Which channels produced teams who signed for a term?

A flexible workspace sells individual memberships at a couple of hundred a month and private suites to teams at thousands a month on a term of well over a year. Both arrive through the same search terms and appear as one row in the signup report.

Ranking channels on signups therefore rewards whichever channel produces the most individuals, when the business is often carried by a small number of team deals. Ranking on contracted revenue reverses that in one step, and a membership-type column mapped as Product / service shows individual and team revenue as separate lines.

Because terms average well over a year, contracted value rather than first month's fee is the honest number. It is also the number the operator's own occupancy and revenue forecasts run on.

Flexible office brokers bring qualified teams and take a commission on the signed term. That makes them one of the few channels whose cost per deal is precisely known, and it is usually a large number.

Putting broker-sourced deals on the same report as paid search shows what the operator's own demand generation costs per contracted desk-month against the commission. That comparison can be the most useful output of the whole exercise.

Nobody signs for space they have not stood in. Whatever the campaign, the inquiry form or the chat widget achieved, the decision happens when somebody walks the floor, sees the meeting rooms, judges the noise level and works out whether their team would be comfortable there. The tour booking is therefore the real conversion event in this business, and the gap between tours booked and tours attended is one of the largest and least examined leaks in the whole funnel.

That gap is not evenly distributed across sources. A broker-introduced team arrives on time because the broker has qualified them, arranged the slot and intends to be standing in reception. A search inquiry from somebody comparing four buildings on a Tuesday lunchtime may or may not appear at all, and a promotion offering a free day pass produces bookings that are essentially free options nobody feels obliged to exercise. A channel can therefore look strong on tours booked and weak on tours attended, and the community manager running the diary already knows exactly which ones those are.

Putting the tour booking list in the lead file, tagged with the source of each booking, prices each channel on signed agreements rather than on appointments in a calendar. The operator can then see each channel's tour count beside the agreements it signed and, more usefully, how many tours the front desk has to host to sign one team. That is a staffing question as much as a marketing one and it has never had a number attached to it.

People experience a workspace before they join it, and usually by paying for something small. A visiting team books a meeting room for an afternoon. A consultant takes a day pass while their office is being refitted. A company holds an off-site in the event space. Each of those is a transaction with a name, an email address and a card payment attached, and each of those people has now spent hours inside the building forming an opinion about whether they would like to work there.

Operators treat this revenue as ancillary, which it correctly is in accounting terms and is not in marketing terms. The forty dollars the day pass generated is not interesting to anybody. The fact that the person who bought it signed for six desks nine months later is extremely interesting, and nothing in the workspace platform connects the two records, because one is a casual booking taken at the front desk and the other is a membership agreement raised by the sales team. They are different objects in different parts of the system.

Because the match runs on the individual's contact details and has no time limit attached to it, that connection is entirely recoverable from files the operator already holds. Adding casual bookings to the lead file, tagged as day passes or meeting rooms, ranks them against paid search, broker introductions and referrals on contracted value rather than on the small sums they themselves generated. Because the earliest record is credited, a team whose first contact was a day pass is credited to it, and the meeting room and day pass business may turn out to be one of the better-performing acquisition channels the operator runs.

People choose workspace by commute, and by very little else once the shortlist is drawn. The catchment for a building is measured in walking minutes from a station and in tolerable journey times from where the team actually lives, which means a campaign that fills one location can be almost irrelevant to another two miles down the road. Operators running several sites nevertheless often buy demand centrally, report on it centrally, and allocate it to whichever site the inquiry form happened to mention or whichever one had a slot free that week.

The consequence is that a strong overall result can quietly conceal a site sitting at sixty per cent occupancy that has been subsidized all year by another at ninety-five. Since the economics of a flexible workspace are dominated by occupancy against a fixed lease or a management agreement with a landlord, that concealment is expensive in precisely the way a headline number is designed not to show. By the time it surfaces in the accounts, a full year of demand generation has already been allocated on the wrong basis.

A location column on the agreement export, mapped as Location, shows each site's contracted value and how much of it traces to a channel, using a field the billing system already records. It does not rank channels within each site, but each building is then judged on the revenue its own catchment produced rather than on its share of a portfolio total, and a site whose revenue sits mostly in Direct / Unknown is visibly one the central campaigns are not reaching. The operator can start to see whether an underperforming site has a demand problem, a pricing problem or a product problem, which are three entirely different faults calling for three entirely different responses. A blended report cannot tell them apart and never suggests that it should.