Study Data · Part 4 · Indicator Deep-Dive Analysis

What AI certifications do you have?

Indicator C016 · By Phil Hatch · Akholi · October 9, 2026

← Study Data ← Previous indicator Next indicator →

Indicator Specification Summary of Findings Distribution of Responses Descriptive Statistics by Segment Tests of Difference Between Segments Association with Cancellation Rate Association with Customer Satisfaction Association with the Engagement Indicator Mean Association with Intent to Leave Comparison of Associations Across the Four Measures Clustering and Effective Sample Size Equivalence Tests of Null Results Robustness Checks Measure Independence and Part-Whole Correction Comparison with the Study Norm Interpretation of the Data Limitations Definitions Sources and Provenance

Indicator Specification

Table 1. Indicator specificationTable

Field	Value
Indicator code	C016
Question as asked	What AI certifications do you have?
Failure category	Not one of the 100 rated indicators
Place in the survey	Respondent Demographics > 1.2 Experience and Expertise > Credentials
Indicator class	Respondent level
Indicator type	Categorical
Response scale	Levels: None, Databricks Generative AI Fundamentals, Google Cloud Generative AI, Azure AI Fundamentals, DeepLearning.AI certificate, AWS Certified Machine Learning
Answering base	788 of 850 respondents gave one of the listed answers, across 150 of 150 engagements. None answered I do not know or I prefer to not answer
Non-answers	I do not know 0, I prefer to not answer 0, blank 62

Summary of Findings

Table 2. Key results: association of the indicator with the four study measuresTable

Measure	Effect size	n	Value	p
Cancellation rate	Not tested: fewer than two levels of 10 or more engagements
Customer satisfaction	Not tested: fewer than two levels of 10 or more engagements
Engagement indicator mean	Not tested: fewer than two levels of 10 or more engagements
Intent to leave	Not tested: fewer than two levels of 10 or more engagements

Of the 850 respondents, 788 answered the question. None, a listed answer, was given by 437 (55.46%). The other 351 named one of five AI certifications, each held by between 66 and 78 respondents: Databricks Generative AI Fundamentals, Google Cloud Generative AI, Azure AI Fundamentals, a DeepLearning.AI certificate and AWS Certified Machine Learning.

The answers did not differ by role, vendor type or any segment family recorded at engagement level. With None as a level, they did not separate the respondents on cancellation, customer satisfaction, the engagement indicator mean or intent to leave.

Distribution of Responses

Table 3. Distribution of responses, respondent levelTable

Answer	Respondents	Share of answered (base 788)	Cumulative share	Engagements with at least one	Share of 150 engagements	Distressed
None	437	55.46%	55.46%	148	98.67%	Not applicable
Databricks Generative AI Fundamentals	78	9.90%	65.36%	60	40.00%	Not applicable
Google Cloud Generative AI	72	9.14%	74.49%	59	39.33%	Not applicable
Azure AI Fundamentals	68	8.63%	83.12%	58	38.67%	Not applicable
DeepLearning.AI certificate	67	8.50%	91.62%	55	36.67%	Not applicable
AWS Certified Machine Learning	66	8.38%	100.00%	56	37.33%	Not applicable
Answered, total	788	100.00%		150	100.00%

None was the answer of respondents on 148 of the 150 engagements. Each engagement was assigned the answer given by the largest number of its respondents, with None counted as an answer. Twenty-five engagements were tied and carry no assigned answer, and None was the assigned answer on 115 of the remaining 125.

