Questions people ask, all in one place.

Who writes it?

One person — Hammad Arain, the founder, with input from a panel of practicing solos in the early review pool. Not a team of associates billing time. Not generated by an AI model and shipped without review.

Will you contact me after?

Only if you reply. Reply with a follow-up question and you get a written answer in the same thread. Don't reply, and you don't hear from us — no nurture sequence, no “checking in” emails, no calls.

What if my answers raise more questions?

We email one batch of clarifying questions before writing. Answer at your pace; the 48-hour clock resumes when you reply. We don't interrogate; we ask the things that actually change the recommendation.

What if I want changes to the report?

Reply to the delivery email. Factual corrections (you don't actually use Clio, you switched in March) get a revised PDF within 24 hours. Disagreements with a recommendation stay in the report — we'd rather you have the full reasoning to push back on than a sanitized version.

Why is it free?

Prevaldi is new and we need case studies. The scope and quality match the eventual paid version — only the price changes. We'd rather earn a reputation by writing thirty good reviews than by selling thirty mediocre ones.

Who writes it?

One person — Hammad Arain, the founder, with input from contractors active in the early review pool. Not a team. Not generated by an AI model and shipped without review.

Will you contact me after?

Only if you reply. Reply, and you get a written answer in the same thread. Don't reply, and you don't hear from us — no nurture, no calls, no “checking in.”

What if my answers raise more questions?

We email one batch of clarifying questions before writing. Answer at your pace; the 48-hour clock resumes when you reply.

What if I want changes to the report?

Reply to the delivery email. Factual corrections get a revised PDF within 24 hours. Disagreements with a recommendation stay in the report — we'd rather you have the full reasoning to push back on than a sanitized version.

Why is it free?

Prevaldi is new and we need case studies. The scope and quality match the eventual paid version — only the price changes.

What does this tool actually do?

Takes shorthand time entries — the kind you scribble at end of day — and rewrites them as defensible billing narratives in the voice and detail level your clients expect. Same hours, same dates, cleaner language.

Will it invent work I didn't do?

No. The tool only rewords the entries you paste. If your shorthand says “30m call w/ JT re discovery,” the polished version describes that call — it doesn't add a follow-up email you didn't write. If an entry is too vague to polish honestly, the tool flags it instead of fabricating.

Do my entries get stored?

No. The text you paste is sent to Anthropic's API to generate the response and discarded. We log a hashed IP for rate-limit accounting and the tool slug, nothing else. Even so, redact client names and matter captions before pasting — use the Redactor tool if you want one-click placeholders.

What if I need to reconstruct entries I forgot to log?

Different tool. Billing Narrative Polisher rewrites entries you already have. Time Reconstructor rebuilds entries from your calendar, sent emails, and notes when you forgot to log them in the first place. The two tools pair — reconstruct first, polish second.

Can I trust the output for a difficult bill?

Read it before sending. The tool is a fast first pass, not a substitute for review. On a contested bill — one you expect a client to push back on — read each entry against your underlying notes to make sure the narrative still describes what actually happened.

What does the Billing Recovery Sprint do?

Takes 90 days of your sent-mail and calendar fragments and rebuilds them into a clean billing log of work you actually did but never invoiced. Each reconstructed entry carries a confidence score so you know which to double-check before sending.

How much time does a typical sprint recover?

Forty to eighty hours is the common range. At a $250 hourly rate that's $10,000 to $20,000 of revenue you already earned and never billed. The sprint pays for itself in the first ten hours you invoice back.

When is this available?

Not yet. It's on the waitlist — add yourself and you'll get one email the moment it ships. No nurture sequence between now and then.

What inputs will you need from me?

An export of your sent mail and calendar for the 90-day window, your existing time log if you keep one, and your matter list. We do not need privileged content, attachments, or client communications themselves — metadata and subject lines plus your existing log are enough.

What does it cost?

Planned at $2,499 one-time. If a typical sprint recovers $10K–$20K of billable time, the math is straightforward — but if your last six months of billing already feel tight, the recovery will be smaller and the audit may not be worth it.

Why does this audit exist?

Thomson Reuters is shutting Casetext down and pushing users toward Westlaw at $750–$1,000 a month. For Am Law that's a rounding error. For a solo who built their research stack around Casetext because it was affordable and good enough, it's unworkable. The audit picks the replacement that fits your jurisdictions and patterns, not the most expensive option on the menu.

What's actually in the report?

Your current research workflow, mapped. The two or three alternatives that fit your actual practice (Paxton, Fastcase, Bloomberg solo, free PACER + targeted Westlaw day passes, etc.) with monthly cost and what each loses vs. Casetext. A migration sequence — what to set up first, what to test, what to cut over last.

