There is a specific feeling that sends people looking at scheduling software. The week is full, the pipeline is not, and something has to give.
The shopping starts before the diagnosis, and that is the mistake. A scheduler with qualification in it removes meetings. If your meetings are not the problem, removing some of them is not neutral. It is harmful, and you find out three months into a subscription.
So do the diagnosis first. An hour, last month's calendar, a spreadsheet. It gives one of three answers, and one of them is buy nothing at all, which is a genuine result rather than a rhetorical device.
Counting meetings measures the wrong thing
The number everybody reaches for is the count. Thirty-one external meetings last month. It feels like evidence and it is not, because a count has no direction. Thirty-one that each produced a dated next step is an excellent month. Thirty-one that produced four is a different business with the same calendar.
A full calendar is not the complaint. A full calendar with nothing moving is, which is why the busiest people you know are not the most frustrated ones.
The measurement that carries information is a ratio: meetings that went somewhere, over meetings taken. Almost nobody has it, for structural reasons rather than lazy ones. A calendar records that a meeting existed. A CRM records deals. Neither records the meeting that led nowhere, because nothing in the toolchain has a reason to write that down.
The audit, in about an hour
An export and a spreadsheet. No trial, no integration, nothing installed. Longer than an hour means you are being too thorough, and thoroughness is how this ends up abandoned in a tab.
- Export last month from your calendar: subject, start time, duration, attendee list. Takeout for Google, the file export for Outlook. One month, not one quarter, because a quarter never gets finished.
- Delete every internal meeting and every recurring one. Standups and weeklies are a real cost, and not something a scheduling tool has any view on.
- What is left is your external meetings, and there are fewer of them than you expected. One row each.
- Who asked for it. Inbound booking, referral, your own outbound, or a partner. Pick the nearest.
- Did it produce a next step. A proposal sent, a second meeting in the diary, a trial started. Somebody promising to think about it does not count, and being generous here defeats the exercise.
- Knowing what you know now, would you have taken it? Yes or no, in under five seconds a row, because the fast answer is the honest one.
- The length in minutes, and how many people were on your side. Then total the no column, and do not soften the third question afterwards.
What it costs, in hours
Take a two-person team with forty external meetings last month, nine of them marked no.
Six of the nine were thirty-minute calls with both people on them: 6 × 30 × 2 = 360 minutes. The other three had one person: 3 × 30 = 90. Meeting time alone is seven and a half hours.
Then add what surrounds a meeting. Fifteen minutes of prep and recovery per person per call: 6 × 2 × 15 = 180, plus 3 × 15 = 45. Another 225 minutes, so three and three quarter hours.
Eleven and a quarter hours in a month, on meetings this team has just said it would not repeat. Held flat, 135 hours a year. The ratio is the part to keep: nine out of forty, a shade under a quarter.
The same caution applies here as in the per-seat pricing post: the figures are only as good as the inputs. These came out of your own spreadsheet five minutes ago, which makes them worth more than any statistic on a vendor's landing page, this one included.
Three answers, and what each one means
Under about one in ten: this is not your problem
Three meetings out of forty is a normal error rate, not a filtering failure. Your judgement about who to meet is already working. What you feel is a volume problem wearing a calendar problem's clothes: too few of the right conversations, which makes the wrong ones conspicuous.
Friction makes that worse, predictably. A form in front of your calendar removes some proportion of bookings, and when nine in ten are good, most of what it removes is good. This is the case where our own product is the wrong purchase, and it deserves the plainest sentence available: do not buy Setupp. Spend the money getting more of the right people to the booking page.
Around a quarter: worth fixing, and a form usually does it
The ordinary case, and the one this category exists for. A quarter is enough hours to matter and enough of a pattern to filter on, and the pattern usually sits in the first column: the bad ones come through the same door.
What fixes it is three questions asked before the calendar renders rather than after the slot is taken. That order is the mechanism, and it is the subject of qualifying leads before they book. Approximate it this week with a free form and a link you share by hand, before paying anyone to enforce it.
Over about a third: the problem is upstream of the calendar
If more than fourteen of the forty were mistakes, something is generating them and it is not your booking page. A booking page is a door. It has no say in who walks up to it.
Look at the first column. If the bad meetings share a source, the source is the finding: an advert promising a free consultation to an audience that cannot buy, or a directory listing ranking for a far broader query than the service you sell. The promise and the traffic are the two dials, and neither lives in a scheduler. A filter on the end of a badly aimed funnel moves the disappointment to the form.
The two things the audit cannot see
The audit is a record of meetings that happened. Both blind spots follow from that.
What happened to the people you turned away
There is no column for the enquiries you declined, the emails you never answered, or the visitors who opened your booking page and closed it. Most teams cannot fill that one in at all, because nothing recorded it.
The gap points one way. The audit shows meetings you should not have taken and can never show one you should have taken and did not, so it flatters whatever filtering you already do. Reading your own spreadsheet and concluding your judgement is sound proves something narrower than it looks.
Start recording it today: one line per enquiry you turn down, with the reason. A name from that list turning up as somebody else's customer is a more interesting finding than anything in the first spreadsheet.
Whether your judgement is any good
The third column is a prediction. Not a feeling: a prediction, made after the fact by the same instinct that will decide next month's bookings. Anything that predicts can be scored against what happened, gut feel included, and most people never score theirs because nobody made them write it down first.
The method is boring. Before each external meeting next month, write one line: worth taking, yes or no, and how confident you are. Afterwards record what happened. Then check whether the confident yeses really did convert better.
That is what the product does to its own scoring, and the mechanism is the same whether the prediction came from a model or a person. Setupp scores a booking out of ten before the meeting; the host later marks it completed or a no-show. calibration.ts then does one sum: attendance among bookings scored seven or higher, minus attendance among the rest. Positive means the score carries information, near zero means it is decorative, negative means it points the wrong way.
The part worth stealing is what that file refuses to do. Below eight recorded outcomes it reports no headline rate at all, and below four in a band it leaves the band blank. Eight is not significant: it is the point where one flipped result moves the number by twelve points rather than thirty-three. A rate built from three meetings is noise wearing a percentage sign.
Hold your own month to that standard. Twelve meetings with predictions attached is a start, not a verdict. If the confident ones did convert better, your instinct is worth writing down as criteria, and that is when a tool becomes worth paying for. If they did not, a scheduler automating it applies the same poor judgement faster.
A score nobody checks is a horoscope. So is a gut feeling, and a gut feeling is harder to check, because nothing writes it down.
What to do with the answer
Three outcomes, three quite different afternoons.
- Under one in ten. Close the tab. The audit did its job by telling you not to buy anything.
- Around a quarter. Start with the order rather than the tool: three questions in front of the calendar, tested with a form you already have. Qualifying leads before they book is that argument in full.
- Over a third. Take the spreadsheet to whoever owns the traffic. A marketing conversation with evidence attached goes a great deal better than one without.
In the middle band and want the shape of it in your own trade? The agency version runs a spend floor through the same arithmetic.
And the honest ending, since you have done an hour of work at our suggestion. We sell software that filters meetings, so publishing a method for finding out you do not need it is odd on purpose. If your audit says your meetings are fine, they are fine. Keep the spreadsheet, run it again in six months, and spend the hour on something better.