ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
CUBYTS - AI COST CALCULATOR
2
Enter your own volumes in the shaded cells. Everything else recalculates.
3
Shaded cells are yours to change. Unshaded cells are calculated - do not overwrite.
4
5
1. CLOUD PROVIDER
6
ProviderAzureChoose from the drop-down. All three are compared at the bottom.
7
8
2. ENGAGEMENT
9
Engagement duration6months
10
Sprint length2weeks
11
Engineers8people
12
Sprints in the engagement13sprints
13
14
3. CODE AND DELIVERY
15
Repositories3repos
16
Tracked branches per repository2branchesEach branch is indexed separately - no sharing between them
17
Pull requests per sprint30PRsThe largest single driver of cost
18
Files changed per pull request4filesReview cost is per file, not per pull request
19
Pushes per pull request2pushesEvery push re-reviews the changed files
20
Commits per sprint, all branches450commitsIncludes branch commits that never open a pull request
21
Repository-branches to onboard6repo-branches
22
File reviews per sprint240reviews
23
24
4. STORIES AND WORK ITEMS
25
Starting story backlog50stories
26
New stories per month50stories
27
Edits per new story per month3editsAvailable from the Jira changelog
28
Existing stories edited per month30%of backlogAvailable from the Jira changelog
29
Stories long enough to analyse70%of editsStories under 50 words are not analysed and cost nothing
30
Total stories created300stories
31
Story edits over the engagement1,215edits
32
Story edits per sprint93.5edits
33
Story edits analysed per sprint65.4edits
34
35
5. DOCUMENTS
36
Confluence pages30pagesTechnical design, architecture, HLD, LLD
37
Confluence pages revised per month30%of pages
38
Pages of text in SharePoint / Teams6,000pagesText only. Images, decks and recordings do not count
39
Document revisions per sprint4.2revisions
40
41
6. RECURRING COST, PER SPRINT
42
Priced at the provider selected in section 1.
43
CapabilityVolumeUnit rateCost
44
Code review240.0$0.2861$68.66file reviews
45
PR documentation, dependency risk & drift30.0$4.7700$143.10pull requests
46
Engineering agents8.0$5.0000$40.00engineers
47
Code indexing450.0$0.0711$32.00commits
48
Requirement drift analysis65.4$0.0420$2.75story edits
49
Work-item processing93.5$0.0012$0.11story edits
50
Document processing4.2$0.0190$0.08revisions
51
Platform services1.0$14.0000$14.00per sprint
52
Total per sprint$300.70
53
54
7. ONE-OFF SET-UP COST
55
ItemVolumeUnit rateCost
56
Repository onboarding6$12.3800$74.28repo-branches
57
Story backlog ingestion300$0.0012$0.36stories
58
Confluence ingestion30$0.0300$0.90pages
59
SharePoint / Teams ingestion6,000$0.0038$22.80pages of text
60
Total one-off$98.34
61
62
8. YOUR TOTAL
63
Azure - total for the engagement$4,007
64
Per month$667.90
65
Per engineer, per month$83.49
66
67
9. ALL THREE PROVIDERS
68
Same inputs, priced against each provider.
69
ProviderPer sprintOne-offTotal
70
GCP$424.00$111.47$5,623
71
Azure$300.70$98.34$4,007
72
AWS$423.57$110.87$5,617
73
74
10. HOW TO READ THIS
75
- Accuracy is roughly plus or minus 50 per cent. Prices and call counts are confirmed against production usage. Token volumes scale with repository, file and commit patch sizes, so the figures move up or down with those.
76
- Median values throughout. No maximum is presented as a baseline.
77
- Pull request volume, files per pull request and pushes per pull request drive most of the total. Story and document counts drive very little.
78
- SharePoint and Teams ingestion scales with whatever content is connected, not with engineering activity. Agree the folder scope up front.
79
- Costs shown are AI model charges only. Hosting, storage and database costs are covered separately.
80
81
Prepared 20 August 2026. Confidential - Cubyts & Deloitte only.
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100