Tempo<\/a> application from the Atlassian Marketplace, and planning can begin.<\/p>\n\n\n\nPlanning time and capacity<\/h2>\n\n\n\n Of course, you also need to configure a few settings in Tempo afterward. For example, you need to enter public holidays depending on the country of use and define the respective working time models of the employees. In our case, we currently have about two different part-time models for students, while everyone else is on a full-time model. I won\u2019t go into more detail here, as my wife and I found our way very well at the time using Tempo\u2019s documentation and guides. You simply have to \u201cwork through\u201d them.<\/p>\n\n\n\n
Once the application has been successfully set up, you can start planning. To do this, simply go to the respective project board and create epics and issues there:<\/p>\n\n\n\nYou then need to assign three things to these issues so that Tempo can process them:<\/p>\n\n\n\n
\nWho is responsible for completing the task?<\/li>\n\n\n\n By when does the task need to be completed?<\/li>\n\n\n\n How many hours are estimated for it?<\/li>\n<\/ol>\n\n\n\nHere\u2019s an example of how I assigned a time block to myself for the initial rough planning:<\/p>\n\n\n\nLater, this naturally needs to transition into more detailed planning. In my opinion, the maximum limit for a single task is 80 hours. However, I would rather recommend 40 hours.<\/p>\n\n\n\n
Once you have completed these three things, you can already import the whole setup into Tempo:<\/p>\n\n\n\nMacht man das konsequent f\u00fcr alle Projekte, ist die Kapazit\u00e4tsplanung ein Kinderspiel. Man geht in Tempo einfach auf \u201eReports\u201c und w\u00e4hlt dort \u201ePlanned Time\u201c aus:<\/p>\n\n\n\nThen you simply select the desired observation period, and the report is ready.<\/p>\n\n\n\nSince I never explicitly plan general hours, my recommendation is to never plan utilization above 80% in principle. Due to an absence caused by a hospital stay, most employees were scheduled for more hours than their capacity allows in this particular month. Of course, this should remain the exception. Thanks to Tempo, however, this becomes visible early on, allowing you to respond in a targeted manner.<\/p>\n\n\n\n
Detailed reporting<\/h2>\n\n\n\n Once the workflow has been set up, everything else is just a click away. You can now reap the rewards of your work and obtain detailed reporting.<\/p>\n\n\n\n
What used to take several hours and many manual steps now happens with a single click: all four categories and their logged hours are displayed automatically and clearly.<\/p>\n\n\n\nWith another click, I export them to Excel and automatically transfer them via a script into my respective monthly report, which I prepare in PowerPoint.<\/p>\n\n\n\n\nNote: The y-axis intentionally does not display absolute values, as I do not necessarily want to share them and they are not relevant for the article.<\/p>\n\n\n\n
In the past, my decisions were often based on intuition; now they are based on clear facts and data. What I definitely had to acknowledge here:<\/p>\n\n\n\n
My subjective gut feeling often doesn\u2019t align with the actual data. Since we are a very small technical team and a lot of the organizational overhead is handled by PMT Holding, which AutoLab is part of, I had the impression that general hours didn\u2019t take up that much time. To my great surprise, however, I found that in the 2024\/25 financial year they accounted for a whopping 22.7% of all hours. In my opinion, that was far too much, and something had to change.\n\nAfterward, we looked at which marketing activities actually generate reach and how we could better integrate onboarding into existing projects. Many small adjustments were made. The result speaks for itself: we were able to reduce the share of general hours in total working hours from 22.7% to 14.3%.<\/p>\n\n\n\nBased on the data, we were able to identify another weak point: in some cases, we were spending too many hours on project inquiries. Especially in image processing, inquiries often come in at a very early stage, when there are no specifications yet and it\u2019s really just about an initial proof of concept. These potential projects are therefore very small-scale with low revenue. If you then invest several days just to process the inquiry, you actually end up losing money, even if you win the contract. So a change was needed here as well.<\/p>\n\n\n\n
The following adjustments were made:<\/p>\n\n\n\n
\nDefining a standardized inquiry process to identify potential early on<\/li>\n\n\n\n Developing questionnaires to help us work with the client to arrive at clear requirements more quickly<\/li>\n\n\n\n Faster estimates based on historical data (thanks to Jira\/Tempo)<\/li>\n<\/ul>\n\n\n\nUsing these approaches, we were able to reduce the processing time per project inquiry by 60.2%!<\/p>\n\n\n\nOf course, my annual report contains numerous other examples and analyses. However, the two examples mentioned are certainly the most striking\u2014and the ones that are literally saving us time and money.<\/p>\n\n\n\n
