Tool-agnostic by design: every check is described by what the report returns, and Sage, NetSuite, Epicor, Dynamics, SAP, Oracle, QAD, Acumatica, or a homegrown system can produce it. Each tool is a skill you run in your own AI assistant plus a CSV workbook you keep. One of them (the Buy-In Resistance Map) is the primary recommendation of the founder's doctoral study of ERP implementation in manufacturing SMEs, turned into an instrument. Before the toolkit, score the project with the ERP Implementation Readiness Scan and the data with the Six ERP Failure Modes Self-Scan.
Who runs it: the cost accountant, controller, or whoever owns the cost roll.
What it saves: The mis-quote and the margin drop nobody can explain. A standard that moved without a dated reason is a leak; this names it the week it happens.
What it hands a reviewer: A dated record of every standard cost change above tolerance, with the reason, the owner, and the quotes or orders it touched. It is the costing evidence any margin review, audit, or AI quoting tool would need.
Who runs it: the operations lead, controller, or ERP administrator; anyone can contribute rows.
What it saves: The decision made from a spreadsheet nobody else can see, the migration the ERP project did not know about, and the single-owner file that leaves with the person.
What it hands a reviewer: The inventory of every spreadsheet or side tool acting as a system of record, scored on risk, with an owner and a retire, keep, or move-to-ERP decision. It is the shadow system map every ERP project and every AI readiness review starts by building.
Who runs it: the executive sponsor and the project lead, with one conversation per department.
What it saves: The go-live that technically works and nobody uses. Resistance that is named before kickoff is a design input; resistance discovered at training is a crisis.
What it hands a reviewer: A department-by-department record of what each group stands to lose, why they might resist, what was offered in response, and whether they can say what the system gives them. It operationalizes the primary recommendation of the founder's doctoral study of ERP implementation in manufacturing SMEs: identify the causes of resistance to organization-wide buy-in.
Who runs it: the quality lead, operations lead, or whoever owns the procedures.
What it saves: The SOP the floor stopped following two years ago, the training given on a procedure that no longer matches the system, and the audit finding that follows both. For AI, it stops an assistant from confidently quoting a procedure nobody uses.
What it hands a reviewer: Every procedure with its last review date, its owner, whether the floor actually follows it, and whether the system matches it, scored for freshness. It is the document control evidence an ISO, FDA, or customer auditor asks for and the trusted-document list any AI assistant must be built from.
Who runs it: the selection team: sponsor, process owners, finance, IT.
What it saves: The ERP or SaaS purchase decided by the best demo. A scored evaluation on your own processes, with the same questions put to every vendor, is what keeps the decision defensible when it is questioned in year two.
What it hands a reviewer: A side-by-side scorecard of every vendor on the same criteria, weighted by what matters to your business, with demo evidence, reference-call notes, and a five-year cost line. It is the selection record a board, a lender, or an auditor asks for.
Who runs it: the controller and the data owners named for each master and balance.
What it saves: The first month-end on the new system that cannot close because opening balances do not tie, the item master that loaded on the wrong cost basis, and the customer who was invoiced twice. Each tie-out is a pass or fail with a signature before the go-live date is confirmed.
What it hands a reviewer: A pass or fail per data set at each mock load and at cutover: record counts, control totals, opening balances by ledger, sample traces, and the signature of the business owner. It is the go-live acceptance evidence an auditor asks for and the Deploy stage of GROUND with the proof attached.
These are the worries, in the words a plant president, a controller, an ERP project lead, or a contracts owner actually uses. Each one names the first move and the free tool that produces the evidence. BBI's own answers; no outside statistic is claimed.
You are right to doubt it until the evidence says otherwise. Run the ERP Implementation Readiness Scan tonight (ten minutes); if it says not yet, the Cutover Readiness Scorecard gives you the weekly go or no-go and the Cutover Tie-Out Workbook gives you the proof per data set.
Say so with evidence instead of instinct. The Six ERP Failure Modes Self-Scan gives you a READY to DO NOT DEPLOY band and the exact checks to run; the AI Readiness Checklist turns one use case into the brief that gets a yes or an honest not yet.
Start the AI Use-Case Register this week; it is the one page an auditor, a prime, or a cyber insurer asks for first. The Vendor AI Terms Watch answers the follow-up question about whose model trains on your data.
