Every hotel tech company has one: the channel that costs money every month and never seems to produce a deal anyone can point to. A trade show booth, a LinkedIn ads budget, a content programme, a reseller agreement. Someone in the leadership meeting asks whether it is worth keeping, nobody can prove it either way, and the decision gets pushed to next quarter.

This self-assessment is built to end that loop. It gives you ten scored questions, three score bands and a clear action for each band. It works for any channel, paid or unpaid, and takes about 20 minutes if your CRM is in reasonable shape. If it is not, that is your first finding.

Why this decision is harder in B2B than it looks

The instinct is to look at last-touch leads from the channel over the past quarter and cut it if the number is low. In hotel tech that test is almost always too short. Dreamdata's 2025 LinkedIn Ads Benchmarks Report found that the typical B2B customer journey runs 211 days and 76 touches, involves 6.8 buyer stakeholders and spreads across an average of 3.7 channels. On LinkedIn specifically, the report measured an average of 320 days from first ad impression to revenue. A channel judged on 90 days of data is being judged on less than half of one buying cycle.

Buyers also decide earlier than your CRM can see. 6sense's 2025 Buyer Experience Report, based on responses from more than 4,000 buyers, found that 94% of buying groups ranked preferred vendors before first contact, and that the vendor buyers prefer before engaging with sellers still wins 80% of deals. Channels that work at that stage, such as content, events and peer conversations, rarely show up as the source of an opportunity. Hotel buying adds seasonality on top, which we covered in how seasons shape hotel tech sales cycles.

The opposite error is just as real. Some channels look busy in the reports and do nothing. According to The Hustle, Uber turned off two-thirds of a roughly $150m ad budget and saw no negative effects, after finding ad networks claiming credit for sign-ups that would have happened anyway. Airbnb told investors something similar in 2021: Marketing Week reported it cut sales and marketing spend by 28% year on year in the first quarter, mainly performance marketing, with 90% of traffic unpaid or direct. And yet Skift's review of Airbnb's SEC filings shows search marketing growing again from 2022, with brand and performance marketing up 32% in the first half of 2026 against revenue growth of 17%. Cutting a channel is a reversible decision only if you measure what happens after.

How to score yourself

Pick one channel. Answer each question with 0, 1 or 2 points. Be strict: if you have to guess, score 0. Use the last 12 months of data, not the last quarter, so you cover at least one full hotel season.

Part A: Is it actually not working? (questions 1–4)

1. Do you have a written goal for this channel? 0 = no goal, 1 = a vague goal ("awareness"), 2 = a number and a stage it is meant to influence, such as 15 first meetings with management companies per year.

2. Have you given it at least one full buying cycle? 0 = under 6 months, 1 = 6–9 months, 2 = 9 months or more, or longer than your median time from first touch to closed-won.

3. Do you track it beyond last touch? 0 = only lead source, 1 = first and last touch, 2 = you can see every opportunity where the channel appeared anywhere in the journey, including contacts from the same hotel group.

4. Have you asked closed-won customers how they heard of you? 0 = never, 1 = sometimes, 2 = a self-reported attribution field on every won deal, checked against the CRM.

Part B: Was it set up to work? (questions 5–7)

1. Does the channel reach your ICP? 0 = you do not know who sees it, 1 = partly (for example, a show full of vendors and consultants), 2 = you have evidence your target segment is there, such as attendee lists or audience data.

2. Was it executed properly? 0 = run part-time with no owner, 1 = an owner but inconsistent effort, 2 = a named owner, a consistent cadence and follow-up within 48 hours.

3. Does the offer match the stage? 0 = demo requests asked of cold audiences, 1 = mixed, 2 = the call to action fits where buyers are when they meet the channel.

Part C: What would it cost to stop? (questions 8–10)

1. Do you know its fully loaded cost? 0 = media or fees only, 1 = fees plus some time, 2 = fees, travel, team hours and tools, per opportunity influenced.

2. Can you test the cut before making it? 0 = all or nothing, 1 = you can reduce it, 2 = you can pause it in one region or segment and compare against the rest.

3. Would anyone notice if it disappeared? 0 = nobody has asked about it in a year, 1 = sales would complain but cannot name deals, 2 = customers, partners or reps can name specific deals it helped.

Score bands: what your total means

A low score on Part A is the most common result. It is also the most useful, because it means the debate in the leadership meeting was never about the channel. It was about measurement.

How to cut a channel without guessing

If you land in the 14–20 band, or you are forced to cut for budget reasons, do it in a way you can learn from.

• Run a holdout. Pause the channel in one region or segment for one quarter and keep it running elsewhere. Compare first meetings and opportunities created, not leads.

• Write down what you expect to happen. If you believe the channel does nothing, you expect no change. Put a number on it before the test so nobody rewrites the result afterwards.

• Watch the slow metrics. Branded search, direct traffic, inbound demo requests and "heard of you from" answers move months after a cut. Check them at 3 and 6 months.

• Tag the losses. If deals start slipping to a competitor that is still at the show you left, your closed-lost reasons will say so, provided the tags are good. Our guide to closed-lost tagging for hotel tech shows how to set them up.

• Keep the relationships. Cutting a booth does not mean cutting the event. You can still book meetings around it at a fraction of the cost, as we described in what hotel tech trade shows really deliver.

Three patterns we would not cut on score alone

Some channels score low because of what they are for, not because they fail. A reference customer programme, a partner agreement with a PMS that has your target hotels, and founder content all work early in the journey that 6sense describes, where 94% of buying groups have a shortlist before they talk to you. For these, score questions 2 and 10 twice as carefully. If buyers and partners name them, keep them even if the CRM cannot.

The reverse also holds. A channel with plenty of leads, all from the wrong segment, can score well on activity and still deserve the cut. Uber's case is the reminder: reported numbers can be confident and wrong.

Get a second opinion on the channel

If your score sits in the 0–7 band, the problem is usually upstream of marketing: unclear ICP, broken attribution or a handoff that loses leads. Our free revenue leak audit looks at where pipeline is leaking across channels, so you can decide what to cut with numbers instead of opinions.

Frequently Asked Questions