Almost every manufacturer past ₹80 crore produces a monthly MIS pack, and a striking number of them are produced by one person over four days, run to thirty pages, arrive on the ninth, and get read by nobody except the person who asked for them. I have been handed a lot of these. The effort is real and the return on it is close to zero.
The problem is rarely the content. It is that the pack is trying to be a record when what the room needs is a decision aid.
What should a production MIS contain?
Late and long are the same problem
A pack that takes nine days to produce describes a month you can no longer influence, so it gets read as history rather than as a prompt. And a pack that runs thirty pages has no argument in it, because the author could not decide what mattered and included everything to be safe.
Both come from the same place: the numbers are being assembled rather than read off. When assembly is the bottleneck, effort goes into producing the pack and none is left for thinking about it.
Five numbers, and why these five
On-time dispatch. The only number your customer also keeps. If it disagrees with theirs, that disagreement is the most valuable thing in the pack.
OEE on the constraint machine. Plant-wide OEE averages away the thing you care about. One machine governs your output; measure that one and let the rest be context.
Scrap as a percentage of material issued. Material is usually the largest cost and scrap is the part of it you can influence this quarter.
Inventory days. Working capital sitting in the racks, in a unit the board already understands.
Days sales outstanding. The other half of the cash conversation, and the one that tends to be somebody else's job until it is a problem.
We have written about choosing these in the five numbers every factory owner should know. What changes at board level is not which numbers but the discipline of keeping it to five.
Every number gets a name against it
A figure with no owner produces a discussion and no action. A figure with a name against it produces a conversation with that person, which is uncomfortable and useful. This single change does more for a management review than any amount of charting.
The corollary is that you cannot put a number in the pack that nobody can influence. If on-time dispatch is owned by nobody because it depends on four departments, that is the finding, and it matters more than the percentage.
"We cut it from thirty pages to one and the meeting got longer, which was the opposite of what I expected and exactly what we needed."Pump component manufacturer, Coimbatore
Same definition every month
The quiet killer is definitions that move. On-time dispatch measured against the original promise in January and the revised promise in March is two different metrics wearing one name, and the trend line between them is fiction.
Write the definition down once, put it in the appendix, and change it only deliberately with the change noted. This sounds pedantic until the first time a trend reverses because somebody adjusted a cut-off.
The successor taking over a plant usually asks for this pack first, and it is a good instinct: building it forces the underlying measurement to become real. That argument is in taking over the family manufacturing business.
Facto produces these five off the records the floor already enters, so the pack is read rather than assembled and it lands on day two. If your MIS currently takes a week, talk to our team.