Descriptive Statistics by Segment

Table 4. Descriptive statistics by segmentTable

Segment	Base	Answered	Answer rate	Engagements	None, share	Databricks Generative AI Fundamentals, share
Study set	850	788	92.71% of 850	150	55.46%	9.90%
Outsourcing model
CXM	232	216	93.10% of 232	42	59.72%	7.87%
ITO	244	231	94.67% of 244	40	54.55%	9.09%
BPO	209	188	89.95% of 209	37	56.91%	10.64%
KPO	165	153	92.73% of 165	31	49.02%	13.07%
Institution type
Asset management	132	120	90.91% of 132	24	55.00%	8.33%
Wealth management and private banking	140	131	93.57% of 140	23	61.07%	9.16%
Investment banking (advisory and capital-markets origination)	105	96	91.43% of 105	19	56.25%	12.50%
Insurance (life, property and casualty, health)	78	73	93.59% of 78	14	60.27%	6.85%
Commercial and business banking	82	74	90.24% of 82	13	45.95%	10.81%
Payments and merchant acquiring (networks, acquiring, processing)	71	69	97.18% of 71	13	46.38%	14.49%
Credit card and payments issuing	73	69	94.52% of 73	12	56.52%	8.70%
Retail and consumer banking	73	68	93.15% of 73	12	64.71%	2.94%
Markets and trading (sales and trading, global markets)	57	52	91.23% of 57	11	51.92%	9.62%
Private equity and private capital	39	36	92.31% of 39	9	47.22%	22.22%
Process family
Corporate and back-office functions	248	227	91.53% of 248	47	52.42%	9.69%
Technology engineering and operations	204	194	95.10% of 204	35	57.22%	7.73%
Client servicing and communications	94	90	95.74% of 94	16	65.56%	10.00%
Trade, settlement and reconciliation	80	70	87.50% of 80	15	51.43%	17.14%
Payments, cards and disputes	73	67	91.78% of 73	12	55.22%	5.97%
Credit, underwriting and suitability	57	54	94.74% of 57	9	59.26%	3.70%
Regulatory, audit, tax and legal	51	47	92.16% of 51	9	51.06%	14.89%
Financial crime and client onboarding	43	39	90.70% of 43	7	48.72%	17.95%
Class of work
LOB Operational	451	415	92.02% of 451	76	55.66%	10.36%
Corporate Functional	399	373	93.48% of 399	74	55.23%	9.38%
Engagement status and lifecycle stage
Active, Build-Out	175	163	93.14% of 175	31	56.44%	9.82%
Active, Transition/Cut-Over	250	232	92.80% of 250	43	55.17%	9.91%
Active, Production	103	98	95.15% of 103	20	59.18%	9.18%
Canceled	322	295	91.61% of 322	56	53.90%	10.17%
Respondent role
Individual Contributor	667	608	91.15% of 667	150	55.43%	9.38%
Engagement Manager	150	150	100.00% of 150	150	56.67%	11.33%
Portfolio Director	31	30	96.77% of 31	30	50.00%	13.33%
Customer Management Executive	2	0	0 of 2	0	0	0
Vendor type
IT services major	316	298	94.30% of 316	59	54.70%	10.40%
BPO specialist	198	181	91.41% of 198	37	51.93%	11.05%
Consulting firm	117	112	95.73% of 117	23	51.79%	8.93%
Boutique or specialist provider	87	78	89.66% of 87	17	64.10%	7.69%
Pure-play agentic AI firm	75	65	86.67% of 75	14	56.92%	9.23%
Not reported, recovered by engagement	57	54	94.74% of 57	46	64.81%	9.26%

