Skip to content

    SAMPLE: a fictional Ontario law firm. The firm and everything about it below, its people and every number included, are fictional. Not a client of Adwaizer.

    AI Audit report: Quillmere Osgrove LLP (sample)

    Prepared by: Adwaizer AI Consulting Inc. · Engagement: Adwaizer AI Audit (2 weeks, fixed price CAD 2,100 plus GST/HST for a Canadian client) · Report version: sample, October 2026

    What this page is. This is the report a client keeps at the end of the Adwaizer AI Audit, shown for a mid-market law firm. Quillmere Osgrove LLP is invented to show the format and depth; the firm does not exist, and neither do its people, clients or numbers. In a real audit, every figure comes from the client's own records or interviews, and every estimate says what it rests on.

    How to read the numbers. Nothing in this report was measured. Figures marked (owner's estimate) are what the fictional managing partner or firm staff told us in interviews. Figures marked (estimate) are ours and name their assumption next to them. Costs are in Canadian dollars (CAD), before tax.

    Regulatory terms. Terms such as conflict of interest, client identification and verification, and confidential information are used as the Law Society of Ontario's Rules of Professional Conduct and By-Laws, and the Personal Information Protection and Electronic Documents Act (PIPEDA), use them; each is cited in section 11. This report is not legal advice or a compliance review, and it does not replace the firm's own reading of the rules or the Law Society's guidance.


    1. Executive summary

    Quillmere Osgrove LLP is a full-service law firm in Ontario with about 120 people across a main office and one branch. It practises business law, real estate, wills and estates, employment law and civil litigation. About 160 new matters open each month (owner's estimate). Every one of them passes through the same chain: an intake call or email, a conflict search by the conflicts clerk, client identification and, where funds move, verification, then file opening and an engagement letter. Each step re-types the same names into a different place. Lawyers also lose time finding the firm's own best precedent, which lives in shared folders and in lawyers' own drives. The managing partner's concern is that associates have started using AI tools on their own, with no firm-wide rule, while the intake chain is already the firm's slowest step.

    We mapped 10 workflows and scored 6 AI opportunities. Our recommendation starts with firm-wide controls, then one tool at a time:

    1. First, a firm-wide AI use policy and approved-tool list (roadmap week 1), built on the Law Society's own generative AI checklists, so that every later step runs inside it.
    2. Then a file-opening gap check (opportunity O3) that compares each new matter record with its intake notes, identification record and engagement letter, and lists what is missing for the file-opening clerk. It changes no record and gives records staff back an estimated 3 hours a week.
    3. Next, conflict search preparation (O1): AI pulls every party, related company and name variant out of the intake email or form into a search request; the conflicts clerk runs the search in the firm's own system and a lawyer decides. Estimated 7 hours a week back across intake staff.

    We advise against any AI step in deciding a conflict, verifying a client's identity, or filing research a lawyer has not checked line by line (section 5). The 90-day plan in section 6 delivers the policy in week 1, the gap check in weeks 2 to 5 and conflict search preparation in weeks 6 to 11. Estimated cost of the three opportunities we would build: CAD 14,000 to 25,000 one-time plus CAD 4,000 to 10,000 a year to run (section 7).


    2. The firm (fictional)

    ItemDetail
    NameQuillmere Osgrove LLP (invented)
    LocationA main office in a mid-sized Ontario city and one smaller branch
    SectorLaw firm licensed by the Law Society of Ontario: business law, real estate, wills and estates, employment, civil litigation
    PeopleAbout 120: 16 partners, 26 associates, 2 articling students, 14 law clerks, 6 licensed paralegals, 1 conflicts clerk, 3 file-opening and records staff, 4 accounting staff, 5 IT and administration staff, about 43 legal assistants and reception staff
    MattersAbout 160 new matters a month (owner's estimate), about 4,800 active files (owner's estimate)
    RevenueNot disclosed for this sample
    Current toolsA cloud practice management system (matters, time, billing, conflict database); a document management system; Microsoft 365 (Outlook, Word, Teams); two paid legal research databases; a web intake form; precedents in shared folders
    Who we spoke withManaging partner, director of operations, conflicts clerk, records supervisor, two law clerks, the IT manager, and one partner and two associates from each of business law and litigation (eleven 30 to 45 minute interviews)

    3. Current-state findings

    F1. Intake details are typed four times. A prospective client's name and the other parties are taken down by reception or a legal assistant, typed into the intake form, typed again into a conflict search request, and again at file opening. Intake emails often name companies by trade name only, and spelling varies (owner's estimate).

