A self-reported source captures what the customer remembers and chooses to share about how they found you, not an automated log of every website click or message thread. That memory is useful—it shows which channels people recognize and which referrals matter enough to mention—but it sits alongside automated tracking rather than replacing it. By asking the right question with neutral answer options, you collect one useful signal without forcing a single-channel story or losing customers who genuinely cannot remember.
Why self-reported source is different from last contact channel
The channel where a customer finally messages you is not always the channel where they first discovered your business. Someone might see your flower-delivery ad on Facebook in January, remember the name in April when they need a Mother's Day arrangement, search your business name on Google, land on your website and submit a contact form. The website session shows a Google referrer and the form arrives through your site, but the customer discovered you months earlier on Facebook.
Self-reported source asks the customer directly. The answer reflects their memory and interpretation, and both matter. A person who says "Facebook" might mean they saw your post, clicked a tagged photo from a friend, or joined a local group where your business was recommended. That lack of precision is real: customers do not track their own journeys with UTM codes. The self-report tells you what stuck in their mind, not a timestamped sequence of every touchpoint.
Automated tracking captures different information. A tagged profile link records the click. Website analytics reports the referring domain. Your inquiry log shows whether the first message arrived as an Instagram DM, a phone call or a form submission. None of these signals alone proves where discovery happened, and none automatically contradicts the customer's own answer. Track social media leads by combining automated and self-reported signals rather than assuming one source of truth.
When to ask the question
Ask after the customer has made clear contact—submitted a form, sent a message or called—and before the conversation moves to detailed service questions. Intake is the natural moment: you already need their name, the type of request and contact details, so adding one source question costs little friction.
The question can be optional. A required field with no Prefer not to say or Not sure option forces customers to guess or pick a random answer just to proceed, and that forced data is less useful than a blank marked unknown. Someone booking an urgent building-material delivery may skip the question entirely to reach the delivery-time field faster, and that choice is information: speed mattered more than sharing how they found you.
Do not ask the source question again on every repeat order from the same customer. Once you have recorded their answer for the first inquiry or purchase, tag that customer record and reuse it. If you want to understand repeat behavior separately, track it as a return visit or repeat order in your simple social media inquiry log, not by asking the same discovery question at every transaction.
To connect auto repair marketing with repeat-service inquiries, use the Auto Repair Shop Marketing: Turn Service Inquiries Into Repeat Business guide.
Neutral question wording and answer options
The exact wording shapes what customers report. "How did you hear about us?" is broader and more neutral than "Where did you find us?" or "What brought you here today?" The first phrasing invites discovery stories—"A neighbor mentioned you" or "I saw your truck"—while the last phrasing can tilt toward the immediate task and the middle phrasing assumes active searching.
Answer options should cover the channels you actually use, leave room for answers you did not predict, and avoid forcing a single choice when the customer recalls multiple touchpoints. A checkbox list allows multiple selections; a dropdown with Other and a text box captures answers outside your list.
| Answer option | When to include it | Coding note |
|---|---|---|
| You maintain a page, post in groups or run ads | Includes page, groups, Marketplace and ads | |
| You maintain a business profile or post regularly | Includes posts, stories, reels and messages | |
| Google search | Customers might search your business or service type | Distinguish branded and category search later |
| Friend or family | Referrals and word-of-mouth happen in your market | Can include neighbor recommendations |
| Saw your truck/sign | You use vehicle wraps, yard signs or job-site signs | Groups physical visibility |
| Repeat customer | You serve returning customers | Separates discovery from subsequent orders |
| Other | Always | Requires a text box for the specific answer |
| Not sure / Don't know | Always if the question is optional | Honest uncertainty, not missing data |
If your automotive-accessories shop runs Facebook ads, maintains an Instagram account and depends on word-of-mouth, include all three platforms plus search, referral and Other. If you have never posted on Instagram, do not include it in the list just because it exists; an option customers never choose is clutter, and an option that does not match your activity confuses the results.
Add a short text box after Other so customers can write "Nextdoor," "YouTube," "Craigslist" or another channel you did not list. Read these free-text answers weekly and add a new checkbox option when the same unlisted answer appears frequently. If five customers write "Nextdoor" in the Other box during one month, add Nextdoor as a standalone option the following month.
Allowing multiple answers without losing the primary driver
Some customers remember more than one touchpoint. A person might check Google search and Friend or family because they heard the recommendation first, then searched your name to find your phone number. Both are true, and both are useful: the referral planted awareness, and search provided the contact path.
A checkbox list captures multiple selections, but your summary report will be clearer if you also ask which channel mattered most or code a primary source yourself using a consistent rule. One approach: if the customer checks multiple boxes, add a follow-up question—Which of these was most important in your decision?—with radio buttons showing only the options they selected. Another approach: always code the earliest remembered touchpoint as primary when the sequence is clear from the answers, and mark the rest as secondary touches.
Do not assume one selection is automatically the primary driver. If someone checks Instagram, Facebook and a friend's recommendation, you cannot reliably guess which happened first or mattered most without asking. Missing this nuance is better than inventing a false hierarchy.
Coding answers and separating discovery from contact channel
Store the raw customer answer in one field and your coded interpretation in a separate field. The raw answer preserves what the customer actually said; the coded field groups similar responses for reporting.
