,

Implementing Audit Management Software: The First 120 Days and Migrating Off Excel

Signing the contract is the easy part. What decides whether an audit management platform earns its subscription happens in the weeks after signature: whether the function settles its methodology before it starts configuring screens, whether the workpaper archive comes across intact, whether the admin role gets real hours instead of a side-of-desk assignment, and whether auditees are trained before a request portal replaces their inbox. Gartner’s own research on this market makes the point directly. Its September 2026 survey of 259 audit departments found that 84% have adopted audit management software, then added the finding that should worry every buyer: departments that buy the same platform realize very different value depending on whether they standardize their methodology, integrate workflows and reduce their reliance on spreadsheets and email.

This guide is a vendor-neutral walk through the first 120 days of an implementation: the methodology decisions to lock before configuration starts, a week-by-week plan, what to migrate from Excel or SharePoint against what belongs in a read-only archive, a data mapping table, the admin role most functions underbudget, a training plan by role, how to bring auditees along, the measures that show whether it is working at 30, 90 and 180 days, and the failure modes behind most disappointing rollouts. It is part of the site’s independent buyer’s guide to internal audit software, follows the review method set out there, and pairs with selecting an audit management system for readers who have not chosen a platform yet and audit software vs Excel and SharePoint for readers still deciding whether to move at all.

What this guide is for. This is an implementation guide, not a single-product review. It sets out the sequence, roles and decisions common to the platforms reviewed on this site, illustrated with public evidence from several of them, so a function moving off spreadsheets or off an older system can plan the first 120 days before it signs anything.

Evidence. Research-based: vendor documentation and release notes, public procurement records, third-party pricing data, verified user reviews on Gartner Peer Insights and G2, and analyst coverage. We have not implemented any of the products named in this guide hands-on.

Last verified. 27 September 2026.

In this guide

Decide your methodology before you configure anything

The single decision that predicts whether an implementation goes well is not which product a function bought. It is whether the methodology existed before configuration started. Gartner’s September 2026 release named the traits shared by the audit functions that get value from their software: they standardize the methodology, integrate the workflow instead of reproducing paper forms on screen, use the system to actually reduce spreadsheets and email rather than running both in parallel, align audit, assurance and IT before they select a system, and start from workflow needs rather than a feature list. Every one of those is a decision, not a configuration setting, and every one of them is cheaper to make on a whiteboard in week one than to unwind in a live system in year two.

  • Risk rating scale and audit universe structure. Decide how entities are scored and grouped before the universe gets built in the system; a scale invented during configuration rarely survives contact with the next risk assessment.
  • The RCM or workpaper standard. The columns, tick-mark legend and review levels the function already uses become the configuration brief; the site’s risk and control matrix template is a reasonable starting point if none exists yet.
  • Engagement life cycle stages and sign-off levels. Planning, fieldwork, review and reporting each need an owner and an exit condition, not just a status field.
  • Issue rating and escalation rules. The scale, the aging clock and who gets escalated to, matching the fields in the site’s issue log template if the function has been tracking these in a spreadsheet.
  • Reporting hierarchy. What an engagement report contains versus what reaches the audit committee, so the system’s report and dashboard layouts get built once, not rebuilt after go-live.

None of this needs to be perfect. It needs to be written down and signed off by the chief audit executive before the implementation team opens the admin console, because a system configured to match whatever happened to be open first tends to mirror a spreadsheet’s structure without a spreadsheet’s adaptability, which is a worse outcome than either extreme on its own.

The 120-day plan

The five phases below add up to 17 weeks, close enough to 120 days that the shorthand has stuck across the vendors reviewed on this site. Treat the weeks as a starting allocation, not a promise: a two-person function with a small migration can compress it, and a function integrating single sign-on, several data sources and a professional-services partner should expect to stretch it.