Do I need to send you my research history?

No. The intake form asks about your jurisdictions, practice areas, and the kinds of searches you ran most often. We don't need access to your Casetext account or your search log.

How fast is delivery?

PDF in your inbox within 48 hours of the intake form. One round of clarifying questions if your intake leaves something ambiguous — the 48-hour clock pauses until you reply.

Will the recommendation work for my state bar?

The audit names which alternatives ship with the state-specific coverage you need. Fastcase comes free with many state bar memberships — if yours is one, that's the first option we'll evaluate. Where coverage is patchy, we'll tell you and propose a fallback.

What does this tool do?

Takes a wall of unread client email and sorts it. Each email gets a priority, a suggested response template for the routine ones, and a flag if it looks like something you can't ignore — a missed deadline mention, a fee dispute, a client threatening to switch firms.

Will it draft replies for me?

It drafts templates for the routine categories — “acknowledge receipt,” “share status,” “provide document.” These are starting points, not finished replies. Anything substantive — strategy, advice, fee discussion — gets flagged for you to handle directly, not templated.

Should I paste actual client emails into this?

Run the Redactor tool first. Replace client names and identifying details with placeholders before pasting here. The triage logic works on the content of the message, not the identity of the sender — placeholders don't degrade the output.

Does it understand the difference between an actual emergency and a panicky client?

Reasonably well. It flags messages that name a deadline, a court date, a settlement number, or a service-of-process event as high priority regardless of the client's tone. It's calibrated to under-flag rather than over-flag — there's no point in a triage tool that calls everything urgent.

What if I have 200 emails in the backlog?

Paste them in chunks. Each run is rate-limited and capped at 50K characters of input. Two or three runs covers most realistic backlogs. The output is structured so you can paste the priority lists into a doc and work through them.

Does this replace my firm's conflict-check procedure?

No. This is a text-comparison worksheet — it surfaces overlaps between a new matter description and your existing client list. It does not replace running names through your case management system, calling parties, or doing the substantive conflicts analysis that's your professional responsibility.

What does it actually catch?

Same-name matches, similar-name matches (typos, transliterations, married names), and substantive overlaps in matter description that suggest a potential conflict — adverse party, related entity, prior representation in a similar matter.

Should I paste real client names into this?

Yes — the tool is built for it, and the comparison only works on real text. The input is sent to Anthropic's API, not stored on Prevaldi servers. But it goes off your machine. If your firm's policy forbids client names leaving your network for any reason, run the Redactor first and substitute placeholders.

Will it flag things that aren't actually conflicts?

Sometimes — common surnames especially. The tool errs toward flagging more rather than fewer. Treat every flag as “look at this,” not “this is a conflict.”

What format should my client list be in?

Plain text, one client per line, with a short matter description if you have it. No need for a structured export — copy-paste from your case management system works fine.

What does the Contract Clause Identifier do?

Reads a contract clause-by-clause, flags risky language, and lists standard protections that are missing. Output is a sortable list of clauses with risk levels, a short summary, and a “favors” line that says which party the contract leans toward.

Will it redline or rewrite the contract for me?

No. This is an identifier, not an editor. It tells you which clauses to look at and why; the redline is your job. The plan-to-use case is “read this contract in 4 minutes before my call,” not “rewrite this contract while I'm on a call.”

Can I paste a redacted contract?

Yes — that's the recommended approach. The analyzer reads clause language, not party identity. Replace party names, addresses, and governing state with bracketed placeholders and you still get the full analysis. Three example contracts on the page show the pattern.

How long can the contract be?

Fifty thousand characters of input — about 80 pages of typical contract language. For longer contracts, paste the meaningful sections (the operative clauses) and skip exhibits and signature pages.

What about contract types you don't list?

Select “Other” or leave the type field on auto-detect. The analyzer recognises clause patterns across most US commercial contract types; the type selector helps it know which standard protections to expect, but it isn't a hard filter.

What does this tool flag?

Imminent deadlines (anything within 30 days), stale matters (no activity for X months — you set the threshold), data-quality issues (missing fields, ambiguous dates, name typos), and matters with deadline patterns that suggest a missed prior date. The output is a sorted risk report you can paste into your task list.

Will it actually catch a missed statute of limitations?

It will flag matters where the dates suggest one might be approaching, and it will flag matters where the date data is too ambiguous to evaluate. It will not tell you the SOL for your jurisdiction or your matter type — that's research the tool can't do from the inputs. The output is a checklist of matters to look at, not a legal opinion.

What format should my matter list be in?

