 163 responses (20.25%). C157, “Rate the factual reliability of the AI tools' outputs on this engagement before any human review.” (30 responses), C126, “Rate the maturity of your employer's playbook for taking agentic use cases from concept to governed production.” (14 responses), C098, “Rate how well the customer's legacy systems support the integration this engagement requires.” (13 responses) and C164, “Rate how reliably the tools retained instructions and constraints given earlier in long tasks and sessions.” (10 responses) followed.

The most frequent theme was “Comparison to a prior engagement”, in 9 responses (5.52%). The second was “Respondent proposes a remedy”, in 7 responses.

About a third of all ratings of the tools' ability to explain and evidence their outputs fell in the distressed range (262 of 800 answers, 32.75%). The rating was associated with all four study measures, and every association held under each of the 22 robustness checks. Canceled engagements rated far below the three active stages (mean 3.2, n = 302 respondents).

The written base is large, and 88.96% of it was negative. For work a financial institution must be able to defend to an auditor, an examiner or a customer, an output without a reason attached cannot be relied on even when it is correct, and several respondents said exactly that. The cost appeared as hand-built records and as staff who could not stand behind work they believed was right. The positive responses show the gap can be closed, since some engagements produced traces that audit accepted at the first sitting. Written sentiment was associated with cancellation and with intent to leave.

The evidence is the reported experience of the vendor's delivery teams, collected in a single cross-section. The associations reported here are not causal.

Tests by respondent role and by vendor type treated respondents as independent. The section on clustering reports how far clustering within engagements reduced the information in respondent-level figures.

Customer management executives gave 2 answers, a base of anecdote strength.

Shares of written sentiment are shares of responses and not of the study set of respondents. The written score takes five fixed values, so its mean and median are coarse. Responses clustered within engagements (163 responses on 90 engagements). Intervals for written sentiment therefore used a bootstrap by engagement.