What is a go-to-market strategy for a startup?
A go-to-market strategy for a startup is four decisions made in order: which segment you sell to first, how that segment buys (founder-led, sales-led or product-led), what you charge and per what unit, and which single channel you test first. Each decision narrows the next, so making them out of order is how teams end up with a free plan nobody upgrades or a sales process for a $20 product.
| Decision | What you are choosing | The fact that decides it | Revisit when |
|---|---|---|---|
| 1. First segment | One group of companies with the same problem, trigger and buyer | Who already reacts most strongly to the problem or product | The segment is saturated, or a different group keeps buying anyway |
| 2. Motion | Founder-led, sales-led or product-led | Whether one person can try and buy alone, and how many people must approve | Deals start needing security reviews or procurement |
| 3. Pricing | The unit you charge for and the first price | What the buyer counts as value (seats, usage, outcomes) and what the motion costs you | Customers keep asking for a different unit |
| 4. First channel | One way of reaching the segment | Where that segment already looks for help, and what the motion can afford | After a fixed test of a few weeks with a written target |
If you want the one-page document to write these decisions into, use our go-to-market strategy template with a filled-in example. This post is about making the four calls well.
Which customer segment should a startup sell to first?
Sell first to the narrowest group that feels the problem most, then widen. Paul Graham, co-founder of Y Combinator, calls this keeping a fire contained: in Do Things That Don't Scale (July 2013) Graham describes how Facebook was first only for Harvard students, a market of a few thousand people, and argues it is always worth asking whether there is a subset of the market where you can get a critical mass of users quickly.
Data beats opinion here. Rahul Vohra, founder and CEO of Superhuman, wrote in First Round Review (12 November 2018) that only 22% of surveyed users said they would be very disappointed without the product in summer 2017. Narrowing to the personas in that group (founders, managers, executives and business development) took the score to 33%, and within three quarters of product work it reached 58%. Our guide to running the very-disappointed survey covers the method.
To pick a first segment this week:
- List every customer, trial user or warm conversation you have. Even 10 is enough to start.
- Tag each with industry, company size, the role of the person who cared, and what happened just before they reached out (a hire, a new tool, a failed project).
- Circle the group where the same trigger and the same buyer repeat most. That group is your first segment, written as one sentence. For example: operations leads at 20 to 100 person logistics firms that just added a second warehouse.
- Write down who you are ignoring for now. Saying no is half of the decision.
Sales-led, product-led or founder-led: which should you choose?
Choose by how your buyer buys. If one person can sign up, get value alone and pay with a card, product-led can work. If several people must approve, or the price only makes sense with a contract, you need sales. Before either is proven, the founder sells, because the founder is the only one who can change the product, the pitch and the price after each conversation.
| Motion | Fits when | Warning sign |
|---|---|---|
| Founder-led | Roughly your first 20 customers (our rule of thumb); you are still learning what to say and charge | The founder has no hours left for product |
| Sales-led | Several approvers, security or legal review, contract pricing | Deals too small to pay for the time each one takes |
| Product-led | One user can try it alone and see value in the first session; inviting colleagues makes it more useful | Lots of sign-ups, few people coming back |
Many companies blend them over time. Figma's prospectus says larger customers frequently have rigorous procurement processes and require considerable time to evaluate the platform. Elsewhere the same filing says Figma recognized early that it needed a direct sales model to serve larger customers.
Worked example: what each motion needs to reach $10,000 a month. Divide the target by the monthly price per customer:
- At $50 a month: 10,000 / 50 = 200 paying accounts. If (an example assumption) 5% of sign-ups pay, that is 200 / 0.05 = 4,000 sign-ups. Only a product-led funnel produces that many without a sales team.
- At $500 a month: 10,000 / 500 = 20 customers. If 1 in 4 qualified calls closes (again an assumption), that is 80 calls, or 1.6 a week over 50 weeks. A founder can run that.
- At $2,000 a month: 10,000 / 2,000 = 5 customers, 20 qualified calls at the same rate. Few calls, but each deal may need several approvers, so plan for sales-led.
Put your own price and close rate into the same three lines. If the founder-led number fits in your week, start there; our 30-minute daily founder-led sales routine shows how to fit it in.
How should a startup price its first offer?
Price per the unit your buyer already counts value in, and set the first number high enough that your chosen motion pays for itself. The two prospectuses below show the unit choice clearly. Slack's 2019 filing says paid customers typically paid monthly or annually, based on their number of users. Figma's filing says it launched in 2015 and began charging in 2017, and that it launched its Organization plan in 2018.
Both companies still show the split between self-serve and sales on their pricing pages. Figma's pricing page (checked 5 Nov 2026) lists a free Starter plan and prices per seat, with Organization Full seats at $55 a month billed annually and Enterprise Full seats at $90 a month billed annually, and a Contact sales button on both. Slack's pricing page (checked 5 Nov 2026) has Get started buttons for Free, Pro and Business+, and Contact sales for Enterprise+.
Do the arithmetic on a price like that. Ten Organization Full seats at $55 is 10 x 55 = $550 a month, or 550 x 12 = $6,600 a year. A contract that size can pay for a sales conversation; a single $16 Professional Full seat can't.
