What is snowball sampling?

Also called: Chain-referral sampling, Referral sampling

Snowball sampling is a non-probability recruiting method where each participant refers others who fit the study, so the sample grows through referrals. It is used when the people you need are hard to find through a customer list or panel.

Updated

When product teams use it

Snowball sampling earns its place when the right participants are rare or invisible in your data. Typical cases:

  • B2B roles that rarely log in, such as the procurement lead or security reviewer who signed off on the purchase.
  • Prospects and non-customers in a niche field, for example hospital pharmacists or customs brokers, who aren't on general panels in useful numbers.
  • People who stopped using a category entirely, who no longer appear in your product data.
  • Internal stakeholders at a customer account: one champion introduces you to the analyst and the executive sponsor.

How to run it

  • Start with seeds: two to four participants who fit and who differ from each other, so the referral chains start in different places.
  • At the end of each session, ask a specific question: "Who else on your team approves new tools?" works better than "Know anyone else?"
  • Screen every referral with the same criteria as the seeds. A referral is a lead, not a qualified participant.
  • Track who referred whom, so you can see when one chain is taking over the sample.
  • Offer the same incentive to referred participants, and decide in advance whether referrers get anything extra.

The bias to watch for

People refer people like themselves: same company, same seniority, same opinions. Five interviews from one referral chain can feel like agreement when it's one network repeating itself. The fix is more seeds from different starting points, a cap on referrals per chain, and a note in the readout on how the sample was built.

Referrals also carry social pressure. A participant introduced by their manager may be less willing to criticise a tool the manager chose, so make it clear that sessions are confidential.

Go deeper

Questions

Is snowball sampling a type of purposive sampling?

It's usually treated as a related non-probability method. The seeds are chosen purposively, then referrals grow the sample. The difference is control: with purposive sampling you pick everyone, with snowball sampling participants pick the next people for you, which is why screening referrals matters.

Can snowball sampling be used for surveys?

Yes, by asking respondents to forward the survey to peers who fit. It helps reach a niche audience but makes the sample impossible to describe precisely, because you don't know who received the link. Treat the results as directional, not as a measure of the whole group.

Next step

Run the study, not just the method

Great Question handles recruiting from your own customers, scheduling, incentives, interviews, tests and analysis in one place.

Book a demo

No credit card required.

Related terms

Full glossary
  • Purposive sampling

    Purposive sampling is a non-probability sampling method where the researcher deliberately picks participants because they have characteristics or experiences the study needs, such as admins who just migrated from a competitor, instead of recruiting at random.

  • Sampling bias

    Sampling bias happens when the way you choose participants makes some kinds of people more likely to end up in your study than others, so the sample stops representing the users or customers you want to draw conclusions about.

  • Convenience sampling

    Convenience sampling is a non-probability sampling method where you recruit whoever is easiest to reach, such as customers who answer your first email or colleagues down the hall, instead of choosing participants to represent a defined population.