Table 5. Descriptive statistics by segment, remaining answer levelsTable

Segment	Base	Answered	Answer rate	Engagements	Google Cloud Generative AI, share	Azure AI Fundamentals, share
Study set	850	788	92.71% of 850	150	9.14%	8.63%
Outsourcing model
CXM	232	216	93.10% of 232	42	5.56%	8.33%
ITO	244	231	94.67% of 244	40	9.09%	9.09%
BPO	209	188	89.95% of 209	37	12.23%	7.45%
KPO	165	153	92.73% of 165	31	10.46%	9.80%
Institution type
Asset management	132	120	90.91% of 132	24	9.17%	10.83%
Wealth management and private banking	140	131	93.57% of 140	23	6.11%	7.63%
Investment banking (advisory and capital-markets origination)	105	96	91.43% of 105	19	9.38%	5.21%
Insurance (life, property and casualty, health)	78	73	93.59% of 78	14	12.33%	9.59%
Commercial and business banking	82	74	90.24% of 82	13	10.81%	14.86%
Payments and merchant acquiring (networks, acquiring, processing)	71	69	97.18% of 71	13	10.14%	7.25%
Credit card and payments issuing	73	69	94.52% of 73	12	10.14%	8.70%
Retail and consumer banking	73	68	93.15% of 73	12	5.88%	7.35%
Markets and trading (sales and trading, global markets)	57	52	91.23% of 57	11	11.54%	9.62%
Private equity and private capital	39	36	92.31% of 39	9	8.33%	2.78%
Process family
Corporate and back-office functions	248	227	91.53% of 248	47	8.37%	8.37%
Technology engineering and operations	204	194	95.10% of 204	35	8.76%	7.22%
Client servicing and communications	94	90	95.74% of 94	16	5.56%	10.00%
Trade, settlement and reconciliation	80	70	87.50% of 80	15	11.43%	8.57%
Payments, cards and disputes	73	67	91.78% of 73	12	13.43%	13.43%
Credit, underwriting and suitability	57	54	94.74% of 57	9	11.11%	11.11%
Regulatory, audit, tax and legal	51	47	92.16% of 51	9	6.38%	6.38%
Financial crime and client onboarding	43	39	90.70% of 43	7	12.82%	5.13%
Class of work
LOB Operational	451	415	92.02% of 451	76	10.12%	10.36%
Corporate Functional	399	373	93.48% of 399	74	8.04%	6.70%
Engagement status and lifecycle stage
Active, Build-Out	175	163	93.14% of 175	31	9.82%	9.82%
Active, Transition/Cut-Over	250	232	92.80% of 250	43	11.21%	7.76%
Active, Production	103	98	95.15% of 103	20	7.14%	8.16%
Canceled	322	295	91.61% of 322	56	7.80%	8.81%
Respondent role
Individual Contributor	667	608	91.15% of 667	150	9.87%	8.72%
Engagement Manager	150	150	100.00% of 150	150	7.33%	8.00%
Portfolio Director	31	30	96.77% of 31	30	3.33%	10.00%
Customer Management Executive	2	0	0 of 2	0	0	0
Vendor type
IT services major	316	298	94.30% of 316	59	8.72%	8.72%
BPO specialist	198	181	91.41% of 198	37	11.60%	8.29%
Consulting firm	117	112	95.73% of 117	23	8.93%	8.93%
Boutique or specialist provider	87	78	89.66% of 87	17	3.85%	7.69%
Pure-play agentic AI firm	75	65	86.67% of 75	14	13.85%	10.77%
Not reported, recovered by engagement	57	54	94.74% of 57	46	5.56%	7.41%