Plain text or CSV-style. One matter per line with key dates beside it works. The scanner is forgiving — it'll handle different date formats (3/12/26, March 12 2026, 2026-03-12) and inconsistent column ordering. Don't include privileged content; matter-name, key dates, and a one-line description is all the scanner uses.

Should I paste this into the tool every week?

Weekly is overkill; monthly is closer to right. If you want a standing version of this, the Practice Health Quarterly (currently on the waitlist) bundles deadline scanning with billing health, IOLTA review, and intake quality into one quarterly PDF.

What about court-specific calendaring rules?

Out of scope. The scanner doesn't compute deadlines from triggers (e.g. “add 30 days from service of complaint”). It works with the dates you've already calculated and entered. For trigger-based calendaring, use your practice management system's calendaring module — Clio, MyCase, PracticePanther all have one.

What does this tool give me?

A structural outline of a demand letter — facts, damages, liability theory, demand, deadline, signature block — with the right sections in the right order and the kinds of language an attorney expects to see in each. It is a starting skeleton, not a finished letter.

Will it write the actual demand language for me?

It drafts placeholder language for each section so you can see what belongs where, but you'll rewrite most of it. The point is to remove the “staring at a blank page” problem — not to produce a letter you sign and send.

What inputs do I need?

The incident facts (what happened, when, where), the damages (medical bills, lost wages, property damage — concrete numbers if you have them), the liability theory (negligence, breach of contract, statutory), and the recipient (insurer, opposing counsel, defendant directly).

Should I paste actual case facts in?

Use the Redactor first. Replace client names and identifying details with placeholders before pasting. The skeleton-building logic runs on the structural facts (the “what happened”), not on identities — placeholders preserve the structure.

Will this work for my practice area?

It's calibrated for personal injury, breach of contract, and employment demands. For specialised practice areas (construction defect, professional malpractice, securities, etc.) the skeleton will be roughly right but you'll need to adapt the standard sections.

What does the Discovery Drafter produce?

A starter outline of discovery topics organised by category — interrogatories, requests for production, requests for admission, deposition topics — based on the case description and the discovery types you select. The output is a topic list, not actual interrogatories ready to serve.

Will the output be ready to serve?

No. It's a thinking aid — “here are the topics worth covering” — not a draft you sign. You still write the actual interrogatories, draft the document requests, and tailor each item to your jurisdiction's rules and your case strategy.

How much case detail should I include?

Enough to make the topics specific — the parties (use placeholders), the cause of action, the disputed factual issues, the timeline. Don't paste privileged work product, your client's deposition transcript, or anything covered by attorney-client privilege.

Does it know my jurisdiction's discovery limits?

It is calibrated to federal practice and most state practice but does not enforce numeric limits (e.g. “25 interrogatories per side” under FRCP 33). You'll need to cull the topic list to fit your jurisdiction's caps. The output is comprehensive; pruning it is your job.

Can I use this for ediscovery topics?

Partially. It will flag the categories of electronic documents worth requesting (email, calendar, Slack, financial records, etc.) but not the technical specifics of a 30(b)(6) deposition topic on data custody or a request for a forensic image. For complex ESI, use this as an outline starter only.

What does the Estate Brief Generator give me?

A structured drafting brief built from your raw client meeting notes — organised by trust section (Revocable Living Trust, Will, Power of Attorney, Healthcare Directive, etc.) with the client's stated wishes mapped to each section. The point is to skip the “re-reading my notes” step and start drafting.

Will it produce the actual estate documents?

No. It produces a brief that a drafting attorney or paralegal uses to start document preparation. Document assembly happens in your drafting platform (WealthCounsel, ElderCounsel, your own templates). The brief tells the drafter what the client wants, section by section.

Should I paste real client names from my notes?

Use the Redactor first. Replace names and identifying details with placeholders before pasting. The brief-building logic reads the substance of the notes (assets, beneficiaries, fiduciaries, distribution patterns), not the identities — placeholders preserve everything that matters.

What if my notes are scattered across multiple meetings?

Paste them all together. The brief deduplicates and reconciles overlapping notes — if the client said one thing in the first meeting and changed their mind in the second, the brief flags the conflict for you to confirm rather than guessing which version is current.

Does it handle complex estate plans (irrevocable trusts, GST, special needs)?

It will surface the relevant client wishes from your notes but it is calibrated to common estate plans. For complex planning (GST allocation, special needs trusts, dynasty trusts, ILITs), the brief gets you to a starting point — the substantive structuring is still attorney work.

What does this audit actually deliver?

A 10–14 page written assessment of your intake, billing, and matter infrastructure with the specific changes you need to make before your first hire arrives. Not generic advice — concrete next steps on your engagement letter template, your matter-opening process, your trust handling, and your billing cadence. PDF in 48 hours.

