# How do you think about channel strategy? How do you decide where to invest your GTM efforts?
> How to answer the channel strategy question, with how to read the buyer’s path and channel economics, a worked example, and the rule for cutting a channel.

Source: https://mokk.ai/pmm-questions/gtm-strategy/channel-strategy-investment (updated 2026-09-22)
Markdown: https://mokk.ai/pmm-questions/gtm-strategy/channel-strategy-investment.md

Start from how buyers find and choose products like yours, then compare channels on reach into that buyer, the cost of a customer who stays, speed to results, and fit with your team. Fund two or three properly, run small fixed-budget tests on the rest, and move money on payback and retention by channel. State the rule that would make you cut a channel before you need to use it.

## What a channel answer needs beyond a list

Everyone can name channels. The question asks how you decide with evidence that is always incomplete.

The gap in most answers is the reasoning behind the pick. Where did the data come from? Why trust it? What would change your mind? A short chain of evidence behind two or three choices scores better than a tour of every channel you’ve used.

## Reading the evidence on where customers come from

Each source of channel data is biased in a known direction. Tracked, last-click attribution overcredits search and retargeting, because those sit at the end of the journey. Asking customers how they heard about you overcredits memorable moments like a friend’s recommendation. Where the two disagree, you learn something about how demand forms.

Retention by channel matters as much as acquisition cost. A cheap customer who leaves in a month costs more than an expensive one who stays for years. Look at both before ranking anything.

## The channel reasoning, spoken aloud

**Strong answer (invented example)** "Two sources tell me how customers found us, and I’d use both. The product here is a credit-builder card for people in their twenties. Our tracking says most signups come from app store search and paid social. When we ask during onboarding, though, a big share say a friend or a creator they follow told them about it. Both can be true. People hear about us from a creator and search for us later, and search gets the credit. So before I cut the creator budget for looking expensive, I’d check retention by channel. If referred users keep the card active at twice the rate of paid social users, paid social is the expensive channel, even with the cheaper signup. I’d put most of the budget into creators and a referral bonus, and keep app store search funded because it captures the demand creators start. Paid social drops to a fixed test, five thousand dollars over six weeks, and it passes only if the cost per card still active at ninety days comes in under our limit. I’d review every quarter and hold a small reserve for new tests."

## Comparing channels on the numbers that matter

Channel measures, what each tells you, and where each misleads
| Measure | What it tells you | Where it misleads |
| --- | --- | --- |
| Cost per signup | Efficiency at the top of the funnel | Ignores quality and retention |
| Cost per retained customer | What you pay for someone who stays | Needs time, and early cohorts are small |
| Payback period | How fast spend comes back | Shifts whenever pricing changes |
| Self-reported source | Channels people remember | Undercounts search and ads people ignored |
| Last-click attribution | The final step before signup | Overcredits search and retargeting |

## Conclusions the channel data won’t support

- **Cutting a channel for weak last-click numbers** when it creates the demand that search later captures.
- **Scaling a winning test in a straight line.** Channels saturate, and the second round of spend rarely performs like the first.
- **Comparing channels on signups** when their customers retain at very different rates.
- **Treating last year’s winner as proven.** Costs rise as competitors find the same channel.

On the [rubric](https://mokk.ai/pmm-rubric), Strong means you separate correlation from cause without being asked. Exceptional means you go looking for evidence against your own pick. Pausing a channel in one region for a month and watching what happens to signups there is the cleanest version.

## Follow-ups on channel investment

How interviewers test a channel answer
| Follow-up | Why it comes up | Response |
| --- | --- | --- |
| Buyers touch several channels. How do you attribute? | Attribution maturity | Combine tracked and self-reported data, and use holdout tests for big bets |
| The founder loves a channel you’d cut. | Influence with data | Show retention and payback by channel, then propose a fixed test |
| How do you test on a small budget? | Discipline | A fixed budget, a fixed time, and a pass mark set beforehand |
| When do you stop funding a channel that used to work? | Honesty about sunk cost | When payback misses the limit two quarters running |

## More questions in this category

- [Late-changing launch scope](https://mokk.ai/pmm-questions/gtm-strategy/features-unfinished-before-launch)
- [Explaining technical products](https://mokk.ai/pmm-questions/gtm-strategy/technical-product-non-technical-buyers)
- [Integrating an acquired product](https://mokk.ai/pmm-questions/gtm-strategy/integrate-acquired-product)
- [Self-serve and enterprise motions](https://mokk.ai/pmm-questions/gtm-strategy/self-serve-and-enterprise-motions)
- [All go-to-market strategy questions](https://mokk.ai/pmm-questions/gtm-strategy)

## Frequently asked questions

**Which channels should my answer mention?** Only the ones that fit the buyer in your example. Naming every channel suggests you don’t know how to choose, and it uses up the time you need for the reasoning.

**Do I need exact acquisition cost figures?** No. Show which numbers you would compare and why, and use invented figures inside an example if that helps. The logic of weighing cost against retention is what gets scored.

## Practice

**Answer it out loud.** A Mokk session asks this question, follows up on your answer, and scores it on five competencies.