The winter build list: what to fix before spring runs you over

Here is what usually happens to a contractor's winter. Week one is collapse. Weeks two and three are last season's paperwork. Then there is a stretch of genuine quiet where the big plan is going to get made, and somewhere in there the quiet turns into equipment maintenance, next year's bids, a tax appointment, and one afternoon of looking at software demos. By the middle of March the field is calling and nothing in the office changed.

That is not a discipline problem. Ten weeks is not very long, and the office work does not come with a deadline the way a pour does. The thing that fixes it is picking a short list before January and refusing to add to it.

So: five things worth building over the winter, in the order we would do them, and two that are not worth your January.

1. The daily report, because it has to be habitual before it is useful

This goes first for a reason that has nothing to do with how valuable it is and everything to do with timing. A daily log only works if it is boring by the time the season starts. Build it in July and you get a half-adopted form and gaps on exactly the dates somebody asks about later.

Decide the fields now, cut the form until it fits one screen, make it work with no signal, and run it on whatever work you have in March so it is a habit by the time the real volume lands. We wrote the whole field list out separately — here is what belongs on it and what to leave off.

2. One handoff that keeps failing — and only one

Not the task you hate. The handoff: the place where information exists but does not travel. Field to office. Estimate to invoice. You to your foreman. Those are where the money leaks, and they are invisible because nobody owns the gap.

The test is cheap. For two weeks, write down every time somebody has to ask for information that already exists somewhere in your business. The thing that shows up most often is your winter project. It is almost never the thing you would have guessed, and it is almost never the thing you hate most — the task you hate is usually the one you already do carefully, precisely because you hate it. We went through the full method in what to automate first.

One handoff. If you pick three, you will finish zero.

3. Whatever the office retypes

Every contracting business has at least one chain where the same information gets entered by hand two or three times: a quote that becomes a work order that becomes an invoice; a timecard that gets re-keyed into payroll; a delivery ticket typed into a spreadsheet and then into an accounting system.

This is the cheapest category of work to fix and the easiest to prove, because the before-and-after is countable. It is also where the error cost hides — re-entry does not just cost the minutes, it generates the wrong-number-on-an-invoice problem that costs an afternoon and some credibility to unwind.

If you want a yearly figure on what this is costing you before you commit a winter to it, our paperwork cost calculator will do the arithmetic on your own hours. Every assumption in it is editable and visible, because the point is your number, not ours.

4. The one thing you have to remember, that only lives in your head

Lien and notice deadlines. Insurance certificate expiries on subs. Prequalification renewals. Bid dates. Certified payroll due dates. Warranty periods about to run out. Equipment inspections.

Every one of these is a date, every one of them is expensive to miss, and in most small contracting businesses every one of them lives in the owner's memory. This is the least glamorous item on the list and probably the highest return per hour you will spend all winter, because it is a one-afternoon build: a single list of dates with an owner and a reminder that fires early enough to act on.

It is not AI and it is not impressive. It is a calendar somebody looks at. Do it anyway.

5. Reading the documents you sign

This is the one that has changed the most in the last two years, and it is the one we would actually spend technology money on.

A spec book is six hundred pages and the five paragraphs that decide whether you get paid are scattered through it: the payment article, the retainage percentage, the notice provisions and how notice must be delivered, the quantity-determination clause, the schedule requirements that have to be formally accepted before any progress payment is due. Reading all of it properly takes days you do not have during bid season. Skipping it is how a contractor finds out in August what they agreed to in April.

Document reading and cross-checking is the thing current technology is genuinely good at — high volume, low judgment, never gets bored on page 480. It is not good at deciding what to do about what it found; that is still you. We built IAOIntel around exactly that split, and we were honest about the boundary in what AI is actually good at on a construction job.

The winter version of this project is small: take the three contracts you are most likely to sign a version of next season, and build a one-page summary of each one's payment, notice, retainage and quantity clauses. Do it once in the quiet and you will never again be reading the payment article for the first time with a bid due Thursday.

Two things not to spend your winter on

A platform migration. Moving your whole business onto a new system is a year-long project disguised as a winter project, it always lands mid-season, and it puts every one of your existing processes at risk at the same time. If something truly has to be replaced, replace it in the quiet — but do not confuse replacing one tool with migrating the company.

A rebuild of the thing that already works. There is a strong temptation in January to improve the system you are proud of, because it is the one you understand well enough to have opinions about. It is also, by definition, not where the losses are. Go where it is leaking, not where it is interesting.

How to know in April whether the winter worked

Pick the measure before you build, and make it something you can observe rather than calculate. Which crews filed a log yesterday. Whether anybody retyped that number this week. How many days the last pay application took to go out. Whether you can name the filing deadline on the last job you finished without going to look it up.

If you cannot name the measure, the project is not defined well enough to start — and that is a useful thing to find out in December rather than in June. One honest note on scale: a first automation that holds for a month teaches you what the second one should be, and you cannot learn that from a plan. So the goal for a winter is not a finished office. It is one system running in the real business before the first spring job, and a clear idea of what the next one is.

Want a second opinion on your winter list? Walk us through your week →