Even a small logging inaccuracy eats into your margin. Learn why real-time time tracking is the most important tool for small construction companies.
An unrecorded working hour never shows up in red on a report. It causes no error message and does not get caught in an audit. It is simply absent, and because it is absent, the project looks more profitable than it was.
This is why missing hours are a harder problem than most other measurement problems in a construction company. A wrong figure gets noticed. A missing one does not.
The consequence appears in the next quote. When the recorded hours for a contract are ten per cent below reality, the next similar job is priced ten per cent too low.
Under-recording is not laziness. It comes from four structural causes, and each needs a different remedy.
The entry is made at the wrong time. When hours are filled in on Friday for the whole week, they are filled in from memory. Memory rounds down and flattens the days into equal lengths. Monday's seven and a half hours and Wednesday's nine both become eight. The exceptions vanish too: the day spent waiting two hours for materials is exactly the one profitability tracking would learn most from, and it is the first to be forgotten. And if the week covered three sites, memory does not separate them to the hour.
The work does not look like work. Fetching materials, moving between sites, cleaning up and maintaining tools are all working time, but they do not feel like the contract. They go unrecorded because they are not felt to belong to the project.
Recording costs more than it gives. If logging an hour requires signing in, finding the project in a list and filling three fields, it happens once a week. If it takes two taps, it happens daily.
Nobody looks at the entries. If hours are never reviewed and nothing follows from them, recording becomes a ritual. Rituals get done at the minimum.
The figures are worth calculating with your own crew, because an abstract percentage convinces nobody.
Take a construction company of six. Each person leaves one hour a week unrecorded. That is six hours a week, roughly 270 hours a year.
At a billable rate of 55 euros, that is about 15,000 euros of unbilled work a year. For comparison, a construction ERP for six users costs less than 2,000 euros a year.
One hour a week is also a conservative estimate. When entries are made weekly from memory, the loss is typically larger, because the rounding happens for every day rather than once a week.
The most common attempt at a fix is to estimate the missing time afterwards. The site manager looks at the week and adds a couple of hours where there probably were some.
That does not correct the error, it relocates it. An estimate does not know which project the hours belonged to, so it distributes them wrongly. For profitability tracking the situation is now worse than before: the total is closer to the truth, but the split between projects is further from it.
The total is what interests the bookkeeper. The split between projects is what interests you, because it tells you which kind of work pays and which does not.
This is the heart of the matter, and it is easy to miss because the effect is delayed.
The accuracy of cost estimation rests on your own history. Industry averages tell you what others do, your own history tells you what you do. If the history is skewed downwards, every calculation that uses it is skewed downwards.
The loop runs like this. Hours are recorded short, so the job looks cheap. The next comparable job is priced accordingly. It is done at a loss, but the loss is invisible because the hours are again recorded short. The third job is priced lower still.
The loop usually breaks only at the year end, when the accounts do not agree with the projects. By then a year has passed.
Travel and moves. The morning drive from the yard to the site, moving between sites mid-day, collecting materials. Here the question is not only recording but whether the company has agreed what travel counts as working time. If that has not been agreed, every worker decides it differently and the figures are not comparable with each other.
Extra work. The customer asks for a small change on site, it is done there and then, and it is recorded nowhere. The hours land on top of the base contract but never reach an invoice, because nothing says they existed. This is the single largest source of unbilled work in small construction companies.
Finishing. The tail of a contract is full of short visits: a warranty fix, a forgotten item on the snag list, a customer's remark. The project has mentally ended, so recording stops before the work does. These hours also determine the real margin, because they come entirely out of it.
Under section 32 of the Finnish Working Hours Act, employers must keep working time records. They cover hours worked, including overtime and additional hours, and the compensation paid for them, and they must be kept for at least two years from the end of the calendar year in which the right to compensation arose. Two practical risks follow from that.
The first is a dispute. If a worker claims overtime and the employer has no records, the employer is in a weak position. An entry created during the day carries different evidential weight than a table filled in afterwards.
The second is that overtime goes unnoticed until it is too late. When hours appear only at the end of the month, the overtime has already accrued and nothing can be done about it. When they appear daily, supervision can level the load before the balance grows.
Time tracking most often fails because it is experienced as surveillance. That deserves to be taken seriously, because there is something to it: an entry really does say what a person did.
Two things change attitudes more than anything else.
Say what the data is used for. If hours steer how the next contract is priced and staffed, say so out loud. Nobody is enthusiastic about recording data whose purpose is unclear.
Make it useful to the person recording. When a worker sees their own hours, overtime balance and leave in the same place where they record them, recording stops being one-directional.
On location data, be direct. If it is collected, say why and when. If it is not, say that too. Ambiguity is worse than either answer.
This is a subtler problem than under-recording, and it is more common than people think.
When choosing a project means picking from a list and the list is long, people pick the first familiar-looking one. When the same customer has ordered three different jobs, hours easily land on the wrong one.
The effect is double: one project looks better than it is and another worse. Neither figure is usable, and the total looks right, so the error goes unnoticed.