Why We All Plan Poorly\u2014and What Really Helps<\/h2>\n\n\n\n As mentioned at the outset, the intention behind our time tracking and logging was to obtain more realistic assessments. The analysis tools shown earlier also provide enormous added value, but targeted planning can be crucial for a company's survival. If a major project is completely miscalculated, the economic damage can be enormous.<\/p>\n\n\n\n
Once again, I would like to take this opportunity to refer to my absolute favorite book, \u201cThinking, Fast and Slow.\u201d It vividly demonstrates just how incredibly poor we are at estimation and planning. My personal experience from the project mentioned at the beginning is therefore also in line with the scientific consensus. A few examples from the book (from Chapter 23, The Outside View):<\/p>\n\n\n\n
\nThe construction of the Scottish Parliament was originally estimated to cost around 40 million pounds. Over the course of several \u201cfinal\u201d estimates, the costs rose gradually\u2014ultimately reaching over 400 million pounds.<\/li>\n\n\n\n An analysis of rail projects spanning several decades shows that passenger numbers were overestimated in more than 90% of cases. On average, forecasts were more than double the actual figures, while costs were, on average, about 45% higher than planned.<\/li>\n\n\n\n The same pattern holds true even for private projects: U.S. homeowners estimated their kitchen renovations at an average of about $19,000\u2014but actually spent around $39,000.<\/li>\n<\/ul>\n\n\n\nAs it turns out, we humans systematically underestimate the effort required for projects. That\u2019s exactly how I felt at the beginning with AutoLab.<\/p>\n\n\n\n
Fortunately, both Thinking, Fast and Slow and its follow-up, Noise, describe how to protect yourself from such misjudgments. The topic is a rabbit hole in itself\u2014you could easily spend days exploring it. I\u2019ll therefore keep this brief and focus on what I consider the most effective method: estimating based on base rates.<\/p>\n\n\n\n
The idea is simple: instead of making estimates \u201coff the top of your head,\u201d you base your calculations on actual data from comparable projects. For example, you take the average costs or workloads of similar rail projects and then slightly adjust them to fit your own circumstances.<\/p>\n\n\n\n
That\u2019s exactly how we handle project costs today. Thanks to our time tracking down to the minute, we now have an excellent database.<\/p>\n\n\n\n
When we receive a request today for a 3D scan using two laser profilers, we look at a similar project we\u2019ve already completed. Let\u2019s say that project took 240 hours. We then assess the differences\u2014such as the geometric complexity of the component. If the new part is more complex, we add a 25% buffer, for example, and arrive at around 300 hours.<\/p>\n\n\n\n
In practice, it has since become very clear that this offers several advantages:<\/p>\n\n\n\n
\nThe estimates are significantly more accurate than they used to be<\/li>\n\n\n\n The estimates are significantly more accurate than they used to be<\/li>\n\n\n\n Our customers receive reliable quotes faster<\/li>\n<\/ul>\n\n\n\nI highly recommend that everyone base their estimates on data in this way.<\/p>\n\n\n\n
Project Monitoring<\/h2>\n\n\n\n Another phenomenon that is frequently observed in projects is what is known as Parkinson's Law:<\/p>\n\n\n\n
Work expands to fill the time available for its completion.<\/p>\n\n\n\n
Or, to put it simply: If I budget 100 hours for a task, it will most likely take 100 hours\u2014even if, in theory, it could be completed in 70 hours.<\/p>\n\n\n\n
The reasons for this are hardly surprising:<\/p>\n\n\n\n
\nPeople tend to be less focused because there is no real time pressure<\/li>\n\n\n\n Tasks become unnecessarily complex (\u201coverengineering\u201d)<\/li>\n\n\n\n Perfectionism<\/a> creeps in<\/li>\n\n\n\nDecisions are postponed<\/li>\n<\/ul>\n\n\n\nThis is exactly where good project monitoring becomes crucial. In the past, it was essentially a blind flight for us. Once planned, we just hoped it would somehow work out. If it didn\u2019t, we usually realized it too late.<\/p>\n\n\n\n
Heute ist das komplett anders. Dank Jira und Tempo haben wir jederzeit einen Echtzeit-SOLL-IST-Vergleich auf allen Ebenen. Ich sehe zu jedem Zeitpunkt:<\/p>\n\n\n\n
\nHow many hours were planned?<\/li>\n\n\n\n How many have already been booked?<\/li>\n\n\n\n Where are we going off track?<\/li>\n<\/ul>\n\n\n\nHowever, what matters most is not transparency alone, but what we do with it. As soon as we see a task starting to get out of hand, we take proactive action. This may mean:<\/p>\n\n\n\n
\nNarrow the scope<\/li>\n\n\n\n Redefine tasks<\/li>\n\n\n\n Consciously accept a \u201cgood enough\u201d solution<\/li>\n\n\n\n Or, if in doubt, refine it early on<\/li>\n<\/ul>\n\n\n\nSo we continuously make adjustments to stay on track, rather than being caught off guard at the end.<\/p>\n\n\n\n