For a first price with nothing to compare against, use our method for pricing with no benchmark. Then write three things into your GTM plan: the unit (per seat, per account, per usage), the first price, and the price at which you would add a sales step.
Which channel should a startup test first?
Pick the one channel where your first segment already looks for help and that your motion can afford, and run it alone for a fixed test before adding a second. A product-led company at $50 a month needs channels that bring many sign-ups cheaply, such as content, communities and integrations. A founder-led company at $500 to $2,000 a month can afford one-to-one outreach to a short, well-chosen list.
Early on, the channel is often the founder doing it by hand. Graham's essay says Stripe was famous within YC for aggressive early user acquisition: when someone agreed to try it, the Collison brothers would say "Right then, give me your laptop" and set them up on the spot. The same essay says the founders signed early users up for traditional merchant accounts manually, behind the scenes.
Write the test before you start: channel, segment, number of contacts, weeks, and the result that counts as a pass (for example, 50 contacts over three weeks and five booked calls; set your own numbers). If outbound on LinkedIn, X or email is the channel, Siela's Lead Finder suggests roles, industries, company sizes and locations from your website, and scores every lead hot, warm or cold with a one-line reason.
What did Slack, Figma and Stripe actually do?
Slack, Figma and Stripe each made the four decisions differently. The table uses only what the companies' own filings say, or, for Stripe, what Paul Graham wrote in 2013.
| Company | Source and date | First motion | What was added, and when |
|---|---|---|---|
| Slack | S-1 filed 26 April 2019 | Self-service, for free and paid plans; the filing says growth was largely due to word-of-mouth recommendations | A direct sales force and customer success team from 2016, focused on larger organisations |
| Figma | S-1 filed 1 July 2025 | Launched in 2015; adoption grew as designers brought Figma into their organisations; charging began in 2017 | First sales rep hired in 2018, the same year as the Organization plan |
| Stripe | Paul Graham essay, July 2013 | Founders recruiting users one by one and installing it for them | Not covered by the essay |
Two numbers show where each company's money came from once sales was added. Slack had more than 88,000 Paid Customers (organisations with three or more paid users) as of 31 January 2019, and the 575 paying more than $100,000 a year accounted for about 40% of revenue in fiscal 2019. Figma says about 70% of its 2024 revenue came from customers on Organization and Enterprise plans. In both cases the self-serve product found the users and sales served the largest accounts. For how Slack's launch itself was staged, see our breakdown of Slack's preview release.
In what order should you make the decisions?
Make the decisions in the order of the first table, one a week, and write each down before moving on:
Week 1: segment
Tag your last 10 to 30 customers or conversations and write the one-sentence segment, plus who you are ignoring.
Week 2: motion
Answer two questions for that segment: can one person try and buy alone, and how many people approve a purchase? Pick founder-led until you have about 20 customers.
Week 3: price
Choose the unit and a first price, then run the $10,000-a-month arithmetic above with your own close rate.
Weeks 4 to 6: one channel
Run one written channel test with a pass mark. Change one thing at a time: segment, message or channel, never all three.
If the channel test fails, go back one step, not four. Usually the segment was too wide or the price did not fit the motion.
Frequently asked questions
What is a SaaS go-to-market strategy?
It is the plan for which companies a software product is sold to first, how they find, try and buy it, what it costs per seat, account or usage, and which channel brings the first customers. For SaaS the biggest single choice is whether buyers can sign up and pay alone or need a sales conversation.
When should a startup hire its first salesperson?
Once the founder has closed enough deals to know who buys, why and at what price, and can no longer keep up. In their filings, Slack says it added a direct sales force in 2016, two years after its 2014 launch, and Figma says it hired its first sales rep in 2018, three years after launching in 2015.
Is product-led growth right for a B2B startup?
It suits B2B products that one person can start using alone, that show value quickly, and that become more useful as colleagues join. If buying needs a security review, a contract or several approvers, a product-led funnel alone tends to stall there. Slack and Figma both started self-serve and later added sales teams for their largest customers.
What is the difference between a GTM strategy and a marketing plan?
The GTM strategy decides who you sell to first, how they buy, what you charge and which channel you test. A marketing plan sits underneath it and covers the campaigns, content and budget for the chosen channels. Writing the marketing plan first usually means spending on channels before you know who the buyer is.
How do you know if a go-to-market strategy is working?
Set a pass mark before each test, such as a number of booked calls or paying accounts from a fixed number of contacts in a fixed number of weeks. If the test passes, add volume to the same segment and channel. If it fails, change one thing, usually the segment, and run it again.
Sources
- Slack Technologies, Form S-1 (SEC) (filed 26 Apr 2019, checked 5 Nov 2026)
- Figma, Form S-1 (SEC) (filed 1 Jul 2025, checked 5 Nov 2026)
- Paul Graham: Do Things That Don't Scale (July 2013, checked 5 Nov 2026)
- Rahul Vohra, First Round Review: How Superhuman Built an Engine to Find Product Market Fit (12 Nov 2018, checked 5 Nov 2026)
- Figma: Plans and pricing (checked 5 Nov 2026)
- Slack: Pricing (checked 5 Nov 2026)
- Paul Graham: Bio (checked 5 Nov 2026)