Yearly running cost Assuming 12,000 samples a year
By hand $132K / year 2,400 hours of work
Subscription $2K+ / year
This workflow $619 / year Machine usage only; setup, hosting and review are extra. How far should the testing go? Test evidence you upload, or also manage evidence requests and file approved workpapers.
One sample includes up to 3 supporting documents and 10 pages.
Test uploaded evidence A testing sheet in your template, with source references. $1,224 – 3,720 setup Add evidence requests and filing Also request evidence and file approved workpapers. $1,872 – 5,640 setup
Where do you keep engagement files?CaseWare AuditBoard A document portal such as Suralink SharePoint or a shared drive Other Your auditors review and sign off each test and decide what is an exception. It never concludes on an account, changes a sample or signs a workpaper.
What is included Match uploaded evidence to each selected sample, read the tested attributes, compare them with the recorded amount, date and period, mark agreed items and exceptions, and draft the testing sheet in your template.
Start with your sample selection, the test procedure and attributes, your workpaper template and a set of past tests with known results. The connected option also drafts itemised evidence requests for the client and reads and files evidence in your audit platform.
How reliable should the test results be? Set targets for matching evidence, reading tested attributes and flagging exceptions.
Standard Correct matching and attribute reading on clear evidence, with unreadable or ambiguous items left for an auditor. Evidence matched to the right sample: ≥97% Test attributes read correctly: ≥97% Included Strict A second independent read of each tested attribute and more tests on past samples. Evidence matched to the right sample: ≥99% Test attributes read correctly: ≥99% Setup +$288 – 840 My own targets Agree your own acceptance requirements with the provider. Quote separately
How these standards are measured Test on held-out past samples with labelled evidence, attribute values and exceptions, including multi-page scans with several samples, missing documents, partial payments and documents from the wrong period. Measure matching, attribute reading and exception flagging separately. A result without a document and page counts as a failure.
Targets for your selected standard What is checked Target Evidence matched to the right sampleSupporting documents linked to the correct sample item, divided by all labelled sample-document links in the test set. A document linked to the wrong sample counts as a failure even if its values look right. ≥97% Test attributes read correctlyAttribute values such as amount, date, counterparty, description and period that match the labelled value on the document, divided by all labelled attribute values. An attribute marked unreadable for a person to check is not counted as wrong. ≥97% Labelled exceptions flaggedLabelled exceptions, such as a different amount, a date outside the period, a missing approval or no matching document, that are flagged, divided by all labelled exceptions. ≥95% Results with document and pageEvery attribute result must cite the document and page it came from, so the reviewer can open the evidence in one step. ≥100%
Drafted results are prepared work for an auditor to review, not audit evidence on their own. Your engagement team still reviews each sample, decides on exceptions and signs off.
How quickly should each sample be tested? Choose how soon after the evidence arrives each sample result is ready for review. Auditor review time is separate.
Within 2 minutes Included Within 60 seconds Setup +$144 – 480 Within 30 seconds Setup +$288 – 840
Timing details Time from evidence upload to a drafted sample result, including reading pages, matching, comparison, queueing and retries. Very large scans may need agreed limits. Confirm sample volume and hardware with your provider.
The target applies to at least 95% of agreed test runs, with 8 in progress at a time.
How much do you want to spend per sample? Set the AI processing budget for testing each audit sample.
Up to $0.25 Included Up to $0.15 Setup +$72 – 360 Up to $0.10 Setup +$216 – 600
Cost details Includes model calls, retries and a shared hosting allocation. Audit platform licences, your auditors’ review time and the calling agent are separate. Re-running a sample after new evidence arrives is another run.
Reference machine cost: $0.09 – 0.17 per sample at 1000 samples a month. The selected cap is a target to test, not a replacement for this estimate.
Where do you want it to run? Run it in your cloud or on your own server. Choose whether evidence pages may go to an approved AI service.
Run it onYour cloud Local environment AI model accessApproved model API Private model only
Data and access details Runs in a cloud account your firm controls, with access by engagement team and access logs.
Only the evidence pages and attributes for a sample go to the selected external model. Agree retention and confirm this fits your firm’s confidentiality rules and engagement terms.
Keep client evidence, sample selections and workpapers inside your environment. Send only the pages needed for a sample to the approved model, under agreed retention terms. Private model only keeps evidence on your hardware; the workflow still connects to your own audit platform.
How do you want to use it? Choose where you want to use it. You can select more than one.
A review web page Upload the sample selection and evidence, review each result next to the document and export the testing sheet. Included My existing AI agent Let your current agent run a test on a sample selection and read back the results and exceptions. Setup +$72 – 240 Inside my spreadsheet Run the test and see results with links to the evidence from inside your spreadsheet workpaper. Setup +$216 – 600