Why audit before hiring instead of after?

Because workflows that worked when one person held everything in their head don't survive contact with a second person. If your new hire arrives on Monday and you haven't sorted the intake handoff first, by Wednesday they're building their own version of every template and within three months you have two versions of everything. Cleaning house before they start is dramatically faster than retrofitting.

What inputs do you need?

About 15 minutes of intake form — your matter types, your current intake flow, your billing cadence, your software stack, your trust handling, the role you're hiring for. We don't ask for client names, matter captions, financial statements, or anything privileged. The intake is structural, not factual.

Will this work if I haven't picked who I'm hiring yet?

Yes — and arguably it's better that way. The audit names the workflow upgrades you need; the hiring decision (paralegal, associate, virtual assistant, contract billing clerk) is informed by what the audit surfaces about your bottlenecks. Several reviewees have rethought the role after reading the audit.

What does it cost?

$999 one-time. For context, the first three months of a bad hire — searching, training, parting ways, and starting over — typically run a solo $15,000+ in time and salary. The audit is priced as a fraction of getting that decision wrong.

What does the Intake Summarizer do?

Reads raw consult notes and produces a structured matter summary — parties, dates, claim type, damages, jurisdictional questions, and a flagged list of follow-up questions you still need to ask before opening the matter.

What's the “follow-up questions” section?

Things you didn't get answered in the first consult that you need before you can responsibly open or decline the matter — has the client preserved evidence, has anyone served notice, what's the relationship between the parties, is there an existing lawyer. The list is calibrated to surface the questions that change the matter decision, not every theoretical question.

Should I paste real consult notes?

Use the Redactor first. Replace party names with placeholders before pasting. The summary logic reads the substance of the consult (facts, dates, claim theories), not the identities.

What if my notes are messy or out of order?

That's the use case. The tool is built to take stream-of-consciousness consult notes — including the “oh wait, going back to what they said earlier” asides — and produce a clean structured summary. Order doesn't matter; completeness does.

Can I use the output as a memo to file?

It's a starting point. The summary covers the structural facts and flags the gaps; the legal analysis and the recommendation belong to you. For a true memo to file, treat the summary as the “facts” section and add your analysis underneath.

What is the IOLTA Quarterly?

A standing written review of three months of trust account activity, delivered quarterly. Each report flags patterns a qualified bookkeeper or trust accounting specialist should look at — not a substitute for one, a second set of eyes on the numbers.

How is this different from the IOLTA Safety Check?

The Safety Check is a one-time three-month review. The IOLTA Quarterly is the same review on a standing cadence — every 90 days, automatically. Drift becomes visible across quarters in a way it isn't in a single snapshot.

When is it available?

Waitlist. Add yourself and you'll get one email when it ships — no nurture sequence between now and then.

What do I need to send each quarter?

An export of three months of trust account transactions and your active matter list. We do not need bank statements as PDFs, client SSNs, or anything beyond the transaction-level data your accounting software exports.

Is this legal or ethics advice?

No. It's an anomaly report with confidence flags and recommended next actions. Anything substantive (an actual trust violation, a remediation plan, a bar self-report decision) is between you and your ethics counsel.

What does the IOLTA Safety Check actually do?

Reviews three months of trust account activity and produces a written anomaly report — flagged transactions, unusual patterns, reconciliation gaps, and a confidence score on each finding. Delivered as a PDF in 48 hours.

Is this a substitute for a bookkeeper?

No. It's a second set of eyes on the numbers and a fast triage. A bookkeeper engagement to remediate a known IOLTA error usually runs $3,000–$8,000. This audit is meant to surface drift while it's still cheap to fix, before you need that engagement.

When is it available?

Waitlist. Add yourself and you'll get one email the moment it ships — not a newsletter.

What inputs do I need to send?

Three months of trust account transactions (your accounting software's export is fine), your active matter list, and your most recent three-way reconciliation if you have one. We don't need bank statement PDFs or anything else.

What about my state's specific trust rules?

The audit is calibrated against ABA Model Rule 1.15 and the relevant state-specific guidance for the state you practice in. Findings cite the rule and recommend the next action — not legal advice, but a second set of eyes pointed at the rule that applies.

What is the Practice Health Quarterly?

Four reports a year covering the four areas of a solo practice that fail silently: billing leakage, IOLTA drift, deadline exposure, and intake data quality. One PDF per quarter, with flags tracked across reports so trends become visible.

Why these four areas?