PhaseWeeksWhat happensDone when
Methodology decisionsWeeks 1-2Lock the risk scale, RCM or workpaper standard, life cycle stages, and issue and reporting rules; name an internal project lead separate from the vendor’s implementation manager; confirm which existing templates become the configuration baselineThe methodology document is signed off by the CAE and handed to the vendor’s team as the configuration brief
ConfigurationWeeks 3-6Build the audit universe, risk model, workpaper templates, issue workflow, user roles, sign-off levels and report layouts; connect single sign-on; agree what is out of scope for go-liveA test engagement can be opened, worked and closed in the system end to end
MigrationWeeks 7-10Move the open plan, current audit universe, open issues and action plans, and templates into the live system using the data mapping below; export closed engagements into a read-only archive; check a sample against the source recordsEvery field in the mapping table has a home, and a sample of migrated records matches the source line by line
PilotWeeks 11-14Run one or two live engagements end to end in the system, with the old process kept only as a fallback; collect issues from pilot auditors and pilot auditees; fix configuration rather than build workaroundsA pilot engagement has closed in the system with a report issued and no workaround duplicating work in a spreadsheet
Go-liveWeeks 15-17Train the remaining staff and the wider auditee population, retire parallel tracking, set the admin’s ongoing cadence, and put the 30, 90 and 180-day check-ins on the calendarEvery open engagement runs in the system and the old tracker is locked to read-only

The public record shows how wide the real range is. G2’s aggregated user data puts the average time to implement TeamMate at four months, in line with the plan above. Cost tells a similar story at different scales. A City of Norman, Oklahoma council quote dated 28 February 2025 priced a two-user TeamMate+ implementation at $15,630 against a $6,150.88 annual subscription, well above the 20-50% of first-year subscription that Vendr’s buyer data suggests for a typical Wolters Kluwer deal, because small deployments often carry a flat onboarding fee out of proportion to a small subscription. The response from Optro (formerly AuditBoard) to a West Virginia Department of Transportation bid, a bid response rather than a confirmed award, priced implementation at $50,000 against a $164,000 annual quote for 25 users, close to the top of the 10-30% of subscription that Vendr’s Optro data shows. Ask for the implementation figure as a fixed number before signing, not a range.

What to migrate, and what to archive

Not everything in the old spreadsheet or the old system belongs in the new one. The distinction that keeps a migration fast and a new system’s search results useful is between what the function still acts on and what it only needs to keep.

WhatMigrate live, or archive read-onlyWhy
Audit universe and risk assessmentMigrateBecomes the system’s live universe; carry the risk scores across as a starting point for the next assessment, not as a permanent record
Current and board-approved planMigrateThe plan the board approved needs to be workable in the new system from day one
Open issues and action plansMigrate, with full historyAging clocks and escalation history must carry forward; cutover should not reset them
Control and test program templatesMigrateBecomes the new system’s template library, replacing the versions kept on a shared drive
Closed engagement workpapersArchive read-onlyKept for the function’s retention period and prior-year reference, not for active editing
Prior audit committee reports and dashboardsArchive read-onlyHistorical record; the new system will build the next one

Get the closed-engagement export format in writing before signing, whatever the function decides. The site’s guide to the Global Internal Audit Standards maps the engagement documentation requirement, Standard 14.6, in full; in practice it means a workpaper archive that only opens inside a system the function has since left is not a retained record in any useful sense. The clearest illustration on record involves TeamMate: when the Office of the Auditor General of Canada replaced its legacy TeamMate AM with Caseware, it completed a privacy impact assessment, because moving audit records that contain personal information to a new system is a control decision, not a technical detail. Functions with any personal data in their workpapers, which is most of them, should ask the same question before cutover, not after.

Mapping your data: from spreadsheet fields to system fields

Every migration eventually comes down to a spreadsheet of fields, mapped by hand. The table below is a starting map for the fields that cause the most trouble, drawn from the fields most audit trackers already have.

