Rate how well the agents on this engagement are protected from breaking silently when a third party changes something without warning

The mean rating was 4.6 and the median 5.0 (n = 781 respondents). Ratings of 3 or below, the distressed range, made up 31.88% of the answers. The rating was associated with all four study measures. Higher ratings came with less cancellation (r = −0.64, 95% CI −0.72 to −0.53, p < 0.001, n = 150 engagements), higher customer satisfaction among active engagements (r = 0.68, 95% CI 0.55 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 −0.71 to −0.59, p < 0.001 with standard errors clustered by engagement, n = 733 respondents). After Holm correction, ratings differed by engagement status and lifecycle stage (Welch F 37.06, Holm p < 0.001, omega squared 0.393) and respondent role (Welch F 10.32, Holm p < 0.001, omega squared 0.024).

Of the 136 written responses, most from C198, the comment field on the third-party ecosystem, 100 were negative and 34 positive. Respondents described agents that kept running when a provider changed something and produced wrong or empty output without an error, with no monitoring to catch the change and nothing to fall back on. Engagements with a negative written response were canceled more often, 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 249 of the 781 answers (31.88%). The engagement mean fell in the distressed range in 36 of 150 engagements.

The primary responses numbered 117, all written in answer to C198, “Add any comments on the third-party ecosystem that the ratings in this section do not capture.” The remaining 19 responses, 13.97% of the total, were secondary and came from 6 other questions. That secondary share ranked 40th 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 1. Negative responses made up 69.23% of primary responses (81 of 117) and 100.00% of secondary responses (19 of 19).

Sixteen responses described an agent that kept running on changed input and produced wrong, empty or partial output without raising an error. A custodian shifted two columns and breaks were matched against the wrong field for a day and a half. A watchlist file gained a column and screening passed names it should have stopped. A leavers file changed layout and deprovisioning processed nothing for four days. A ticketing interface changed over a weekend and the agent closed change tickets that had never been raised. One respondent summed it up: the failure is quiet and the run comes back green. Nine responses said nothing watched the third-party inputs, so the change was found from a failure or, in one account, from employees before any system.

Eight responses described agents working through interfaces built for people, a statutory portal, a market data terminal, network and consent portals. These broke whenever the provider moved a button or a column. One team ran a nightly script that clicked through a provider's web console and found out about changes when the morning file was empty. Two responses described checks the teams built for themselves: known cases pushed through before the batch each morning, and checks that caught a renamed column before the load ran. One respondent described pulling rates once a week into a workbook so that runs would not fail mid-week, and called the result stale by design and not defensible. Two respondents described the burden on staff, including two engineers lost from the engagement in three months over night call-outs.

Nineteen responses came from the questions on altering outputs, data and processes and their comment fields, and described patching output or working around a broken feed before it went out, in several cases without telling anyone beyond the team. The thirty-four positive responses described stable providers, changes caught loudly by the team's own checks, or a standby route used within the hour, while noting a single model provider, licenses written for named users or knowledge that had left with former staff.

Of the 136 responses, 100 (73.53%, 95% CI 64.62% to 81.51%) were negative and 26 (19.12%) carried the strongest negative score of −10. Positive responses numbered 34 (25.00%), and 2 were neutral. The mean written score was −3.2 (95% CI −4.2 to −2.1, n = 136 responses), and the median was −5.0.

Of the 136 responses, 127 came from respondents who also rated the indicator. Among them, the written score rose with the rating (Spearman rho 0.44, 95% CI 0.26 to 0.61, p < 0.001, n = 127 responses). The written responses corroborated the rating. Of the 92 negative responses with a rating, 56 (60.87%) came from respondents whose rating was in the distressed range.

Respondents who wrote about the indicator and gave a rating numbered 125, with a mean rating of 4.2. The 656 respondents who rated the indicator without writing about it had a mean of 4.7. The difference, −0.5 (95% CI −1.0 to 0.1), included zero. Ratings in the distressed range made up 47.20% of the ratings from respondents who wrote (59 of 125) and 28.96% of the ratings from respondents who did not (190 of 656).