    F2. Conflict search requests are built by hand. The conflicts clerk receives about 40 search requests a week (owner's estimate). Each one needs every party, former name, parent or subsidiary and key individual listed before the search runs; legal assistants spend about 20 minutes preparing each request (owner's estimate), and the clerk sends about one in four back for missing parties (owner's estimate). The Rules require a lawyer not to act where a conflict of interest exists unless the Rules permit it (section 11, [1]), so this step cannot be skipped or rushed.

    F3. File opening waits on missing pieces. A matter cannot open until the identification record, the conflict clearance and the signed engagement letter are on file. Records staff chase missing items by email; about one new file in five waits more than two days (owner's estimate). Client identification is required when the firm is retained, and verification when it handles funds (section 11, [2]).

    F4. Precedents are hard to find. The firm's own approved precedents (share purchase agreements, employment contracts, wills, demand letters) sit in shared folders with inconsistent names. Associates estimate 1 to 2 hours a week each searching for, or re-drafting, a precedent the firm already has (owner's estimate).

    F5. Matter handovers rely on memory. When an associate leaves a file or goes on leave, the receiving lawyer reads the file to find its status; partners estimate about 2 hours per handover (owner's estimate).

    F6. Research memos repeat. Similar research questions are answered more than once because past memos are not searchable by topic (owner's estimate).

    F7. No AI policy exists. Several associates already paste draft emails and clauses, some with client names, into free consumer chatbots. That puts confidential client information into tools the firm has not reviewed, which puts its duty of confidentiality at risk under the Rules and their technology-competence commentary (section 11, [1]).

    Readiness (our scoring, 1 = weak, 5 = strong)

    DimensionScoreWhy
    Data3Matters and conflicts are in one practice management system; precedents and memos are scattered
    Tools4Practice management, document management and Microsoft 365 already paid for and used daily
    People3Associates want help; several partners are wary after news of AI-invented case citations
    Process3Intake and file opening follow fixed steps set by the Rules and By-Laws; the hand-offs between them are manual
    Governance1No AI policy and no approved-tool list; unreviewed consumer tools in use with client text

    These scores are our judgement from the interviews, not a measurement.


    4. Workflow map and scored AI opportunities

    4.1 Workflows mapped (10)

    #WorkflowHours a week, staff (owner's estimate)Error riskData sensitivityFit for AI
    W1Intake calls and emails, web form20MediumHigh (client information)Medium
    W2Conflict search request preparation13High (missed party)HighHigh as a drafter, with clerk and lawyer review
    W3Conflict search and clearance decision8High (regulated duty)HighNone for the decision
    W4Client identification and verification6High (regulated duty)HighLow (a checklist does this)
    W5File opening and engagement letter10MediumHighHigh as a checker only
    W6Precedent search30+ (associates)MediumMediumHigh, over the firm's own approved precedents
    W7Legal research and memos40+ (associates)High (wrong authority)MediumLow until every citation is checked by a lawyer
    W8Matter handover summaries4MediumHighMedium
    W9Docketing time and billing25MediumMediumMedium
    W10Trust accounting and trust reconciliations12High (regulated records)HighNone

    4.2 Scored opportunities

    Scoring: Value (time saved or errors avoided), Effort (how hard to build and run; 5 = easiest), Risk (what could go wrong for clients or for the firm's licensees; 5 = safest). Each is 1 to 5; total out of 15. Scores are our judgement, with the reasoning in the last column.

    #OpportunityValueEffortRiskTotalReasoning
    O3File-opening gap check. For each new matter, compare the matter record with the intake notes, identification record, conflict clearance and engagement letter; list what is missing for records staff.44513Reads only; a person fixes every record; catches gaps the day the file is requested
    O1Conflict search preparation. From the intake email or form, draft the search request: every party, related company, former name and spelling variant. The conflicts clerk checks it and runs the search; a lawyer decides.53412Largest time gain at intake; the clerk and the lawyer stay in charge of the search and the decision
    O2Precedent finder. Search the firm's own approved precedents in plain language and return the document and clause, with its location.43411Saves associate time; answers only from documents the firm approved, each linked to its source
    O4Matter handover summary. Draft a status summary of a file from its documents and notes, checked by the handing-over lawyer.34310Answers F5; client text must stay in reviewed tools
    O5AI-drafted research memos.3328Every authority would need a lawyer's check; see section 5
    O6AI deciding conflicts or verifying identity.2215Regulated professional judgement; the decision belongs to the lawyer under the Rules and By-Law 7.1

    Top 3 for Quillmere Osgrove: O3, O1 and O2, in that order, after the week-1 AI policy. O4 rides along with O2 once the firm's approved-tool list is in place.

