A “How did you hear about us?” survey asks customers which source introduced them to your business. Use it to learn about recommendations, events, content, and other routes customers remember. Ask about one moment—first discovery, the latest visit, or the decision to buy—so you know what the answers mean.
Youform lets you collect a source question, show relevant follow-ups, and send responses to Google Sheets. Start with the short survey below, then use the worked report to turn answers into a useful next action.
Start collecting: Draft your survey with Youform AI, or create a free form and copy the question below.
Useful extras: Download the editable survey (TXT), get the blank coding worksheet (CSV), or see the completed report. The downloads are free and need no signup.
A copy-ready “How did you hear about us?” survey
For customers making their first purchase, use this introduction and one optional single-choice question. Replace the business name and channels with your own.
Thanks for choosing [business]. One optional question helps us understand where new customers first find us. Your answer will not affect your order.
Where did you first hear about [business]?
- A friend, family member, or colleague
- A search engine
- YouTube
- An event or market
- An article or review
- Something else — let the person describe it
- I don’t remember
Ask a follow-up only when it adds useful detail. A search answer can lead to “Which search engine?”; an event answer to “Which event?” Keep these optional. You do not need the name of the person who made a recommendation unless there is a separate, explained reason to collect it.
Tailor the list to your customers: podcasts, LinkedIn, an AI assistant, or a shop window may be more relevant than the channels above. For a service enquiry, change “order” in the introduction to “enquiry” and ask after the person has finished their main task.
The survey download includes the question, options, follow-up rules, stable reporting codes, and a before-sharing checklist.
Ask your AI for options that fit your industry
A restaurant, a software company, and a local tradesperson will have different discovery sources. Use this prompt in ChatGPT, Claude, or your preferred AI to make a short list for your business. Replace the brackets, then check the suggestions against real customer answers.
Edit your industry prompt
Or open the prompt in your AI:
Your AI’s sign-in and prompt support may vary; use Copy industry prompt if needed. Once you have a useful list, draft it in Youform or add it as a Single Select question in your free form.
Choose wording that answers your actual question
| What you need to learn | Question to ask | What the answer describes |
|---|---|---|
| First discovery | Where did you first hear about [business]? | The source the customer remembers first. |
| This visit | What brought you to our website today? | The reason or route for the current visit. |
| Purchase motivation | What most influenced your decision to buy today? | A buying influence, which may be a price, product feature, or recommendation. |
Use one of these when one decision is enough. “How did you hear about us and why did you buy?” combines two jobs and produces an answer you cannot reliably assign to either. Ask them separately if both are needed.
A customer can first discover you through a friend, return through Google, and buy after reading reviews. The discovery question captures the introduction; the click report captures the visit. Keep both to understand the journey.
Build answer options from the words customers use
Begin with an open-text pilot if you do not know the likely sources. Review those answers, group repeated channels, and turn the common ones into a short choice list. Pew Research Center explains how pilots, wording, answer choices, and option order influence survey answers.
- Use recognizable names. “Instagram” or “a friend” is easier to answer than “paid social” or “referral acquisition.” A person may not know whether a post was sponsored.
- Make choices distinct. Avoid a list containing both “social media” and “Instagram” without explaining the difference.
- Include unpaid discovery. A list containing only channels you advertise on can miss recommendations, organic content, and offline discovery.
- Keep Other and uncertainty available. “I don’t remember” is different from “Something else,” and both are different from skipping the question.
- Keep the meaning stable. Use the same reporting code when you shorten a label. Record when the question or options change, so a new list is not mistaken for a change in customer behavior.
Single choice fits “where did you first hear.” If you ask people to select every source they remember, you are measuring multiple mentions: the totals can exceed the number of respondents and the percentages need a different explanation.
Ask after the customer’s main task
For a store, place the question after checkout on a confirmation page that supports your form, or link to it from a post-purchase email. For an agency or service business, ask after the enquiry or booking. Keep the question optional and avoid making payment or a useful reply depend on answering it.
Choose whether the survey is for first-time customers or everyone. For first discovery, invite first-time buyers from your existing customer or order system. Report first-time and repeat customers separately.
Your checkout platform controls where forms can appear. A Youform link can work from an email even when a checkout page cannot accept a general embed. For Shopify Thank you and Order status pages, check Shopify’s current customization options to choose a supported placement.
Record the invitation dates, audience, channel, and number of people invited. Use a known invitation count when calculating a response rate.
Turn raw answers into a report without guessing
Northline Prints is a fictional small print shop. It invited 40 first-time buyers to answer an optional discovery question; 20 submitted the survey. The complete 20-row example (CSV) shows each original answer, its reporting bucket, the coding decision, and a separate observed first-visit source from an illustrative analytics record.