Your spreadsheet or tracker fieldTypical system fieldWhere it bites
Risk or control ID (free text)Control or risk identifierOften becomes a new system-generated ID; keep the old one as a cross-reference field so closed workpapers still resolve
Owner name (free text)Owner, linked to a user recordMust resolve to an actual account, not a text string; this is where stale org-chart data surfaces
Issue rating (High/Medium/Low or 1-5)Issue severity, on the system’s own scaleConfirm the mapping works in both directions before cutover, since aggregated issue reporting depends on a consistent scale across old and new issues during the transition
Due date or target dateTarget remediation dateTest with an item that is already overdue; some systems recalculate aging from this field in ways that surprise a first-time admin
Workpaper file path or reference numberEvidence or workpaper attachmentDecide file naming up front, and whether legacy files are re-uploaded or only linked from the read-only archive
Auditable entity or department code (flat list)Auditable entity, in a hierarchyA flat spreadsheet list rarely maps one-to-one onto a nested universe; rebuilding the hierarchy is real work, not a data-entry task
Reviewer initials in a cellReview or sign-off workflow stepAn audit trail needs a timestamped action from a named account, not a character typed into a cell

Run this exercise before migration week, not during it. A mapping table with seven rows takes an afternoon to build and a week to argue about; discovering the gaps live, with a pilot engagement waiting, costs far more than either.

The admin role, and how many hours it needs

Every function underestimates this line item once. The admin owns the configuration after go-live: new users, permission changes, template updates, the annual methodology refresh, and whatever the vendor changes in the next release. Public pricing evidence shows what the role is worth when it is skipped: Vendr’s buyer notes on Onspring record that doing implementation in-house rather than through a partner runs $50-$75 more per hour, a real cost for work a trained admin would otherwise absorb into a normal week.

Function sizeAdmin timeWhat the admin doesWho else is involved
First system for a small team (1-5 auditors)A few hours a week once live, layered onto an existing roleUser and permission changes, template tweaks, fielding questionsThe vendor’s implementation package, not a separate partner
Mid-size function (6-25 auditors)A named admin at roughly a quarter to half of one roleConfiguration, integrations, the annual methodology refresh, user supportThe vendor’s professional services for major changes
Large or global function (25+ auditors)One or more dedicated roles, often a small audit-technology teamIntegrations, single sign-on, AI feature governance, multi-year roadmapThe vendor’s account team directly, rather than a reseller

The implementation line is usually more negotiable than the subscription itself. The site’s guide to internal audit software pricing covers how to negotiate it, and the multi-year discounts vendors publish through Vendr, which mostly apply to the subscription, are a separate lever from the implementation fee.

Training, by role

One all-hands demo does not train a function. Each role needs something different, at a different point in the 120 days.

RoleWhat they needWhen
Audit leadershipMethodology decisions, report and dashboard configuration, what the board will seeWeeks 1-2, then a refresher before go-live
Audit staff (performers and reviewers)Workpaper creation, evidence linking, the sign-off workflowThe pilot itself, weeks 11-14, plus a refresher at go-live
The adminConfiguration, user and permission management, integrations, how to read release notesWeeks 3-6, before configuration begins in earnest
Auditees and the first lineSubmitting evidence, responding to requests, updating action plans in the portalDuring the pilot and again at go-live, timed to their first real request

Vendor-run courses exist for the admin role specifically. Treat the cost as part of the implementation budget, not an afterthought: a self-taught admin usually costs the function more in slow tickets and reversed configuration mistakes than a course costs in dollars.

Bringing auditees and the first line along

Auditees notice a new system before auditors do, because it is often the first thing that changes about how they experience an audit: a portal replacing an email thread, a due date replacing a verbal understanding. A rollout that only trains the audit team and assumes auditees will work out the portal on their own is the fastest way to see PBC responses still arriving by email months after go-live, with the admin quietly re-keying them into the system so the record looks complete. That workaround is invisible in a status report and expensive in admin hours, which is why the training plan above gives auditees their own row rather than folding them into a single kickoff email.

Run the first real request cycle during the pilot, with a pilot auditor walking a cooperative auditee through the portal directly rather than sending a help article. Keep email open through the pilot as a stated fallback, then close it deliberately at go-live rather than letting it linger. The site’s auditee communication pack has notification and kickoff templates to adapt, the PBC survival guide is worth sending to auditees directly, and the being audited hub collects the rest of the site’s guides for the other side of the desk.

Success measures at 30, 90 and 180 days

“It is working” is not a measure. The three checkpoints below give a function something concrete to report to the audit committee about its own implementation.