Because they are the four largest sources of silent revenue loss for solos — typical leakage runs $30,000 to $80,000 a year per solo when you total unbilled time, missed deadlines, intake leaks, and overlapping software. The subscription costs about 4% of that.

When is it available?

Waitlist. Add yourself for the email when it ships — no newsletter in between.

Can I just buy one quarter to try it?

Planned yes — a first quarter at standard pricing, then the standing subscription if you want to continue. The subscription price gets you the four reports a year plus the cross-report trend tracking, which a single quarter can't deliver.

What's the relationship to the other Prevaldi audits?

The Full Practice Review is a deeper one-time review across all four areas plus software stack and AI opportunities. The Quarterly is a standing watch on the four areas where drift is silent. If you've never had any review done, start with the Full Practice Review (currently free while in beta).

What does this tool produce?

A HIPAA-compliant medical records request letter ready to print on your letterhead. Patient information, provider information, date range, scope, and the authorisation language a records custodian needs to release.

Is the language actually HIPAA-compliant?

The standard authorisation language and the elements required by 45 CFR 164.508 are present. State-specific add-ons (some states have stricter requirements — California, Texas, Florida especially) are flagged but you should confirm your jurisdiction's specifics before sending. The tool is a fast draft, not a substitute for knowing your state's rules.

What inputs do I need?

Patient's full legal name and date of birth, provider's name and address, the date range of records being requested, and the scope (full records, specific dates, specific procedures, billing only, etc.). Use placeholders if your firm's policy keeps PHI off web tools — the letter structure still works.

Will it generate an authorisation form too?

It generates the language a custodian needs. Whether your client signs a separate authorisation form (HIPAA Authorization, your state's specific form) or signs the letter directly is your choice. The output gives you both options.

What about subpoenas for medical records?

Different document. This tool is for client-authorised records requests under HIPAA. For subpoenas duces tecum, you need your jurisdiction's subpoena form and the specific notice provisions your state requires for medical records — those add-ons are not what this tool produces.

What does The Redactor do?

Takes raw notes with client names and identifying details and returns the same text with placeholders substituted in, plus a legend mapping each placeholder to the original. The legend stays in your browser — it never leaves your machine.

Where does the legend live?

Only in your browser session. The Prevaldi server sees the original text only long enough to substitute placeholders; the mapping is generated and stored client-side. When you close the tab, the legend is gone. Save it locally if you need to un-redact later.

What is this actually for?

Preparing notes for use in any other Prevaldi tool — or in any external AI tool — without sending real client identities. The standard workflow is: paste consult notes here, get back placeholdered text, run that text through the Intake Summarizer, the Demand Skeleton, the Comm Triage, etc. Identity-free.

Should I send the redacted output to a client or court?

No — this is an internal preparation tool. The placeholders are obviously placeholders (PARTY_1, MATTER_A, DATE_X). The output is meant to be read by you and by AI tools, not by clients, opposing counsel, or judges.

What does it catch — names only, or other identifiers too?

Names, addresses, phone numbers, email addresses, dates that look like birthdates, SSN-shaped numbers, account numbers, and dollar amounts when they appear to identify a transaction. Misses are possible — read the output before treating it as fully scrubbed.

What does the Status Update Generator produce?

A client-ready status update email built from your rough case notes. Plain language, careful hedging (no “we will win” promises), and no invented facts — if your notes don't say it, the update doesn't say it.

Will it use language that creates fee or outcome promises?

No. The output is calibrated to avoid outcome guarantees, fee modifications, and anything that could be read as an opinion on the merits beyond what's appropriate for a status update. You should still read the output — calibration is not a substitute for your own review of every message you send to a client.

What's the right level of detail for my notes?

Enough so the update is specific. “Filed motion last Tuesday, hearing set for the 28th, opposing counsel signalled they may seek extension” gives you a real status update. “Working on the case” doesn't.

Should I paste real client names from my notes?

Use the Redactor first. Replace names with placeholders, run the tool, then substitute the real names back in when you paste the output into your email client. The status logic doesn't depend on the identities.

Can I use the same draft for multiple clients?

It's tempting and it's a mistake. Status updates that read as obviously templated cost trust. Run the tool per client with each client's actual notes. The tool is fast enough that doing it per client isn't a meaningful time cost.

What does Time Reconstructor do?

Rebuilds billable time entries from the fragments of your day — calendar events, sent emails, scattered notes — when you forgot to log time as you went. Each rebuilt entry has a confidence flag and points back to the source fragments that justify it.

Will it invent time I didn't work?

No. The reconstruction is bounded by the fragments you give it. If your calendar shows a 30-minute block and your sent mail shows you wrote one email during that window, the entry reflects that — it doesn't extend the time beyond what the source shows. Entries that are guesses are flagged as “low confidence — verify before billing.”