Table 6. Descriptive statistics by segment, remaining answer levelsTable

Segment	Base	Answered	Answer rate	Engagements	DeepLearning.AI certificate, share	AWS Certified Machine Learning, share
Study set	850	788	92.71% of 850	150	8.50%	8.38%
Outsourcing model
CXM	232	216	93.10% of 232	42	9.72%	8.80%
ITO	244	231	94.67% of 244	40	8.66%	9.52%
BPO	209	188	89.95% of 209	37	5.85%	6.91%
KPO	165	153	92.73% of 165	31	9.80%	7.84%
Institution type
Asset management	132	120	90.91% of 132	24	7.50%	9.17%
Wealth management and private banking	140	131	93.57% of 140	23	7.63%	8.40%
Investment banking (advisory and capital-markets origination)	105	96	91.43% of 105	19	6.25%	10.42%
Insurance (life, property and casualty, health)	78	73	93.59% of 78	14	5.48%	5.48%
Commercial and business banking	82	74	90.24% of 82	13	10.81%	6.76%
Payments and merchant acquiring (networks, acquiring, processing)	71	69	97.18% of 71	13	11.59%	10.14%
Credit card and payments issuing	73	69	94.52% of 73	12	10.14%	5.80%
Retail and consumer banking	73	68	93.15% of 73	12	11.76%	7.35%
Markets and trading (sales and trading, global markets)	57	52	91.23% of 57	11	7.69%	9.62%
Private equity and private capital	39	36	92.31% of 39	9	8.33%	11.11%
Process family
Corporate and back-office functions	248	227	91.53% of 248	47	10.57%	10.57%
Technology engineering and operations	204	194	95.10% of 204	35	8.76%	10.31%
Client servicing and communications	94	90	95.74% of 94	16	3.33%	5.56%
Trade, settlement and reconciliation	80	70	87.50% of 80	15	7.14%	4.29%
Payments, cards and disputes	73	67	91.78% of 73	12	8.96%	2.99%
Credit, underwriting and suitability	57	54	94.74% of 57	9	9.26%	5.56%
Regulatory, audit, tax and legal	51	47	92.16% of 51	9	12.77%	8.51%
Financial crime and client onboarding	43	39	90.70% of 43	7	2.56%	12.82%
Class of work
LOB Operational	451	415	92.02% of 451	76	6.75%	6.75%
Corporate Functional	399	373	93.48% of 399	74	10.46%	10.19%
Engagement status and lifecycle stage
Active, Build-Out	175	163	93.14% of 175	31	6.75%	7.36%
Active, Transition/Cut-Over	250	232	92.80% of 250	43	9.48%	6.47%
Active, Production	103	98	95.15% of 103	20	7.14%	9.18%
Canceled	322	295	91.61% of 322	56	9.15%	10.17%
Respondent role
Individual Contributor	667	608	91.15% of 667	150	8.39%	8.22%
Engagement Manager	150	150	100.00% of 150	150	7.33%	9.33%
Portfolio Director	31	30	96.77% of 31	30	16.67%	6.67%
Customer Management Executive	2	0	0 of 2	0	0	0
Vendor type
IT services major	316	298	94.30% of 316	59	9.06%	8.39%
BPO specialist	198	181	91.41% of 198	37	9.39%	7.73%
Consulting firm	117	112	95.73% of 117	23	10.71%	10.71%
Boutique or specialist provider	87	78	89.66% of 87	17	6.41%	10.26%
Pure-play agentic AI firm	75	65	86.67% of 75	14	3.08%	6.15%
Not reported, recovered by engagement	57	54	94.74% of 57	46	7.41%	5.56%

Each share is computed on the respondents who answered in that row. Rows with fewer than 10 respondents report counts only.

Tests of Difference Between Segments

Table 7. Tests of difference between segmentsTable

Segment family	Unit	Chi-square	df	p	Holm p	Cramer's V	Test for p	Levels excluded
Outsourcing model	Engagement	20.79	15	0.065	0.458	0.24	Monte Carlo, simulated	None
Institution type	Engagement	23.92	18	0.125	0.753	0.28	Monte Carlo, simulated	a
Process family	Engagement	12.47	15	0.647	1.000	0.21	Monte Carlo, simulated	b
Class of work	Engagement	3.00	5	0.903	1.000	0.15	Monte Carlo, simulated	None
Engagement status and lifecycle stage	Engagement	10.68	15	0.875	1.000	0.17	Monte Carlo, simulated	None
Respondent role	Respondent	6.06	10	0.815	1.000	0.06	Monte Carlo, simulated	Customer Management Executive, n = 0
Vendor type	Respondent	15.74	25	0.925	1.000	0.06	Monte Carlo, simulated	None

Note: a Levels excluded: Payments and merchant acquiring (networks, acquiring, processing), n = 9, Credit card and payments issuing, n = 9, Private equity and private capital, n = 7. b Levels excluded: Trade, settlement and reconciliation, n = 9, Credit, underwriting and suitability, n = 8, Regulatory, audit, tax and legal, n = 6, Financial crime and client onboarding, n = 6.