The practical fix is to narrow the choices to what is actually possible for that worker on that day. When the list has two options rather than twenty, the chance of a wrong pick collapses.
These get conflated, but they call for different action.
Unrecorded work is nowhere. It cannot be invoiced, because nobody knows it exists. The fix is in how recording happens.
Unbilled work has been recorded but left off an invoice. It is known, it shows on a report, and it can be fixed by invoicing it. The fix is in the invoicing process.
The second is easier, because the problem is visible. That is why recording is worth getting right first: it turns an invisible problem into a visible one, and a visible problem is already half solved.
Three signs that show up without a separate investigation.
The annual result does not match the sum of the project reports. If the projects look good but the result does not, there are costs that were never allocated to projects. Hours are the usual explanation.
Every day is eight hours. A real day on site is not. If the material is smooth, it has been smoothed.
Comparable contracts produce different margins. When two similar jobs differ clearly, the difference is either in the content of the work or in the accuracy of recording. The latter is more common.
What an unrecorded hour costs depends on how the work was priced, and the difference is worth understanding before trying to fix anything.
In hourly work an unrecorded hour is directly an unbilled euro. The damage is immediate, measurable, and hits cash within the same month. The good news is that it is also the easiest to fix: once the hour is recorded, it gets invoiced.
In a fixed price contract an unrecorded hour does not reduce the invoice at all. The invoice is agreed. The damage is that the margin looks better than it was, and that figure travels into the next quote. The damage is slower but wider, because it compounds.
In extra work the worst of both combine. The work is done on top of the contract and should be invoiced separately. If the hours are not recorded at all, the money is never collected and the contract margin looks better than it was. If they are recorded, but against the contract rather than as extra work, the invoice still never goes out and the contract margin looks worse than it was. Either way, the extra work disappears from the invoice.
The practical conclusion: if you mostly do hourly work, fixing your recording shows up in cash within a month. If you do fixed price contracts, it shows up only in next year's quotes, but on a larger scale.
A common situation is that recording gets fixed now while the history stays incomplete. That raises the question of whether the old material can be repaired afterwards.
The short answer is to chase the billable hours and leave the rest alone.
Chase what has monetary value. Go through the unbilled contracts of the last three months and check whether the invoiced hours match the work done. There is usually some trace: calendar entries, the site diary, purchases from the same day. If the gap is clear, it is billable.
Do not repair history by guessing. Filling old months with estimates makes the material worse than it is. An incomplete but honest history says that hours are missing. A filled-in history looks whole and is wrong, and nobody can tell which is which.
Mark a cut-off date. Write down the day from which recording is sound. When you calculate unit rates a year from now, you use only the material after it. That single note saves you from the situation where you compare usable and unusable material without knowing it.
In practice three months of sound material is worth more than two years of vague material. History is better started afresh than patched.
When recording is tightened, the same three arguments come back from the crew. They are not excuses, they point at where the arrangement is broken.
"There is no time for this." If an entry takes more than half a minute, this is true. The answer is not to demand more but to measure how long one entry actually takes. If the answer is two minutes, the problem is not attitude.
"I cannot remember which project." This says the entry is made too late. The same person remembers at three in the afternoon but not on Friday. The answer is to move when recording happens, not to improve memory.
"Why does this have to be tracked." This is an honest question and deserves an honest answer. If the answer is "because management wants it", recording will be done at the minimum. If it is "because the next quote is calculated from these figures, and a quote priced too low means pressure on you", it is concrete and understandable.
A fourth objection is usually not said out loud: the fear that the figures will be used to assess individuals. It is worth raising yourself, because an unspoken assumption steers behaviour more strongly than a spoken one.
In FiSAS, time recording is built around the four reasons hours go missing:
- Recording on site in a few taps. The worker picks the site, cost allocation and type of work, marks start, end and break, and sends the entry from their phone on the same day. Recording works without a connection too and syncs when the connection returns.
- The right site ready to pick. Workers see their own tasks and work orders in the calendar, so the entry lands on the site they were assigned to that day.
- Extra work signed off in the field. The client can sign off working hours directly on mobile, so extra hours are approved before invoicing.
- Someone looks at the entries. Supervisors see the hours immediately and approve them, and every worker sees their own entries on their phone.
- The hours get used. Approved hours flow into project cost tracking and onto the invoice, so the same entry serves profitability, invoicing and working time records.
Location is stored with each entry, so as noted above, tell the crew about it directly. Time tracking is included in every plan, including the single-user Maker plan.
The working hour is a construction company's most precise measure of what work actually costs. It is also the measure that leaks most easily, because the leak is invisible.
The fix is not more careful after-the-fact calculation. It is that the entry happens when the work happens, lands on the right project, and is easy enough for the person recording it to do every day.
If you want to know how much leaks in your own company, start with one active site and record hours there daily for a month. Compare the result with what you would have estimated in advance. The difference is what you have been paying for guessing.
You can run the trial in FiSAS free for 14 days, and two weeks of daily entries already show how far your estimates have been off.