What fragments work best?

Calendar export, sent-mail subject lines and timestamps, and your written notes from the day. You don't need email bodies — subject lines plus the calendar context is usually enough. Phone-call logs help when you have them.

What about non-billable time?

The tool separates billable from non-billable based on matter context. Personal calendar events, internal meetings, marketing time — those get categorised as non-billable. Review the categorisation before billing; misclassification is the most common error.

How far back can I reconstruct?

Reasonably reliably for the last 7–14 days. Beyond that, your memory of the context degrades and the source fragments stop carrying enough information. For 90-day reconstructions across an entire practice, the Billing Recovery Sprint (currently on the waitlist) is the right tool — it's the same logic on a much wider input window.

What does the Trust Account Scanner flag?

Patterns across three months of trust account activity — commingling indicators, deposits without a matching matter, withdrawals without a contemporaneous invoice, reconciliation gaps between bank, firm ledger, and client subledgers. Each flag has a confidence level and a recommended next action.

Is this a substitute for monthly reconciliation?

No. Monthly three-way reconciliation (bank statement, firm trust ledger, client subledgers) is what your state bar expects, full stop. The scanner is a separate set of eyes on patterns the reconciliation may not surface — a deposit that matches the bank but is mis-attributed in the client subledger reconciles cleanly and still represents a problem.

What format should my transactions be in?

Plain text or CSV-style. Date, description, amount, matter (if your accounting software exports a matter column). Most case management systems — Clio, MyCase, PracticePanther, LeanLaw — export this directly. Don't paste client SSNs or bank account numbers; the scanner uses transaction-level data, not banking metadata.

Will it tell me my state's specific trust rules?

It is calibrated against ABA Model Rule 1.15 and the major variations across high-rule-density states (CA, NY, TX, FL especially). Findings cite the rule and recommend the next action. Where your state has a less-common variant, the flag still surfaces — you'll need to check your specific rule before acting.

What if I find a real problem?

Talk to your bar's ethics line or an ethics counsel before remediation. Self-reporting requirements vary by state. The scanner names the pattern; the response — whether and how to disclose — is a decision the scanner can't make for you.

What does the Addendum Digest do?

Takes the text of a bid addendum and pulls out every change — scope, cost, schedule, document revisions, deadline shifts. The output is a checklist of what's actually different and a flagged list of items you need to re-quote before the bid deadline.

Will it cross-reference changes against the original plans?

No. The Digest reads the addendum text on its own. It tells you what the addendum changed; whether that change conflicts with a detail on Sheet A-301 you bid against is your job to verify. Use this to triage what's worth a re-read, not to skip the re-read.

What if the addendum is 40 pages?

Paste it. The tool is capped at 50K characters of input, which is roughly 80 pages of normal text. For addenda longer than that, paste the “what changed” section and the spec/drawing change list — those are the parts the Digest extracts from anyway.

Will it catch every change?

It catches stated changes — anything the addendum explicitly calls out. It will not catch implicit changes (a revised drawing that contradicts an unrevised spec section, for example). The Digest is calibrated to surface what the GC wrote; what the GC accidentally left out is on you to spot.

Should I paste this in before or after my own read?

Before. Run the Digest first to get a structured map of what's in the addendum, then do your own read against the map. Subs who read addenda cold often miss the same kinds of changes (scope reassignments, allowance/covered swaps, schedule shifts hidden in the cover letter). The Digest closes that gap.

What does the Audit Bundle include?

The Estimate Review and the Change Order Audit, delivered together as a single combined PDF. Same scope as ordering both individually; one engagement, one fee.

Why these two together instead of separately?

Margin leaks at the front of the job (bids that undercount scope) and at the back of the job (change orders that fail to capture approvals) are connected. Every undocumented change order has a corresponding scope gap that should have caught it during estimating. Looking at both processes in one engagement makes the pattern visible.

What's the typical recovery range?

Combined recoverable typically exceeds $50,000/year on a $5M operation. The bundle is $19,999 one-time — the math is direct on a typical small GC.

What do you need from me to run it?

Three to six recent bids you've submitted, three to six change orders that ran on those jobs, and a written description of your estimating and change-order processes. We don't need contract values, customer names, or anything privileged — the audit reads the patterns, not the parties.

How fast is delivery?

PDF in your inbox within 10 business days of the intake form. One round of clarifying questions if your intake leaves anything ambiguous — the clock pauses until you reply.

What does the Bid Leveller do?

Takes sub bids in any format — pasted email body, copied spreadsheet rows, PDF text — and produces a normalised comparison table. Each bidder's scope, exclusions, allowances, and price laid out so you can see what each one actually included and what they quietly left out.