Segment families recorded at engagement level were tested on engagements, and respondent role and vendor type on respondents. After Holm correction, the answers differed in no family. Individual contributors, engagement managers and portfolio directors named an AI certification in similar shares.

Association with Cancellation Rate

Cancellation rate is the share of engagements canceled by the study date, measured at engagement level (150 engagements, 56 canceled).

Table 8. Association with cancellation rate by answer levelTable

Answer level	Engagements	Canceled	Cancellation rate
None	115	43	37.39%
Databricks Generative AI Fundamentals	4	2	‡
Google Cloud Generative AI	2	0	‡
Azure AI Fundamentals	2	0	‡
DeepLearning.AI certificate	1	0	‡
AWS Certified Machine Learning	1	1	‡

Note: ‡ Base under 10. No statistic is reported.

Table 9. Association with cancellation rate, tests across answer levels by segmentTable

Segment	Engagements	Chi-square
Study set	125	a
Outsourcing model
CXM	36	a
ITO	34	a
BPO	30	a
KPO	25	a
Institution type
Asset management	20	a
Wealth management and private banking	17	a
Investment banking (advisory and capital-markets origination)	17	a
Insurance (life, property and casualty, health)	13	a
Commercial and business banking	11	a
Payments and merchant acquiring (networks, acquiring, processing)	9	‡
Credit card and payments issuing	9	‡
Retail and consumer banking	12	a
Markets and trading (sales and trading, global markets)	10	a
Private equity and private capital	7	‡
Process family
Corporate and back-office functions	42	a
Technology engineering and operations	30	a
Client servicing and communications	14	a
Trade, settlement and reconciliation	9	‡
Payments, cards and disputes	10	a
Credit, underwriting and suitability	8	‡
Regulatory, audit, tax and legal	6	‡
Financial crime and client onboarding	6	‡
Class of work
LOB Operational	63	a
Corporate Functional	62	a
Engagement status and lifecycle stage
Active, Build-Out		b
Active, Transition/Cut-Over		b
Active, Production		b
Canceled		b

Note: ‡ Base under 10. No statistic is reported. a Not tested: fewer than two levels of 10 or more engagements. b Not tested: no variation in cancellation.

Only one answer, None, was the plurality answer on 10 or more engagements (115), so no test across answers was computed at engagement level. At respondent level, with None as a level and standard errors clustered by engagement, cancellation did not differ across the answers (F = 0.67, df = 5 and 149, p = 0.650, n = 788 respondents).

Association with Customer Satisfaction

Customer satisfaction is the percentage recorded against each engagement.

Table 10. Association with customer satisfaction by answer levelTable

Answer level	Engagements	Mean of the measure	SD
None	72	64.9	11.5
Databricks Generative AI Fundamentals	2	‡
Google Cloud Generative AI	2	‡
Azure AI Fundamentals	2	‡
DeepLearning.AI certificate	1	‡
AWS Certified Machine Learning	0	No engagements

Note: ‡ Base under 10. No statistic is reported.

Table 11. Association with customer satisfaction, tests across answer levels by segmentTable

Segment	Engagements	Welch F
Study set	79	a
Outsourcing model
CXM	23	a
ITO	22	a
BPO	18	a
KPO	16	a
Institution type
Asset management	11	a
Wealth management and private banking	11	a
Investment banking (advisory and capital-markets origination)	8	‡
Insurance (life, property and casualty, health)	9	‡
Commercial and business banking	7	‡
Payments and merchant acquiring (networks, acquiring, processing)	8	‡
Credit card and payments issuing	5	‡
Retail and consumer banking	8	‡
Markets and trading (sales and trading, global markets)	7	‡
Private equity and private capital	5	‡
Process family
Corporate and back-office functions	25	a
Technology engineering and operations	22	a
Client servicing and communications	6	‡
Trade, settlement and reconciliation	4	‡
Payments, cards and disputes	9	‡
Credit, underwriting and suitability	7	‡
Regulatory, audit, tax and legal	3	‡
Financial crime and client onboarding	3	‡
Class of work
LOB Operational	39	a
Corporate Functional	40	a
Engagement status and lifecycle stage
Active, Build-Out	27	a
Active, Transition/Cut-Over	35	a
Active, Production	17	a
Canceled		b

Note: ‡ Base under 10. No statistic is reported. a Not tested: fewer than two levels of 10 or more engagements. b Not tested: canceled engagements fall outside the active population.

