The costs of doing nothing
Should we wait another year before we connect the floor? Waiting is a decision with a monthly price, and that price belongs on the same sheet as the project quote.
What a do-nothing baseline actually is
A do-nothing baseline is the cost of running the plant next year exactly as it runs today. It is not a project and it never arrives as an invoice, so it rarely gets compared with the invoice for a change. A total cost of ownership view of any Industry 4.0 option is incomplete until the option it replaces has been priced the same way.
The question here is simple. If nothing is approved, what does each month cost, and what does each month of delay forgo? Two neighboring questions are covered elsewhere. The standing tax that a program pays after the roadmap shows green is described in disconnected smart factory programs still pay an operating tax. The maturity question of why earlier pilots stalled is covered in pilot graveyards are the real maturity baseline.
The monthly lines that keep running
Do-nothing is made of recurring lines that already sit in the ledger under other names. Most of them are OPEX, so they do not look like decisions. Listing them separately is what turns a vague worry into a number a finance owner can compare.
Walking and searching come first because they are the easiest to sample. Time spent looking for a cage, a truck, or a person is paid time with no output, and it repeats on every shift. The size of that line is argued in walking still costs more than most dashboards admit. Empty forklift rounds, a coordination seat staffed only to translate between screens, and weekend crews booked to recover work that was lost because nobody could see the state of the floor during the week belong in the same list.
Quality and energy belong there too. Scrap and rework that a shared view might have caught earlier feed the cost of poor quality. Equipment that stays powered with no production order running shows up in the electricity bill and in the reported scope 2 emissions figure at the same time. None of these lines is large on its own day. The baseline is their sum, repeated every month the plant waits.
What published benefit ranges say about forgone gains
Every benefit range that consulting firms and researchers publish for connected manufacturing has a mirror image. A reported reduction is also the size of the loss that continues if nothing changes. Some published ranges cite machine downtime reductions of 30 to 50 percent and maintenance cost reductions of 10 to 40 percent for the strongest predictive maintenance cases, which are selected examples and not typical results. Ranges for broader smart manufacturing programs usually span productivity, lead time, and quality cost.
Those numbers are hypotheses for a class of plants, not forecasts for yours. They are useful for one thing only. They show that the forgone side is not zero, because a plant that adopts nothing keeps at least part of the loss that the best cases reduced. They do not say how much of it a given plant would capture. The local version of the calculation starts from a downtime log you already trust and from the contribution of one lost hour on the constraint. Where the gains of a connected program tend to show up on the floor is covered in where Industry 4.0 productivity gains show up.
Why failure rates do not make waiting safe
The usual defense of doing nothing is the failure statistic. Published figures for digital twin and transformation initiatives are discouraging, but they differ in source, sample, and definition of failure. Published surveys often put the share of industrial digital transformation efforts that stall before they scale at around 70 percent, and reported failure rates for digital twin programs vary widely with how failure is defined. Treat these as warning signals, not as measured base rates for your plant.
These figures are real warnings, but they point at the shape of the buy, not at standing still. Programs fail when scope is wide, ownership is vague, and the first result is a dashboard nobody uses. A plant that waits keeps its baseline cost and may meet the same failure modes later, because a small, honest attempt never taught it the lessons. The cost of a trial that goes wrong is its own question and is priced in the real cost of a failed pilot. A narrow trial with a written pass line is usually a smaller commitment than a year of baseline on the same loss, and your own numbers will show whether that holds.
Maturity scorecards can hide this trade. A plant can score well on connectivity and architecture and still pay every line above, as described in Industry 4.0 scorecards still grade IT, not the shift. A digital maturity score does not replace a monthly baseline.
What waiting does to the data you could have had
There is one cost of delay that no benefit range includes, and it is the history that never gets recorded. Place and time data cannot be reconstructed afterwards. A plant that starts storing positions and events a year later has a year less to learn from, and that gap does not close by spending more later.
That is why an operational digital twin can be read in three ways when the delay is priced. Live logistics lets the shift find the nearest free forklift or driver today. The spatially oriented dataset under the live view keeps the place-and-time record. Correlation and causality analytics then use that record to show which events keep preceding the same loss. A plant that waits postpones all three at once. It also postpones the evidence needed for lean manufacturing work on waste, because the Check step of PDCA without a record is a guess.
The same applies to everyday records. Without a spatial record, asset management and inventory management keep depending on counts and memory, and the cost of searching stays in the baseline.
How to price the baseline before comparing options
1. List the recurring lines in plain categories - Walking and searching, empty moves, rework and scrap, idle energy draw, duplicate coordination seats, and weekend catch-up shifts. One peak fortnight of sampling is enough to start.
2. Put a money value on one hour - Use the contribution of a lost hour on the constraint and the loaded cost of a paid hour of searching. If neither exists, that is the first finding.
3. Mark which lines a shared place-and-time view could change - Not every line is addressable. Leave out the ones that depend on supplier or market behavior.
4. Treat published ranges as a mirror, not a promise - Use them to confirm that the forgone side is not zero, and let local numbers set the size.
5. Compare the baseline with a narrow trial, not with a plant-wide program - The honest comparison for waiting is a two-to-four-week zone with a named owner. How that trial is priced is in what a two-to-four-week hall pilot actually costs to learn.
A smart factory program is easier to approve when the do-nothing line sits next to it on the same sheet. twinzo prices by data points rather than named-user seats, so a narrow zone stays a narrow commercial footprint. Current commercial detail is on pricing, and the ROI calculator turns your own minutes and cost per minute into an indicative figure. Capability depth for the live layer is on internal logistics optimization and the features overview. Get in touch if you want to price your own do-nothing month against one loss you can already name.