“We usually respond immediately,” says the office management. The team is referring here to the automatic receipt confirmation. An interested person, however, means by response the clarification of her specific question. Both are talking about the same message and yet are measuring something different.
The following workshop proposes an internal evaluation method for real-estate companies in Austria and Germany. All times and cases are fictitious. They are neither industry benchmarks nor claims of performance about any actual office. Legal or contractual deadlines are not determined here.
The first decision: which event stops the clock?
Separate three possible events in your own process: technical receipt confirmation, first human reaction and professional answer to the specified question. A personal message can also merely confirm that further research is being carried out. It is then an important contact, but not yet the final substantive clarification.
Define the events so that two employees would classify the same case in the same way. “Processed” is often too vague for that. Describe which actually documented step is meant. If a question contains several separate points of substance, the scope chosen for measurement must also be known.
A neutral transaction identifier helps with your own evaluation. It does not automatically anonymise an actual customer case. For real data, purpose, necessary scope and lawful processing are therefore determined separately; our calculation workshop uses only fictitious cases.
Four enquiries and two different pictures
EnquiryA is received on Monday at 9 am. The first human reaction follows after 30 minutes, the specified professional answer after one hour. EnquiryB arrives on Monday at 10 am; the human reaction follows after 60 minutes, the professional answer on Tuesday at 10 am, i.e. after 24 calendar hours.
EnquiryC arrives on Monday at 11 am. A person reacts after 15 minutes. At the evaluation cut-off on Tuesday at 12 pm, however, the substantive question is still open, having been open for 25 calendar hours. EnquiryD is received on Monday at 12 pm; the first reaction takes 45 minutes, the professional answer is available after four hours.
For the first human reaction the times are 30, 60, 15 and 45 minutes. Their mean is 37,5 minutes. This number still says nothing about when the respective substantive questions were fully answered.
For the three cases already professionally answered there are one, 24 and four calendar hours. The average is 29 divided by three, i.e. nine hours and 40 minutes. This is explicitly the mean of the three completed cases, not of all four enquiries. The open case C must neither be entered as zero nor invisibly removed from a claimed overall statement.
The open case belongs in the evaluation
State alongside the metric how many transactions were started, how many reached the defined end point and how many are still open. For open transactions the time elapsed so far up to a clear cut-off can be helpful. It is not yet a final response time.
In the example this presentation prevents a premature conclusion. The first reaction was relatively quick in all four cases; the professional processing of one enquiry is not yet completed at the cut-off. The team does not have to select one number that conceals these differences.
Also distinguish the type of enquiry where it is actually relevant to your own learning question. A simple appointment follow-up and a query about documents that first needs professional clarification are different tasks. This distinction should make processes more understandable, not remove uncomfortable cases from the evaluation without a substantive rule.
Do not mix calendar time and your own working time
Another decision concerns the time base. In the model the office works Monday to Friday from 9 am to 5 pm. For EnquiryB, from Monday 10 am to Tuesday 10 am is 24 calendar hours, but eight hours within this assumed working window: seven on Monday and one on Tuesday.
This calculation is not a general rule about reasonable response times. It only shows why a metric must state its basis. Actual opening hours, on-call arrangements, public holidays and agreed procedures must be explicitly defined for a real evaluation.
Use the same definition in a comparison. Comparing an earlier calendar time with a later working-window time would create an apparent improvement even with an identical process. A changed measurement definition therefore belongs in the evaluation note.
Use the result to improve a process
Office management could, in the example, investigate why EnquiryC remains open: is a particular document missing, is responsibility unclear or is a professional follow-up awaited? The metric itself does not answer this question of causes. It highlights the case that should be examined specifically.
An appropriate change can be a clear handover point or a more comprehensible interim feedback. Its effect is then observed later using comparable definitions. An automatic message will not afterwards be relabelled as a human or professional answer.
Whether and how an internal observation may become a public service statement is an additional check. This workshop does not provide a free pass for that. Its own benefit is an honest working status: which reaction was meant, what is actually completed and which uncertainty remains visible?