Open the CSV in Google Sheets or Excel. Keep the original-answer column, add your bucket and note beside it, and count the buckets. Keep observed visits in their own columns alongside the customer’s answer.
| Original answer | Reporting bucket | Coding decision |
|---|---|---|
| A friend sent me your name. | Word of mouth | Keep the remembered recommendation even if analytics records Google. |
| An Instagram ad. | Preserve “ad” in the original answer; review campaign records separately for paid performance. | |
| Google—I searched your name. | Search | Code the reported source. Branded search may follow an earlier influence the customer did not mention. |
| A flyer at the library. | Other | Keep the detail; consider a flyer option in the next version if it recurs. |
| I’ve known the name for ages; not sure where from. | Don’t remember / unclear | Keep the source marked unclear. |
| No answer | Skipped | Keep missing answers separate from explicit uncertainty. |
Use the same rule for every row. Map spelling variants such as “IG” and “Instagram” to one code. Preserve uncategorizable answers for review. If someone names several sources and cannot identify the first, mark the first source as unclear.
For the 20-row example, =COUNTIF(C2:C21,"word_of_mouth") counts four recommendations. =COUNTIF(C2:C21,"word_of_mouth")/COUNTA(A2:A21), formatted as a percentage, returns 20%. Column A contains one ID for each submitted survey; column C contains the source code. Adjust both ranges for your own rows and include submissions with skipped source answers.
Download the blank coding worksheet (CSV) to apply the same structure to your own responses. Keep access appropriate to any customer information you include.
A completed source report you can adapt
Northline Prints — fictional first-customer discovery report. Audience: first-time buyers invited by email. Invitations: 40. Submitted surveys: 20. Source answers: 18, including two uncertain answers; two people skipped the source question.
Invitation-response rate: 20 ÷ 40 × 100 = 50%. The source shares below use all 20 submitted surveys, so uncertainty and skipped answers stay visible.
| Remembered source | Count | Share of 20 submitted surveys |
|---|---|---|
| Word of mouth | 4 | 20% |
| 4 | 20% | |
| Search | 3 | 15% |
| YouTube | 2 | 10% |
| Event or market | 1 | 5% |
| Article or review | 1 | 5% |
| Other | 1 | 5% |
| Don’t remember / unclear | 2 | 10% |
| Skipped | 2 | 10% |
| Total | 20 | 100% |
What the owner can do next
- Preserve the useful signals: recommendations and Instagram are commonly named among these respondents. Review those customer journeys and the messages people saw.
- Investigate one mismatch: row R01 reports a friend while the observed first visit is Google organic. The original introduction may have happened offline; check the question and collection context rather than rewriting it as Google discovery.
- Keep a question open: four of 20 submissions are uncertain or blank. Improve the wording or pilot the options if that pattern persists. Keep them visible in the next report.
- Assign an action: the marketing owner reviews Other answers and the discovery options before the next monthly survey. Keep the existing list until a revision is agreed, and record that revision.
These shares describe the respondents’ remembered sources; the 20 nonrespondents may have different journeys. Use this report to choose what to investigate, then bring in campaign and revenue data for budget decisions.
Download the completed report and blank report template (TXT). Keep the period, audience, denominator, uncertainty, owner, and next action when you adapt it.
Build the survey and review answers in Youform
- Start a form. Add the introduction and a Single Select question with your source options. Leave the source question optional; one answer should represent first discovery.
- Allow Other. Youform’s choice-question settings provide an Other option with a text answer and an option to randomize choices while keeping Other at the bottom.
- Add relevant follow-ups. Configure conditional logic for a useful detail, such as the event name. Send people who do not remember to the ending instead of requiring a guess.
- Choose the distribution route. Publish a direct link or use a supported website embed. Your own email or customer system determines who receives the invitation and when.
- Connect the response destination. In Integrate, authorize the appropriate Google account and connect Google Sheets. Use the resulting response columns as your raw inputs; the downloadable coding/report sheets are a separate analysis structure you adapt.
- Check the paths before inviting customers. Review a normal choice, an optional follow-up, Other, “I don’t remember,” and a skipped question on desktop and phone. Confirm the published link, labels, and destination you will actually use.
If you pass campaign context through hidden fields, keep it separate from the customer’s answer. A UTM on the survey link tracks the invitation. To compare original acquisition data, use your existing order or analytics process and keep that data in separate columns.
Youform Free includes conditional logic, website embeds, Google Sheets, and unlimited forms and responses within fair usage. Your logo, custom domains, partial submissions, and advanced drop-off analytics require Pro. Check the current plans; connected tools may have their own costs.
Create your source survey in Youform. For a broader decision-journey questionnaire, explore the path-to-purchase template; for satisfaction rather than discovery, use the customer-feedback question guide.
Common questions
Should “How did you hear about us?” be required?
Keep it optional when it is extra marketing research rather than information needed to deliver the customer’s request. Include a way to say the person does not remember, and do not make checkout depend on the answer.
Should I use a dropdown or single-choice list?
A visible single-choice list makes a short set easier to scan. Use a dropdown for a longer list when it helps the layout, and check it on a phone. Reduce or group irrelevant options before adding a long menu.
Can the survey replace analytics?
No. It records what respondents remember. Analytics records observed visits and events under its own collection and attribution rules. Keep the two signals separate and investigate differences.
What if several sources influenced the customer?
For first discovery, ask for the earliest source they remember and allow uncertainty. To study all influences, ask a separate multiple-choice question and report mentions; do not treat its percentages as mutually exclusive discovery shares.
Reviewed . Question-writing reference: Pew Research Center. Product setup references are linked beside the instructions. The Northline Prints example and worksheets are original fictional examples.