    4.3 What could go wrong, per top opportunity

    #Main riskHow we would contain it
    O3The check misses a gap, and staff stop doing their own file-opening reviewIt runs beside the records supervisor's checklist for the first 6 weeks; both lists are compared before the manual check is reduced
    O1A draft request leaves out a party or a related company, and a conflict is missedThe clerk checks every draft against the intake source before running it; the draft only adds names, never removes one the assistant typed; tested against 50 past requests before go-live
    O2The finder returns an outdated precedent or one from outside the approved setIt searches only the folder of approved precedents; each answer shows the document's date and owner; the lawyer reads the source before use

    5. Do not do yet

    IdeaWhy not nowRevisit when
    Any AI step that decides whether a conflict exists, or whether consent cures itThe Rules make this the lawyer's decision, including consent after disclosure (section 11, [1])Not recommended
    AI verifying a client's identity, or deciding a verification is completeBy-Law 7.1 sets the verification methods and the records kept (section 11, [2])Not recommended
    AI-drafted research memos or court documents relied on without a lawyer checking every case and statute citedA wrong or invented authority is the lawyer's responsibility; the Law Society's generative AI guidance addresses it (section 11, [3])The firm has a written check step and has run it on 20 past memos
    AI writing or changing trust ledgers, trust reconciliations or other financial recordsBy-Law 9 sets what the firm must record and keep (section 11, [4])Not recommended
    Any client text in tools not on the approved-tool listConfidentiality under the Rules; under PIPEDA the firm stays responsible for personal information it sends to a vendor for processing (section 11, [1], [5])Each tool passes the week-1 review

    6. 90-day roadmap

    WeeksWorkOwner (Quillmere Osgrove)Measure of success
    1Write a firm-wide AI use policy with an approved-tool list, using the Law Society's generative AI checklists; move associates off free consumer chatbotsManaging partner, IT managerPolicy adopted by the partners and acknowledged by every licensee and staff member
    1-2Review each approved tool's data terms (where data is stored, whether it trains on firm text) against PIPEDA's accountability principleIT manager, privacy officerEach tool on the list has a one-page review on file
    2-3O3 design: export a month of file openings with their intake, identification and engagement records; agree what counts as a gapRecords supervisorGap rules signed off
    4-5O3 live beside the manual file-opening checklistRecords supervisorGaps found on the day of the request, tracked against the checklist
    6-7O1 design: agree the search request format with the conflicts clerk; test on 50 past requestsConflicts clerk, a partnerFormat signed off; no party missed in the 50 tests
    8-11O1 pilot in business law and real estate; the clerk checks every draftConflicts clerk80% of drafts need only small edits (target, not a forecast)
    12O2 design: choose the approved precedent set and its owners (build starts after day 90)Two practice group leadersSet listed and owners named

    Measures are set by the client in week 1. We do not promise a saving; we agree how it will be measured before work starts.


    7. Costs (CAD, estimates)

    All figures are estimates for this fictional firm. A real proposal replaces them with vendor quotes and a fixed Adwaizer price.

    ItemOne-timeYearly runningAssumption
    AI Audit (this report)CAD 2,100 plus GST/HST, fixednoneAdwaizer's published audit price for a Canadian client (PROPOSAL-RULES §6)
    O3 file-opening gap check3,000 to 5,000400 to 1,0003 to 5 days of build at a small-firm consulting rate; running cost assumes model usage on about 1,900 new matters a year
    O1 conflict search preparation5,000 to 9,0001,200 to 3,0005 to 8 days including testing on past requests; running cost assumes model usage plus a business AI subscription for intake staff
    O2 precedent finder (build after day 90)6,000 to 11,0002,400 to 6,0006 to 10 days including tagging the approved set; running cost assumes search hosting and model usage for about 60 lawyers, law clerks and paralegals
    Total, three opportunities14,000 to 25,0004,000 to 10,000Sum of the rows above; the audit fee is separate

    Value side (estimate). If O1 cuts the 20 minutes a legal assistant spends preparing each conflict search request to about 8 minutes of checking, on about 40 requests a week, that is about 8 staff hours a week; with the clerk's returns reduced, we use about 7 (estimate: 40 requests and 20 minutes from the owner's estimate; 8 minutes of checking is our assumption, to be confirmed in the O1 pilot). If O3 replaces most of the chasing in W5 with a same-day gap list, that is about 3 staff hours a week (estimate: from the 10 hours in W5). These are the numbers the week-1 baseline will confirm or correct.