Check the cost side before the price side. The Standard Cost Drift Check compares the last two cost rolls and names the items whose standard moved without a dated reason.
Count the files like that before one of them walks out. The Shadow Spreadsheet Census scores every single-owner file that feeds a decision or the ledger and forces a keep, retire, or move decision.
Put every registration, certification, and attestation on one calendar with its expiry and owner. The Compliance Evidence Calendar flags anything inside 90 days on the first business day of the month.
Panic searches, answered. The first move is always to make the problem visible in a dated log before fixing anything, because the fix without the record is the finding you meet in the audit.
Pull the report of every item and location below zero with the last transaction that drove it there, freeze planning on those items, and fix the driving transactions before the next MRP run. Mode 1 of the Failure Modes Self-Scan is the check; the Master Data Change Watch keeps it from coming back.
Log every disagreement with a date and an owner before you fix any of them; the oldest open one is the audit finding. The Month-End Disagreement Check is that log and it takes a few minutes on close day.
Do not go live Monday. Record the variance per data set in the Cutover Tie-Out Workbook, get a business owner's signature on what does tie, and move the date to follow the evidence. A date without criteria is a hope.
Start the AI Output Review Log today and backfill nothing. Record this batch honestly, including that it reached the customer unreviewed, and put a named reviewer between the output and the next action. Then use the AI Readiness Checklist to write the brief leadership will ask for.
Find the price or cost change that caused it, with the date and the person, before you touch the quotes. The Standard Cost Drift Check and the Master Data Change Watch are the two logs that answer that in minutes instead of a week.
Write the criteria and the weights tonight, before the demo sets them for you, and insist on your own items and your own month-end in the demo. The Vendor Evaluation Scorecard has the six criterion groups and the 0 to 4 evidence scale ready to fill.
Two pages answer it: the AI Use-Case Register (what AI runs, on what data, with what review) and the Vendor AI Terms Watch (whether each vendor trains on your data, retention, subprocessors). Start both with one row each today; honest and incomplete beats blank.
Because the cost roll ran and nobody compared it with the last one. The Standard Cost Drift Check takes two exports (prior standard, new standard), lists every item that moved past your tolerance, and forces a reason and an owner for each. It runs after every cost roll.
Export item standard cost as of the last roll and as of now, and diff them. The Standard Cost Drift Check does that in your AI assistant from the two lists and writes the drift log, largest move first, with the quotes and open orders each change touches.
Check the cost side first. A standard that moved without a dated reason is the usual cause, and it shows up in margin weeks before anyone looks at the cost roll. The Standard Cost Drift Check names the items and sizes the exposure with the arithmetic shown.
You do not need a module. Any ERP exports an item cost list. The Standard Cost Drift Check compares two of them and keeps a dated CSV of every change, reason, owner, and exposure, which is the costing evidence a controller or auditor asks for.
Ask one department at a time what they keep in a spreadsheet because the system cannot or does not, and score each file for risk. The Shadow Spreadsheet Census gives you the questions, the five-point risk score, and the keep, retire, or move decision with an owner and a date.
Most companies do not know until they count. The Shadow Spreadsheet Census is the count: one row per file, who owns it, what decision it feeds, whether it is keyed into the ERP or the GL, and whether a backup exists. Run it quarterly.
The record leaves with them, and whatever decision it fed is made blind until someone rebuilds it. The Shadow Spreadsheet Census flags every single-owner file that feeds a decision or the ledger as risk 4 or 5 and puts it at the top of the decision list.
Run the Shadow Spreadsheet Census before the project plan is final. Every high-risk file is a migration or a retirement the plan has to include; finding them after kickoff is how implementations slip.
Name what each department stands to lose before you tell them what they will gain. The Buy-In Resistance Map is a fifteen-minute conversation per department, recorded in their words, with the trial offered and the response agreed. It operationalizes the primary recommendation of the founder's doctoral study of ERP implementation in manufacturing SMEs.
Because the new system takes something from them that made their day work: a spreadsheet they control, a report only they can run, a workaround. Resistance is rational. The Buy-In Resistance Map finds the loss, which is where the resistance is, and records the response.