Does it calculate the total project cost?

No. The Leveller compares scope, not budget. It tells you that Sub A excluded fire-stopping and Sub B included it; it doesn't roll a project total because that depends on the rest of your bid you didn't paste in.

What's the most useful output?

The scope-gap section. The “Sub C is $4K cheaper than Sub B” line is obvious; the “Sub C excluded permit fees and rough-in coordination” line is what saves you from awarding the wrong bid and eating the gap.

What about unit prices?

Out of scope. The Leveller compares scope and totals, not unit-price reasonability. For unit-price validation (“is $4.20/SF for X reasonable for our market”), you need historical bid data the Leveller doesn't have.

What format should I paste the bids in?

Whatever you have. Pasted email body, copied table, exported PDF text — the Leveller is built to handle inconsistent formats because that's what real sub bids arrive in. If a bidder sent a one-page proposal and another sent a 12-tab spreadsheet, paste both. Cleanup is what the tool does.

What does the Change Order Memo tool produce?

A formal written change order memo from a plain-language description of the change. Scope, cost impact, schedule impact, signature block — the structure a GC expects to see before approving and the structure a customer recognises as a real change order rather than a text message.

Will this hold up legally?

The memo gives you the structure courts and contracts treat as a written change order. Whether your specific contract requires additional formalities (signatures from both parties, prime-contract notice provisions, lien-waiver attachments) is contract-specific. The memo is the minimum upgrade from a text-message approval; a contract attorney should review your standard form before you adopt it firm-wide.

What if the change was already verbally approved?

Exactly the use case. The memo turns the verbal into a written record before the GC denies the approval ever happened. Send the memo within 24 hours of the verbal — speed matters more than perfection. The pattern that loses disputes is approval-by-text-with-no-follow-up-paper.

What inputs do I need?

What changed, why it changed (RFI response, field condition, GC direction), the cost impact, the schedule impact, and the parties involved. The tool fills in the boilerplate — the substance is what you bring.

Is this the same as a Pending Change Order (PCO)?

Close. The output is a written CO ready to send for approval. Whether you label it PCO, COR, or CO depends on your GC's terminology and your contract; the structure is the same. Edit the header to match the convention your GC uses.

What does the Change Order Audit look at?

Your change-order documentation practice across three to six recent jobs. Where verbal approvals are turning into “I never agreed to that” weeks later, where the paper trail breaks, what your current process can and can't defend in a dispute, and the minimum upgrade that holds up.

Why this audit specifically?

Change-order chaos is widely cited as 3–4% of margin lost per project on mid-size construction firms. On a $5M operation that's $150,000–$200,000 a year. The audit names the specific pattern your operation uses today and the specific gaps that are costing you that margin.

What do you need from me?

Three to six recent change orders that ran on real jobs (redacted is fine — names and numbers can be placeholdered), a written description of how COs typically flow in your operation, and the contract size range you typically work. We don't need contract values, customer names, or signed copies.

How fast is delivery?

PDF in your inbox within 5 business days of the intake form. One round of clarifying questions if anything is ambiguous — the clock pauses until you reply.

What does it cost and what's the math?

$9,999 one-time. On a $5M operation losing 3% to change-order leakage ($150K/year), the audit pays for itself if it recovers 6.5% of that leak — a low bar on most operations. If you're below $2M in volume, the math gets tighter; talk to us first about whether this or the smaller T&M Leak Audit is the right size.

Who is the Enterprise Operations Review for?

GCs running $25M+ in annual volume across multiple project types or offices. The standard 10–14 page Operations Review is sized for a lean firm with one or two project types; at this scale, process variation by office, client segment, and trade is what hides the largest leaks.

Why is the price meaningfully different from the smaller reviews?

At $25M+ in annual volume, a 1% margin recovery is worth $250,000. The audit scope expands to cover that recovery surface — multi-office process variation, job-cost discipline at scale, sub coordination across trades, admin and back-office process. Delivered as a 30–50 page PDF within 10 business days.

What do you need from us to run it?

A multi-stakeholder intake covering estimating, change orders, job costing, sub coordination, and back-office process. Selected redacted bid packages, change orders, and job-cost reports from recent work. A scoping call to confirm what the audit will and won't cover — this is the only Prevaldi product with a synchronous scoping call.

What's not in scope?

Not an audit of your construction itself (quality, safety, schedule performance). Not a financial audit (your CPA does that). Not a software selection engagement (although the report flags overlaps and gaps in your existing stack). The Enterprise Review is an operations review of how the work flows, sized for firms whose operations are too varied for the smaller standard review.