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 37.06, Holm p < 0.001, omega squared 0.393). Active engagements in Transition/Cut-Over rated highest (mean 5.6, n = 235 respondents) and canceled engagements lowest (mean 3.1, n = 293 respondents). Ratings also differed by respondent role (Welch F 10.32, Holm p < 0.001, omega squared 0.024). Portfolio directors rated highest (mean 6.2, n = 30 respondents) and individual contributors lowest (mean 4.5, n = 599 respondents). Outsourcing model, institution type, process family, class of work and vendor type showed no difference.

Written sentiment differed by engagement status and lifecycle stage across the 81 engagements with a written response (Kruskal-Wallis H 16.28, df 3, p < 0.001). Of the groups with 10 or more responses, responses from canceled engagements were the most negative (mean −5.2, 56 responses) and responses from active engagements in Production the least negative (mean −0.6, 17 responses).

No test across respondent roles was run, because fewer than two roles had 30 or more written responses. Individual contributors wrote 111, engagement managers wrote 20 and portfolio directors wrote 5 responses.

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 = −0.64, 95% CI −0.72 to −0.53, Holm p < 0.001, n = 150 engagements). The association held after Holm correction in thirteen 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 = 0.68, 95% CI 0.55 to 0.77, Holm p < 0.001, n = 94 engagements). The association held after Holm correction in twelve 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 = 733 respondents). The association held after Holm correction in thirty-six of the 37 segment levels tested.

Two comparisons set the engagements with at least one negative response against all other engagements. Of the 89 other engagements, 69 had no written response addressing the indicator. These two comparisons therefore do not isolate written sentiment. Of the 61 engagements with a negative response, 31 (50.82%) were canceled, against 25 of the 89 other engagements (28.09%). The difference reached significance (Fisher exact p = 0.006). Customer satisfaction among active engagements did not differ between the two groups (mean 64.50% on 30 engagements against 63.91% on 64, Welch t 0.24, p = 0.813).

Two correlations test written sentiment directly. Across the 81 engagements with a written response, a more negative engagement mean written score came with cancellation (point-biserial r −0.38, p < 0.001). Across 127 respondents, a more negative respondent mean written score came with higher intent to leave (Spearman rho −0.34, p < 0.001). Of the four comparisons, three 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 = 0.975).

Answers clustered within engagements (ICC 0.65). Clustering reduced the 781 respondent answers to the equivalent of about 208 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, 73.53%, ranked 52nd from the highest, against a median of 80.82%. Its mean written score, −3.2, ranked 59th from the most negative, against a median of −4.7.

The indicator most often present in the same responses was C181, “Rate the quality of advance notice third-party providers give before changes that can affect the agents on this engagement.”, in 68 of the 136 responses (50.00%). C187, “Rate the stability of the third-party APIs and data structures the agents depend on, from one provider release to the next.” (52 responses), C183, “Rate the engagement's monitoring for unannounced changes in the third-party APIs, data feeds, and models its agents depend on.” (20 responses), C179, “Rate the quality of the fallback the engagement has ready when a critical third-party feed, platform, or model stops working.” (19 responses) and C186, “Rate the availability of true machine interfaces, rather than screens built for humans, across the third-party systems the agents must use.” (15 responses) followed.

The most frequent theme was “Detection only by failure”, in 71 responses (52.21%). The second was “Schema or format changed without notice”, in 32 responses.

About a third of all ratings of how well agents were protected from breaking silently fell in the distressed range (249 of 781 answers, 31.88%). 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 3.1, n = 293 respondents).

The written evidence distinguishes a loud failure from a silent one. Where a change stopped the agent, the cost was a lost morning. Where the agent carried on, the error passed into reconciliations, screening and access control and was found days later, by a person. Some of the respondents who rated this well worked on engagements that had built a cheap test of their own inputs. Written sentiment was associated with cancellation in both engagement-level tests 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 (136 responses on 81 engagements). Intervals for written sentiment therefore used a bootstrap by engagement.