Only one answer, None, was the plurality answer on 10 or more engagements (115), so no test across answers was computed at engagement level. At respondent level, with standard errors clustered by engagement, customer satisfaction did not differ across the answers (F = 0.77, df = 5 and 149, p = 0.576, n = 788 respondents).

Association with the Engagement Indicator Mean

The engagement indicator mean is the mean of an engagement's ratings across the rated indicators. At respondent level the measure is the respondent's own mean rating. The certification is not a rated indicator and does not enter either mean.

Table 12. Association with the engagement indicator mean by answer levelTable

Answer level	Engagements	Mean of the measure	SD
None	115	4.0	1.5
Databricks Generative AI Fundamentals	4	‡
Google Cloud Generative AI	2	‡
Azure AI Fundamentals	2	‡
DeepLearning.AI certificate	1	‡
AWS Certified Machine Learning	1	‡

Note: ‡ Base under 10. No statistic is reported.

Table 13. Association with the engagement indicator mean, tests across answer levels by segmentTable

Segment	Engagements	Welch F
Study set	125	a
Outsourcing model
CXM	36	a
ITO	34	a
BPO	30	a
KPO	25	a
Institution type
Asset management	20	a
Wealth management and private banking	17	a
Investment banking (advisory and capital-markets origination)	17	a
Insurance (life, property and casualty, health)	13	a
Commercial and business banking	11	a
Payments and merchant acquiring (networks, acquiring, processing)	9	‡
Credit card and payments issuing	9	‡
Retail and consumer banking	12	a
Markets and trading (sales and trading, global markets)	10	a
Private equity and private capital	7	‡
Process family
Corporate and back-office functions	42	a
Technology engineering and operations	30	a
Client servicing and communications	14	a
Trade, settlement and reconciliation	9	‡
Payments, cards and disputes	10	a
Credit, underwriting and suitability	8	‡
Regulatory, audit, tax and legal	6	‡
Financial crime and client onboarding	6	‡
Class of work
LOB Operational	63	a
Corporate Functional	62	a
Engagement status and lifecycle stage
Active, Build-Out	27	a
Active, Transition/Cut-Over	35	a
Active, Production	17	a
Canceled	46	a

Note: ‡ Base under 10. No statistic is reported. a Not tested: fewer than two levels of 10 or more engagements.

Only one answer, None, was the plurality answer on 10 or more engagements (115), so no test across answers was computed at engagement level. At respondent level, with standard errors clustered by engagement, the respondents' own mean did not differ across the answers (F = 0.31, df = 5 and 149, p = 0.909, n = 788 respondents).

Association with Intent to Leave

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 by up to 798 respondents.

Table 14. Association with intent to leave by answer levelTable

Answer level	Engagements	Mean of the measure	SD
None	115	6.8	2.4
Databricks Generative AI Fundamentals	4	‡
Google Cloud Generative AI	2	‡
Azure AI Fundamentals	2	‡
DeepLearning.AI certificate	1	‡
AWS Certified Machine Learning	1	‡

Note: ‡ Base under 10. No statistic is reported.