Can we engage on a single office or division?

Yes. Some clients run the Enterprise Review on a specific office or division as a pilot before scaling it across the firm. Pricing scales accordingly — talk to us at intake about whether a full-firm or a single-division engagement makes more sense.

What does the Estimate Review look at?

Your estimating process across three to six recent bids. Where your bid template hides ambiguity, where scope gets undercounted, which specific phrases your subs read differently than the GC does. The audit names the language patterns showing up on bids you lose AND on jobs that go over.

Why bid review instead of takeoff review?

Because takeoff is a quantity-counting problem (Prevaldi doesn't read plan PDFs and produce quantities; that's a different hard problem). Bidding is a process problem — how the bid is structured, what gets included by default, what gets excluded, how addenda get folded in. Process gaps are where the audit finds margin.

What do you need from me?

Three to six recent bids you've submitted on real jobs (redacted is fine), a written description of your estimating process, and your bid template. We don't need bid totals, customer names, or signed contracts.

How fast is delivery?

PDF in your inbox within 5 business days of the intake form. One round of clarifying questions if anything is ambiguous — the clock pauses until you reply.

What does it cost?

$12,000 one-time. Subs and small GCs have publicly described missing $7K-scale scope items on individual bids; the audit is priced to pay for itself on the next two cycles where it surfaces a gap before bid day rather than after.

What does the Scope Gap Scanner do?

Reads plan notes or spec sections and flags what's missing — scope not assigned to a trade, coordination handoffs not addressed, code items buried in the spec that nobody assigned. The output is a structured list of gaps and commonly-missed items for the trade you're bidding.

Is this a substitute for a full plan review?

No. The Scanner reads the specific notes you paste — it doesn't have the plans, the other spec sections, or the drawings. It catches what you missed in the section you're staring at; the comprehensive plan review is still your job.

What text should I paste?

The plan notes, the spec section, the scope-of-work text, the room finish schedule — whatever has the words that describe what's expected. The Scanner reads the language. Drawings, schedules, and details that are only in PDF graphics won't help it.

Will it tell me whether the scope is constructible?

No. Constructibility is a different problem — sequence, conflict, means-and-methods. The Scanner catches what's missing from the words; the question of whether the words describe something buildable is on you.

Will it confirm code compliance?

No. The Scanner flags code items that aren't called out in the scope (e.g. “the spec doesn't mention fire-stopping at penetrations but the assembly described requires it”). It doesn't certify that the code references it flags are current for your AHJ. Treat code flags as “go look at this,” not “this is the rule.”

What does the Sub Scope Letter tool produce?

A written scope letter ready to send to a subcontractor. Project, sub trade, scope of work, start date, site contact — laid out in the format a sub recognises and a court would accept as a written record of what was agreed.

Is this a substitute for a subcontract?

No. The output is a scope letter — useful for early-stage coordination, for confirming what was agreed verbally, and for clarifying scope before a full subcontract is signed. The substantive subcontract terms (insurance, indemnification, payment, lien provisions) belong in a subcontract a construction attorney drafts.

When should I use this?

Early in the job, before the formal subcontract is in place. Right after a verbal handshake. When a sub starts work and you want a written record of what they're supposed to be doing. Anywhere the alternative is a text-message scope description.

What inputs do I need?

Project name and location, the sub's trade, the scope of work (be specific — “electrical” isn't enough; “rough-in for first-floor receptacles and lighting per Sheet E-201” is), the start date, and your site contact.

Should I send this in addition to or instead of an email?

Both. Send the letter as an email attachment with the body of the email recapping it briefly. The sub gets the letter; you get an email trail with the letter attached for the dispute file.

What is the T&M Leak Audit?

A targeted written review of where unbilled time-and-materials work is leaking through your field documentation. A foreman gets a verbal go-ahead, the work happens, the change order shows up two weeks later as a vague text — the GC denies it and the labor and material is already spent. The audit names the specific points in your workflow where the leak is happening.

What's the recovery range?

Subs on public construction threads have described losing $4,000 in a single month on one crew. That's $48,000 a year per crew — thirty-two times the $1,499 audit fee. The audit pays for itself on three weeks of typical T&M leakage on most operations.

What do you need from me?

A written description of how T&M tickets typically flow in your operation, three to six recent T&M tickets that disputed (redacted is fine), and the contract size range you typically work. We don't need contract values or customer names.

How fast is delivery?

PDF in your inbox within 5 business days of the intake form. One round of clarifying questions if anything is ambiguous — the clock pauses until you reply.

When is it available?

Currently on the waitlist. Add yourself and you'll get one email the moment it ships — no nurture sequence between now and then.