Identify its causes before kickoff, department by department, and give each cause a response and a trial on real work. The Buy-In Resistance Map is that process as a workbook, revisited at each phase gate.
What do you keep today that the new system will take away? What are you afraid will happen in the first month? In your own words, what does it give you that the current one cannot? What would you need to see on real work before you believed it? The Buy-In Resistance Map asks these in order and records the answers.
Score each procedure on four questions: reviewed on time, followed on the floor, matched by the system, used in training. The SOP Freshness Scorecard does this a handful at a time each month and names the stale and the fictional ones.
Ask one of the people doing the work, not the document owner, and record the answer. The SOP Freshness Scorecard makes that one question a monthly habit and turns the answers into a trusted list of procedures that are actually live.
Build the trusted list first. An AI assistant trained on a procedure the floor abandoned two years ago will quote it with confidence. The SOP Freshness Scorecard produces the list of procedures scoring four of four, which is the only set an assistant should be built from.
One of them changed and the other did not. The SOP Freshness Scorecard checks the system-matches-it question for every procedure and records an action (revise, retire, retrain, or change the system) with an owner.
Set weighted criteria before the first demo, run the same scripted demo on your own data with every vendor, and score from evidence only. The Vendor Evaluation Scorecard holds the criteria, the weights, the evidence source for every score, and the five-year cost line.
Fit to your core processes, data and migration tooling, integration and reporting, implementation method, vendor and partner references, and five-year cost. The Vendor Evaluation Scorecard organizes these into six groups you weight once and score for every vendor.
Script the demo on your items, your BOMs, and your month-end, give every vendor the same script, and score 0 to 4 from what was shown on your data, not what was claimed. The Vendor Evaluation Scorecard records the evidence source for every score; a score with no evidence is a 1.
What went wrong, how long go-live took against plan, what they would do differently, and whether they would choose the same partner again. The Vendor Evaluation Scorecard has a reference-call column and asks you to record the answers in the person's words.
Tie them out at each mock load, not at cutover. The Cutover Tie-Out Workbook records source and target counts, control totals, sample traces, and a business owner's signature for every data set, so the first close on the new system is not the first test.
Record counts, a control total the business recognizes, a sample trace of the ugliest records field by field, a pass or fail against a written tolerance, and a signature. Per data set, per load event. That is the Cutover Tie-Out Workbook row.
At least two, and the second must pass every data set before the go-live date is confirmed. The Cutover Tie-Out Workbook tracks each load event; BBI's standing rule is that the date follows the evidence, never the other way.
The business owner of each data set, not IT and not the vendor. The Cutover Tie-Out Workbook has a signed column per row, and an unsigned pass is not a pass.
Each recurring tool follows an event that already happens (the cost roll, the close, the quarter), never a clock time. A run is one row; the first run is two minutes. A skipped run is followed by a one-row run, not a catch-up. Every run ends with one finding you did not type in. The project tools (Buy-In Resistance Map, Vendor Evaluation Scorecard, Cutover Tie-Out Workbook) run in sessions short enough to finish and end with one next physical action with an owner and a date. No points, no badges; the workbook is the score.
Yes, if your ERP can export a list. Every check is described by what the report returns (items below zero, standard cost by item, vendors grouped by tax ID), not by a vendor's menu. If you name your system, the assistant translates the check into its terms.
The skill file is written for Claude, ChatGPT, or Copilot, but the CSV workbook and the first-run card stand on their own. You can run every tool by hand with the workbook alone.
The Buy-In Resistance Map operationalizes the primary recommendation of Dr. Gilliam's 2022 doctoral study of ERP implementation strategies in U.S. manufacturing SMEs (Walden University; record at scholarworks.waldenu.edu/dissertations/12693). The other five come from BBI's ERP operations and costing work. The questions, scores, and workbook designs are BBI's own.
No. The tool runs in your assistant under your account and the CSV stays in your files. Nothing is sent to BBI.
CC BY-NC 4.0. Use it, adapt it, and share it with the provenance line intact. Do not sell it or present it as your own. The LICENSE.txt in each download says the same in plain words.
BBI's ERP readiness review is the verified version: a fixed-scope look at your process decisions, data ownership, change plan, and cutover criteria, with a remediation plan in GROUND order. Bring the workbooks; the engagement starts on day one instead of day fourteen.