Table 15. Association with intent to leave, tests across answer levels by segmentTable

Segment	Engagements	Welch F
Study set	125	a
Outsourcing model
CXM	36	a
ITO	34	a
BPO	30	a
KPO	25	a
Institution type
Asset management	20	a
Wealth management and private banking	17	a
Investment banking (advisory and capital-markets origination)	17	a
Insurance (life, property and casualty, health)	13	a
Commercial and business banking	11	a
Payments and merchant acquiring (networks, acquiring, processing)	9	‡
Credit card and payments issuing	9	‡
Retail and consumer banking	12	a
Markets and trading (sales and trading, global markets)	10	a
Private equity and private capital	7	‡
Process family
Corporate and back-office functions	42	a
Technology engineering and operations	30	a
Client servicing and communications	14	a
Trade, settlement and reconciliation	9	‡
Payments, cards and disputes	10	a
Credit, underwriting and suitability	8	‡
Regulatory, audit, tax and legal	6	‡
Financial crime and client onboarding	6	‡
Class of work
LOB Operational	63	a
Corporate Functional	62	a
Engagement status and lifecycle stage
Active, Build-Out	27	a
Active, Transition/Cut-Over	35	a
Active, Production	17	a
Canceled	46	a

Note: ‡ Base under 10. No statistic is reported. a Not tested: fewer than two levels of 10 or more engagements.

Only one answer, None, was the plurality answer on 10 or more engagements (115), so no test across answers was computed at engagement level. At respondent level, with standard errors clustered by engagement, intent to leave did not differ across the answers (F = 0.75, df = 5 and 149, p = 0.591, n = 742 respondents).

Comparison of Associations Across the Four Measures

Table 16. Associations across the four measuresTable

Measure	Effect size	n	Value	p	Equivalent to zero	Rank among the four
Cancellation rate	Not tested: fewer than two levels of 10 or more engagements
Customer satisfaction	Not tested: fewer than two levels of 10 or more engagements
Engagement indicator mean	Not tested: fewer than two levels of 10 or more engagements
Intent to leave	Not tested: fewer than two levels of 10 or more engagements

No association with a study measure was tested at engagement level, so no association was ranked or compared.

Clustering and Effective Sample Size

Table 17. Clustering and effective sample sizeTable

Statistic	Value	Reading
Clusters (engagements)	150	Engagements with one or more answers.
Mean respondents per engagement	5.3	Answers per engagement, among engagements with one or more answers.
ICC(1)	−0.01	Share of the variance that lies between engagements.
Design effect	1.00	One plus the product of the ICC and the mean cluster size less one.
Nominal respondent n	788	Every answer, with “I do not know” and “I prefer to not answer” counted as levels, as in the plurality rule.
Effective n	788	Nominal n divided by the design effect.
Cluster bootstrap applied	No	Applied to every respondent-level interval where the ICC is 0.05 or above.

Answers did not cluster within engagements (ICC −0.01): respondents on the same engagement were no more alike in their certifications than respondents chosen at random. The design effect was 1.00, and the effective n equaled the 788 answers.

Equivalence Tests of Null Results

Table 18. Equivalence tests of null resultsTable

Measure	n	Omega squared	Equivalence bound	Equivalence p	Equivalent to zero	Achieved power	MDE at power 0.80
Cancellation rate	a
Customer satisfaction	a
Engagement indicator mean	a
Intent to leave	a

Note: a Not tested: fewer than two levels of 10 or more engagements.

No equivalence test was computed, because no association with a study measure was tested at engagement level. The absence of a test is not evidence of no association.

Robustness Checks

Table 19. Robustness checksTable

Check	Measure	Primary p	Check p	Check statistic	Conclusion holds
Kruskal-Wallis in place of Welch F	Customer satisfaction	Not computed	Not computed	Not computed	Not computed
Kruskal-Wallis in place of Welch F	Engagement indicator mean	Not computed	Not computed	Not computed	Not computed
Kruskal-Wallis in place of Welch F	Intent to leave	Not computed	Not computed	Not computed	Not computed
Active engagements only	Engagement indicator mean	Not computed	Not computed	Not computed	Not computed
Active engagements only	Intent to leave	Not computed	Not computed	Not computed	Not computed
Respondent level, cluster-robust Wald F by engagement	Cancellation rate	Not computed	0.650	0.67	Not computed
Respondent level, cluster-robust Wald F by engagement	Customer satisfaction	Not computed	0.576	0.77	Not computed
Respondent level, cluster-robust Wald F by engagement	Engagement indicator mean	Not computed	0.909	0.31	Not computed
Respondent level, cluster-robust Wald F by engagement	Intent to leave	Not computed	0.591	0.75	Not computed

No primary test was computed, so no check is scored as holding or failing. The checks that compare engagement-level tests were not computed. The respondent-level test found no difference at 0.05 for any of the four measures (p from 0.576 to 0.909).