CheckpointWhat working looks likeRed flag
30 daysThe pilot engagement is open and being worked in the system; the admin is fielding configuration questions, not access-reset tickets; no auditor has quietly reverted to a parallel spreadsheetThe pilot’s workpapers are still drafted in Word or Excel and pasted in afterward
90 daysGo-live is complete, every open engagement runs in the system, and issue aging is trusted enough that the audit committee package is pulled from the system rather than rebuilt by handTwo versions of the issue log exist and disagree with each other
180 daysThe function has been through one full reporting cycle end to end, the admin has made a methodology or template change without vendor help, and QAIP metrics come from the system rather than a hand tallyThe team can describe what the system cannot do, but not what it has actually saved

The site’s QAIP playbook and audit committee deck template are natural next reads once the system is producing its own numbers instead of a hand-built slide.

Common failure modes, and how to avoid each

Most disappointing rollouts fail for one of a small number of reasons, nearly all avoidable by following the plan above in order. The site’s guide to audit software buying mistakes covers the selection-stage version of this list; the table below covers what goes wrong after the contract is signed.

Failure modeWhy it happensThe fix
Configuring before deciding methodologyThe system mirrors whatever was open first; the function spends year two reconfiguringLock methodology in weeks 1-2, before configuration starts
Migrating everything, including closed workpapersThe live universe bloats, search gets slower, and retention questions get dragged into day oneArchive closed engagements read-only instead of migrating them live
Skipping the pilot and going straight to go-liveConfiguration problems surface for the whole function at once instead of two auditorsRun the pilot phase as written, even under schedule pressure
Treating the admin role as a side projectConfiguration drifts, tickets pile up, and auditors route around the systemAllocate real hours, sized to the admin table above
Telling auditees instead of training themThe request portal goes unused and PBC responses keep arriving by emailRun auditee onboarding during the pilot, not after go-live
Turning on AI features unsupervised at go-liveUntested outputs reach a live engagement before anyone has learned what to checkFollow the site’s guide to evaluating AI in audit software before trusting an output
No rollback plan for the pilotA broken pilot with nowhere to fall back erodes trust in the whole rollout before go-liveKeep the old process live as a stated fallback, but only through the pilot phase

Most of these compound: a function that skips the pilot has usually also under-resourced the admin, because the same schedule pressure caused both decisions. The checklist below is the single-page version of the whole plan.

The cutover checklist

Copy this into the project’s own tracker rather than trusting memory for the day the old tracker goes read-only.

Implementation cutover checklist

Methodology signed off. The risk scale, RCM or workpaper standard, life cycle stages, and issue and escalation rules are approved by the CAE, in writing, before configuration began.

Data mapped and validated. Every field in the mapping table has a home in the new system, and a sample of migrated records has been checked line by line against the source.

Closed archive readable. Closed-engagement workpapers are exported or accessible read-only, in a format confirmed to remain usable for the function’s full retention period.

Pilot closed clean. At least one engagement has run from planning to report inside the system, with no workaround duplicating the work in a spreadsheet.

Admin resourced. A named admin has the hours set out above, not a side-of-desk assignment with no allocated time.

Training complete, by role. Leadership, staff, the admin and auditees have each had the training set out above, not a single all-hands demo.

Reporting rebuilt in the system. The audit committee package can be produced from the system’s own data, not assembled by hand in a slide deck.

Old tracker set read-only. The spreadsheet or shared drive that ran the function before go-live is locked to edits, with a named place it still lives for reference.

Check-ins scheduled. The 30, 90 and 180-day reviews are on the calendar with a named owner, not left to happen informally.

Worked example: MidState Beverage’s 120 days

MidState Beverage, the three-state drinks distributor with a six-person audit function used elsewhere on this site, spent its first six years running the function out of a shared spreadsheet and a SharePoint folder for workpapers. By year seven the issue log had outgrown anyone’s ability to keep its statuses current, and an external quality assessment flagged inconsistent workpaper review evidence. The audit director ran the site’s RFP method and demo script, picked a platform sized for a function of five or six auditors, and used the plan above almost exactly as written.