    8. Tool choices

    Quillmere Osgrove already pays for a practice management system, a document management system and Microsoft 365, and is not asking to switch. Every opportunity above is designed to work with those tools first.

    NeedOption we would put in front of Quillmere OsgroveWhy it fits
    O3 gap list and the intake queue, if the team wants one board for file-opening workmonday.com (work management), as an option next to the practice management systemOne shared board for missing items, each with an owner and a due date, without changing where matters are kept
    The web intake formJotform (Forms), as an optionStructured intake fields (parties, related companies, former names) that feed O1 directly
    O1 drafting and O2 searchA Claude or OpenAI build by Adwaizer, in tools on the firm's approved list, with a clerk or lawyer check at every stepCustom format and checks need a small build; client text stays in reviewed tools

    We're a monday.com partner, so we may earn a referral fee; the recommendation stands on fit. New to monday.com: https://try.monday.com/q1hi6cfy4tr6. We are experienced with Jotform. Quillmere Osgrove can stay on its current systems for all of the above; these tools are options, not conditions.


    9. What Quillmere Osgrove would buy next

    Next stepWhat it isIndicative scope
    Policy packageFirm-wide AI use policy, approved-tool list and one review per tool, built on the Law Society's checklistsAbout 1 week (estimate)
    Build package 1: O3File-opening gap check, tested and handed over with a one-page runbookFixed scope and fixed price, about 1 week (estimate)
    Build package 2: O1Conflict search preparation with clerk reviewFixed scope and fixed price, about 2 weeks (estimate)
    Build package 3: O2 + O4Precedent finder over the approved set, then handover summariesFixed scope and fixed price, 2 to 3 weeks (estimate), after packages 1 and 2
    Staff trainingTwo 90-minute sessions on safe AI use and checking AI output, built on the week-1 policyHalf a day (estimate)

    Each package is quoted separately after this audit. Quillmere Osgrove can also take this report and build any of it with its own team or another provider.


    10. How the work would be delivered

    • Certified delivery: PMP and PMI-ACP certified; fixed scope and fixed price.
    • Weekly progress: small deliveries every week, each shown in a live demo.
    • No surprises: a risk and issue log shared with you, and written acceptance criteria you sign off.

    11. Sources for the regulatory terms

    Found by web search on 31 Oct 2026; the Law Society's and Justice Laws' own pages are the authoritative text.

    1. Law Society of Ontario, Rules of Professional Conduct, Chapter 3: rule 3.1-2 and its commentary on technology (a lawyer's competence includes understanding relevant technology and its risks to confidential information), rule 3.3-1 (confidential information held in strict confidence), rule 3.4-1 (no acting where there is a conflict of interest, except as the Rules permit, with "conflict of interest" defined in rule 1.1-1) and rule 3.4-2 (consent after disclosure). https://lso.ca/about-lso/legislation-and-rules/rules-of-professional-conduct/chapter-3
    2. Law Society of Ontario, By-Law 7.1: client identification when a licensee is retained, verification of identity when the licensee handles or gives instructions on funds, and the records kept of the documents used. https://lso.ca/about-lso/legislation-and-rules/by-laws/by-law-7-1
    3. Law Society of Ontario, Using technology: lists the practice note "Generative AI: Your professional obligations", the Futures Committee's 2024 white paper "Licensee use of generative artificial intelligence", a quick-start checklist and a checklist for a generative AI policy. https://lso.ca/lawyers/technology-resource-centre/practice-resources-and-supports/using-technology
    4. Law Society of Ontario, By-Law 9: the financial and trust records a licensee must keep. https://lso.ca/about-lso/legislation-and-rules/by-laws/by-law-9
    5. Personal Information Protection and Electronic Documents Act (S.C. 2000, c. 5), Justice Laws: Schedule 1, principle 4.1.3 (an organization is responsible for personal information it transfers to a third party for processing) and principle 4.7 (safeguards). https://laws-lois.justice.gc.ca/eng/acts/p-8.6/FullText.html
    6. Office of the Privacy Commissioner of Canada, Interpretation Bulletin: Accountability (how the accountability principle applies to transfers for processing). https://www.priv.gc.ca/en/privacy-topics/privacy-laws-in-canada/the-personal-information-protection-and-electronic-documents-act-pipeda/pipeda-compliance-help/pipeda-interpretation-bulletins/interpretations_02_acc/

    Sample report. Quillmere Osgrove LLP is fictional; any likeness to a real firm or business is a coincidence. Adwaizer AI Consulting Inc. promises no savings or regulatory outcome; every figure above is an estimate or an owner's estimate as labelled.

    The general AI Audit sampleAdwaizer services