Question-First Cheap Channel Playbook
Find where your buyers ask their questions, answer them honestly, and let the answer sell.
- Difficulty
- Easy
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 95%
Berkowitz's method for near-zero-cost acquisition: B2B products exist because someone has an active problem, and people with problems ask questions in public. Map every venue where those questions get asked (Reddit, Quora, Twitter, closed communities), set keyword alerts, and have real humans answer the question completely — with your solution offered only after the answer is given. Because the answers persist, each one compounds.
Origin
Berkowitz was hired at Deel specifically because she specializes in cheap growth channels. She set up Reddit keyword tracking within her first day at the company; Deel's founders still answer questions in these channels personally.
Core principles
- 01People do not want to be sold to — they want their problem solved.
- 02A cheap channel is any place where someone is asking a question you can answer.
- 03The answer is a durable digital asset: you answer once, and it keeps being read.
- 04Value must be delivered before the pitch, or the community ejects you.
- 05Never automate the interaction — Deel has never automated a single one.
- 06Every channel has an audience ceiling; size it before you invest.
How to run it
- 1
Inventory the question venues
List every place your buyer asks peers for answers: subreddits for their role (founders, HR managers), Quora, Twitter, industry Slack/Discord groups, and closed communities (e.g. Y Combinator alumni networks) you may need a partnership to enter.
Pro tip Reddit is where people go for semi-anonymous peer answers to questions too specific to ask Google — that specificity is exactly your qualification signal.
Watch out Closed Slack communities and DM threads do not compound — your answer is invisible to everyone who arrives later. Weight public, indexable venues higher.
- 2
Size the audience ceiling before investing
For each venue, ask what the realistic upper limit is. A thousand-member subreddit will not deliver 5,000 businesses — it might deliver ten people who have this exact problem. Decide if that is worth the time, knowing it stacks with word of mouth.
Watch out Going in without sizing the ceiling is how teams either overinvest in a dead forum or dismiss a small-but-perfect one.
- 3
Wire up keyword alerts
Set alerts on the keywords that signal your problem (e.g. hiring internationally, contractor compliance). Berkowitz used a rough off-the-shelf assistant tool that someone else had built and simply plugged her keywords in.
Pro tip The tooling can be janky. She describes hers as 'a very janky assistant' she is 'not proud of' — but it worked from day one.
- 4
Answer like a friend, not a vendor
Give the actual answer first: yes it's doable, no it isn't, or yes but you need to consider X, Y, Z. Explain the solution to the problem even though competitors could also solve it. Only then mention that they can chat with you to learn more, and let them decide.
Pro tip Treat the person like a friend of yours who asked. The low-value version — 'we've solved that problem, check out our website' — is what gets you ignored or removed.
Watch out Never automate it. Deel routes these to real team members, including co-founders, to this day.
- 5
Give the whole team access and self-throttle
Let anyone on the team jump into a thread; whoever gets there first flags it. When enough people from your company have added value in one thread, stop — a pile-on reads as a marketing swarm, not help.
In the wild
On roughly her first day, Berkowitz set up keyword tracking across Reddit for questions about international hiring, contractor payments and compliance. Team members — including co-founders — jumped into threads and answered the question in full before mentioning Deel.
→ These channels ran at roughly 15% of Deel's funnel early on. As other channels scaled, they fell to under 5% of the funnel by percentage — but the absolute number of people reached through them kept growing, and Deel still runs it today.
At Berkowitz's previous company, when COVID hit and the US government rolled out the PPP program, customers were asking whether they qualified and how to apply. The team took the time to fully understand the program and wrote a definitive resource — purely to help their own customers.
→ The IRS website itself linked to the article as the answer to the question. It became a major growth driver for the few weeks people were applying for PPP, then died out with the program. The author later joined Deel.
Common mistakes
Dropping a link and calling it participation
'Hey, we've solved that problem, check out our website' adds nothing. The community either ignores you or kicks you out — you cannot piggyback on someone else's community without paying rent in value.
Dismissing the channel as too small to matter
People skip it because it looks like one person here, two people there. But each answer is permanent, publicly readable, and compounds with word of mouth.
Automating the replies
Automated answers destroy the exact trust the channel runs on. Deel has never automated it and says it never will.
Is it for you?
Best for
Early-stage B2B founders and first growth hires with little or no budget whose product solves an urgent, specific, frequently-asked problem.
Not ideal for
Categories where buyers do not publicly ask for help — impulse consumer goods, or anything discovered by browsing rather than problem-solving.
From the transcript
“just add value to people answer their questions”
“yes exactly that's what we did it's important to notice that you know if you're looking at a subreddit of about a thousand people you…”
“so like genuinely treat this person like a friend of yours answer their question be like yes it's doable no it's not doable”
“it was never automated it still isn't today we would never automate our interactions with people”
From the episode
An inside look at Deel’s unprecedented growth
Meltem Kuran Berkowitz (Head of Growth)