In Kombination mit k\u00fcrzer geschnittenen Tasks (ideal < 40 h) entsteht dadurch ein sehr robuster Regelkreis: planen \u2192 umsetzen \u2192 messen \u2192 nachjustieren.<\/p>\n\n\n\n
And that is precisely what prevents workloads from spiraling out of control in practice.<\/p>\n\n\n\n
More than just internal: Direct benefits for customers<\/h2>\n\n\n\n One aspect that is often underestimated is the direct value it adds for the customer.<\/p>\n\n\n\n
Since we have to assign every task to an issue anyway, and Tempo is configured so that time cannot be logged without a comment, this automatically results in a complete, minute-by-minute record.<\/p>\n\n\n\n
This can be exported at the touch of a button and made available to the customer with full transparency.<\/p>\n\n\n\n
In my view, this is the fairest and most transparent solution, especially when it comes to consulting services, support contracts, or traditional hourly packages. The client can see exactly what the time was spent on, with no room for interpretation.<\/p>\n\n\n\n
Another advantage becomes apparent during the course of the project. For example, when a client books a 100-hour consulting package, it typically covers several topics. In practice, it often turns out that certain tasks require significantly more effort than originally anticipated due to unclear requirements or missing information.<\/p>\n\n\n\n
Thanks to our monitoring, we identify such developments at a very early stage and communicate them proactively.<\/p>\n\n\n\n
This is consistently very well received by our customers because our internal project monitoring indirectly helps them better manage their own projects as well.<\/p>\n\n\n\n
Conclusion and outlook<\/h2>\n\n\n\n At the end of the day, it\u2019s clear that our time-tracking system has several distinct benefits:<\/p>\n\n\n\n
\nSignificantly better project planning and far fewer unpleasant surprises<\/li>\n\n\n\n Faster quote generation and, as a result, noticeably better customer service<\/li>\n\n\n\n Close, data-driven project monitoring<\/li>\n<\/ul>\n\n\n\nAll of this adds up to significant time and cost savings. Even for a small team like AutoLab, we\u2019re quickly talking about low six-figure savings in internal costs per year.<\/p>\n\n\n\n
And in my view, that\u2019s just the beginning. The more data we collect, the better our decisions become\u2014and the stronger this effect becomes.<\/p>\n\n\n\n
<\/p>","protected":false},"excerpt":{"rendered":"
Vom Albtraum zur Methode Manchmal merkt man erst mittendrin, wie weit man vom Kurs abgekommen ist. Im Oktober 2022 startete ich ein Projekt, das im November 2023 abgeschlossen sein sollte. Doch was als \u00fcberschaubares Unterfangen begann, entwickelte sich rasch zu einem echten Albtraum. Bis Mai 2023 waren erst 30 % erledigt, und die Erkenntnis traf […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[17],"tags":[7,39,14],"class_list":["post-399","post","type-post","status-publish","format-standard","hentry","category-wirtschaft","tag-autolab","tag-projekt-management","tag-software-development"],"_links":{"self":[{"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/posts\/399","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/comments?post=399"}],"version-history":[{"count":18,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/posts\/399\/revisions"}],"predecessor-version":[{"id":445,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/posts\/399\/revisions\/445"}],"wp:attachment":[{"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/media?parent=399"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/categories?post=399"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/tbenninger.com\/en\/wp-json\/wp\/v2\/tags?post=399"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-1024x475.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-1.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-2-1024x649.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-3.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-4.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-5.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-6-1-1024x238.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-7-1024x435.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-8-1024x317.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/image-16-1024x305.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/BookingsperProjectCategoryWeeks-1024x198.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/MontlyHoursTotal.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/GeneralHoursTotal.png\")
<\/figure>\n<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/Percentage-of-general-hours-to-total-hours.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tbenninger.com\/wp-content\/uploads\/2026\/03\/Hours-per-project-request.png\")
<\/figure>\n\n\n\n