Weeks 1 and 2 went to methodology: MidState’s existing RCM columns and its three-tier issue rating became the configuration brief, signed off before the vendor’s kickoff call. Migration in weeks 7 through 10 moved 40 open issues and the current-year plan; the 65 closed engagements sitting in the SharePoint folder were exported as PDFs into a named archive folder rather than migrated live, with the export format checked against the function’s seven-year retention policy first. The pilot ran on the next scheduled engagement, an accounts payable audit, with the fallback spreadsheet kept open but unused after the first week. Admin duties landed on the senior auditor who already owned the RCM template, at roughly three hours a week once live, in line with the small-team allocation above rather than a separate hire. By 90 days the issue log lived in one place, the outcome the site’s own guide to moving issue tracking off Excel describes; by 180 days the director pulled the next external quality assessment’s workpaper sample directly from the system, instead of assembling it by hand from the old SharePoint folder as in prior years.

Questions about implementing audit management software

How long does an implementation actually take?

There is no fixed number. G2’s aggregated user data puts TeamMate’s average at four months, and the 17-week, five-phase plan in this guide is a reasonable default for a first system. Functions with heavier migration, more integrations, or a professional-services partner doing the work should expect to run longer, not shorter.

Should we migrate closed workpapers into the new system, or leave them where they are?

Leave them, archived read-only, rather than migrating them live. Closed engagements do not need to sit in the active universe, and moving them all in tends to slow search and drag retention questions into week one. Confirm the export or read-only access format before signing, and keep it for the function’s full retention period.

How many hours does the admin role really need?

It scales with team size, from a few hours a week layered onto an existing role for a small team to one or more dedicated roles for a large or global function. The admin table above gives a starting allocation; the figure to avoid is zero. Watch the trend, not just the starting number: admin time tends to rise for a year or two after go-live, as integrations, AI feature governance and a first methodology refresh add work, before it settles into a steadier maintenance load.

What should we do if the pilot goes badly?

Fix the configuration, not the auditor’s workaround, and keep the old process as a fallback only through the pilot phase, not indefinitely. A pilot exists to find configuration problems while only two auditors are affected; extending it past week 14 usually means a methodology decision was never actually made in weeks 1 and 2.

Do we need a data migration tool, or can this be done by hand?

It depends on volume. A small function often maps and re-keys its open issues and current plan by hand, using the vendor’s import templates, in a few days. A function with several data sources, an integration to a GRC platform, or a large historical universe typically uses the vendor’s professional services or a partner instead, at the cost premium over in-house work that the pricing evidence above illustrates.

Should AI features be turned on during implementation?

Not by default. Most vendors ship AI features as optional, and the safer sequence is to go live on the core workflow first and evaluate AI features afterward, with a person checking outputs before they touch a live engagement. Gartner’s Market Guide for Audit Management Software, published 13 April 2026, warns buyers to be wary of vendors overstating what their AI actually does, sometimes called agent-washing, which is one more reason to verify before trusting an output; the site’s guide to evaluating AI in audit software sets out what to check.

internalauditguide.com has no commercial relationship with any vendor named on this page. We take no vendor money, run no affiliate links and accept no sponsored placements, and no vendor saw this page before publication. Product and company names are the trademarks of their owners. Corrections: desk@internalauditguide.com.

Sources and verification

Free PDF · 15 one-page cheat sheets

The Internal Audit Desk Reference, free by email.

GIAS 2024 on one page, the sample-size tables, the risk and control matrix, the five-C finding, the SOX deficiency ladder, the CIA exam at a glance, and nine more. Printable. Written by working auditors.

Free. Confirm your address once; the PDF link follows. Unsubscribe anytime.

Cover of The Internal Audit Desk Reference: fifteen one-page cheat sheets and templates (PDF)
Cover of The Internal Audit Desk Reference: fifteen one-page cheat sheets and templates (PDF)

Free PDF

The desk reference: 15 cheat sheets, one page each.

GIAS map, sample sizes, the RCM, the five-C finding, SOX deficiency ladder, CIA at a glance. Free, printable.

Confirm once; the PDF link follows. Unsubscribe anytime.

Comments

Leave a Reply