Example coding table (illustrative)
| Raw customer answer | Coded primary category | Coded secondary category | Notes |
|---|---|---|---|
| Social | No detail on page vs. group | ||
| Friend told me about you | Referral | Word-of-mouth | |
| Search | Search | Likely branded | |
| Saw your truck in my neighborhood | Physical visibility | Offline | |
| Instagram and a friend | Multiple | Social + Referral | Both marked; primary unknown |
| Not sure | Unknown | Honest uncertainty | |
| (blank) | Not provided | Customer skipped the question |
Keep the latest contact channel in a different field. If the customer says "Facebook" as their discovery source but the inquiry arrived through your website contact form, record both facts separately: discovered via Facebook, contacted via website form. The two fields together show the journey without conflicting. Attribution models in analytics platforms assign credit to touchpoints using defined rules, and self-reported source is one additional input rather than the authoritative single truth. What counts as a social media lead explains why the intake method and the discovery channel are distinct dimensions.
When "unknown" or "direct" is not proof of failure
An Unknown or Not sure answer does not mean your marketing failed. It means the customer does not remember a specific channel or encountered your business in a way that did not register as a marketing touchpoint. Someone moving into a neighborhood might see your contractor sign on three job sites over six months, register the name subconsciously, then search for "siding contractor near me" when they are ready. They found you through distributed physical presence and search together, but might report only search because the signs did not trigger a conscious memory.
A blank source field or Prefer not to say is not zero evidence; it is missing data. Do not sum all non-social answers as proof that social delivered nothing. Track known sources, count unknowns separately and avoid filling gaps with assumptions. If ten customers report Facebook and fifteen report Not sure, you know Facebook delivered at least ten attributable contacts, not that the unknowns came from nowhere.
Website analytics often show a "direct" traffic segment—sessions with no measurable referrer. Direct traffic includes people who typed your URL, clicked an untagged link, opened a bookmark or arrived through a path analytics cannot see, such as a link in a text message or a PDF. Some of these visits followed social posts, but without a UTM parameter the session appears direct. UTM tracking for social profile and post links reduces untagged traffic, but no tagging system is perfect. Self-reported source can surface social discovery that left no digital trace.
Self-reported source setup checklist
- [ ] Choose question wording that fits your intake process and sounds neutral.
- [ ] List answer options matching the channels you actually use or expect to hear about.
- [ ] Add Other with a text box and Not sure or Prefer not to say as standalone options.
- [ ] Decide whether to allow multiple selections or require a single choice. If allowing multiple, plan how to code the primary source.
- [ ] Separate the raw customer answer field from your coded category field in your records.
- [ ] Store self-reported source in the same inquiry record where you track contact channel, request date and outcome.
- [ ] Review Other free-text answers weekly and add a new checkbox option when the same answer appears repeatedly.
- [ ] Do not ask the same discovery question again for repeat orders from a known customer.
- [ ] Distinguish discovery source from the channel used for the latest message or call; record both in separate fields.
What to do with self-reported source data
Summarize source answers monthly alongside your tracked inquiry counts and outcomes. If your monthly social media report shows twenty Facebook inquiries from automated tracking and eighteen customers self-reported Facebook, the two numbers will rarely match exactly but should be in the same range. A large gap—automation shows two Facebook inquiries, but fifteen customers say Facebook—suggests your tracking missed something, such as untagged profile links or messages that bypassed your tagged website.
Use self-reported answers as a clue requiring further investigation, not proof of what generated an inquiry. If multiple customers write a channel you were not monitoring, investigate that reported referral path using available business evidence to understand whether it warrants observation.
Compare reported source against actual outcomes to see whether certain channels deliver higher close rates or order values. If ten customers report Google search and eight of those booked a job, while twenty report Facebook and five booked, search produced a higher close rate in this comparison. That difference might reflect audience readiness—people searching are often closer to buying—or it might be random variation in a small sample. Avoid declaring one channel the winner without enough data, but do watch for consistent patterns over several months.
Self-reported source does not replace tracking, and tracking does not replace self-reported source. Both together give you visibility into what customers remember, how they arrived and which signals you can measure directly. When social posts produce no inquiries, check both your tracking and your intake question: perhaps inquiries are happening but you are not capturing where they came from, or perhaps the contact path from posts to inquiry is unclear and customers give up before asking.
Frequently asked questions
Should the source question be required or optional?
Optional is safer unless you have a specific business reason to require it. A required field with no "not sure" option forces customers to guess, and forced guesses dilute the usefulness of the data. If completing an order is more urgent than sharing discovery details, let the customer skip the question rather than risk abandoning the form. An optional question with Prefer not to say or Not sure as explicit choices captures honest uncertainty, which is more useful than a random selection.
What if a customer selects multiple channels?
Allow it, because multiple touchpoints are real. Someone might see your Instagram post, hear about you from a friend and then search your business name before calling. All three influenced the decision. If you need a single primary source for reporting, add a follow-up question asking which was most important, or apply a consistent rule such as coding the earliest remembered touchpoint as primary. Do not force a single choice at intake unless your system cannot handle multiple selections; losing the nuance is worse than adding one follow-up field.
Can I compare self-reported source to UTM data in analytics?
Yes, but recognize they measure different things. UTM parameters on a link show that someone clicked that specific tagged URL and arrived on your site, while self-reported source reflects what the customer remembers about discovery. A customer might click a tagged Facebook link in January, return via Google search in March and report "Facebook" because that is where they first heard of you. The January UTM and March search session are both real, and the self-report is also accurate. Store all three signals—tagged click, session referrer and self-reported answer—so you can compare them without expecting perfect alignment.
One calm closing thought: asking customers how they found you works when the question is short, the answer options match the channels you use, and the data lives in the same place as your inquiry and outcome records. Self-reported source will not explain every visit or perfectly match your tracked clicks, but it surfaces the channels customers recognize and the referrals that matter enough to mention. Start by adding the question to your next intake form, then read the answers weekly to see what you are learning and what you are missing.
A clear picture of where inquiries come from—tracked, reported and unknown—makes it easier to decide where to spend your time and budget next.
Free SEO, GEO and SMM strategy, explained with examples of potential customers.