Measure Independence and Part-Whole Correction

Each of the four measures was computed from data the indicator does not contain. Any association would therefore be a finding and not an arithmetic consequence.

Comparison with the Study Norm

A categorical indicator has no signed correlation with the study measures and therefore no profile against the study norm.

Interpretation of the Data

Most respondents held no AI certification, and the certifications held were spread evenly across the five named and across the roles. With None included, the answers did not separate the respondents on cancellation, customer satisfaction, the engagement indicator mean or intent to leave.

Limitations

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.

None was the plurality answer on most engagements and no other answer was the plurality on 10 or more, so the engagement-level tests that carry the other indicators could not be computed, and the findings rest on the respondent-level tests.

The certification is the respondent's report and not a record. A respondent who held more than one AI certification gave one answer.

Tests by respondent role and by vendor type treated respondents as independent.

Definitions

Table 20. DefinitionsTable

Term	Definition
Response scale	The answer is one of the listed AI certifications shown in the indicator specification, or None. The indicator is not a rating. No distressed range applies.
Answering base	Respondents who gave one of the listed answers, None included.
Non-answers	“I do not know” and “I prefer to not answer” are counted separately. They are excluded from the answering base and from the shares.
Indicator class	Respondent level: a property of the respondent rather than of the engagement.
Failure category	The five failure categories of the Study Taxonomy apply only to the rated indicators. The indicator carries none.
Assigned answer	The answer given by the largest number of an engagement's respondents, with None counted as an answer. A tied engagement carries no assigned answer.
Clustered test	A test whose standard error allows for the answers of respondents on the same engagement to be alike.

Sources and Provenance

Table 21. Sources and provenanceTable

Source	Value
Study data	Study Data.xlsx, Data sheet, column C016, sha256 219eba0457274d04
Statistics engine	compute_statistics.py, sha256 738319eef764602b
Statistics record	Computed Statistics.json, record C016, sha256 1ae50a3637fc9370
Computed	October 13, 2026, 04:57 UTC
Template	Indicator Template v2, version 2.0 of 10 October 2026
Figures placed by script	1,122

Continue the report

Executive summary →

Overview

Executive Summary

Four findings from 150 engagements

Read →

Introduction

Scope, evidence, and how the report is organized

Read →

Part 1

Engagement Outcomes

Poor results on every measure, and what separates them

Read →

Part 2

The Five Failure Categories

Where failure concentrates, category by category

Read →

Part 3

Governance, Oversight, and Control

Controls that existed in name only

Read →

Part 4

The Delivery Workforce

Early in their careers, dissatisfied, and many intending to leave

Read →

Part 5

Manual Alteration

How the delivery team altered the work to sustain it

Read →

Part 6

Executive Action Plan

Seven immediate actions and a first-year plan

Read →

Appendices

Definitions, indicator inventory, method, and disclosures

Read →

Data

Study Data

Deep-dive analyses of the study indicators, published in stages

Read →

What AI certifications do you have?

On this page

Indicator Specification

Summary of Findings

Distribution of Responses

Descriptive Statistics by Segment

Tests of Difference Between Segments

Association with Cancellation Rate

Association with Customer Satisfaction

Association with the Engagement Indicator Mean

Association with Intent to Leave

Comparison of Associations Across the Four Measures

Clustering and Effective Sample Size

Equivalence Tests of Null Results

Robustness Checks

Measure Independence and Part-Whole Correction

Comparison with the Study Norm

Interpretation of the Data

Limitations

Definitions

Sources and Provenance

Continue the report