Key Takeaways
- A large share of online sharing happens through private channels analytics cannot track, around 69% of sharing according to RadiumOne research reported by MarTech.
- This 'dark social' traffic is usually dumped into your 'direct' bucket, so analytics mislabels word of mouth and referrals as people typing in your address.
- Last-click attribution credits the final touch and ignores the podcast, post or recommendation that actually started the journey.
- The simplest fix beats expensive tools: ask every lead 'how did you hear about us?' and record the answer.
- Use analytics and that one question together, not one instead of the other.
Your analytics is lying to you, and it is not its fault
Every business owner wants the same thing from their numbers: a clear answer to what marketing is working. Most analytics tools cannot give you that, and the gap is bigger than people realise.
A large share of online sharing happens through private channels that analytics simply cannot see. RadiumOne research reported by MarTech put dark social at around 69% of all sharing activity. When that traffic arrives, it usually gets filed under 'direct', as if everyone typed your web address from memory.
This guide explains why attribution is broken, what dark social really is, and why the simplest fix, asking every lead how they found you, beats the fancy tools most small businesses do not need.
What 'direct' traffic really is
When you open your analytics and see a chunk of 'direct' visitors, it is tempting to read it as loyal customers typing you in. Some of it is. Most of it is something else.
- Private shares: A link sent over WhatsApp, email or a group chat. The source is stripped, so it lands as direct.
- Offline sources: Someone heard you on a podcast, saw a flyer, or got a recommendation, then searched and visited later.
- Tracking gaps: Links without proper tags, app browsers, and privacy settings all hide the true origin.
In other words, 'direct' is mostly a polite word for 'we do not know'. Treating it as a real channel leads to the wrong decisions.
Dark social, explained simply
Dark social is the name for all that private sharing analytics cannot track. It is not sinister. It is just how people actually recommend things.
Think about how you share a useful link. You probably copy it and drop it into a message to a friend or colleague. You do not post it publicly with a tracking tag attached. Multiply that by everyone who likes your business, and you have a huge stream of word of mouth that never shows up properly in your reports.
This matters because dark social is often your best traffic. A personal recommendation converts far better than a cold ad. If your analytics hides it, you might cut the very activity that drives those recommendations.
Why last-click attribution gets it wrong
Most default reporting uses last-click attribution. It hands all the credit to the final thing someone did before they converted, which is usually a branded search or a direct visit.
That sounds reasonable until you picture a real customer. They hear about you on a podcast, mention you to a friend, see a post weeks later, then finally search your name and enquire. Last-click credits only the name search. The podcast that started it all gets nothing.
The result is predictable. Late, cheap touchpoints look like heroes. The real demand creators look worthless and get cut. We have seen owners almost scrap the activity that was quietly feeding everything else, simply because the model could not see the connection.
The fix that beats every tool: just ask
Here is the contrarian bit. You do not need a six figure attribution platform. You need one question, asked consistently, and recorded properly.
- On your form: Add an optional 'how did you hear about us?' field to your enquiry form. Keep it short, with an 'other' box.
- On sales calls: Ask it naturally early in the conversation. People are happy to tell you.
- In onboarding: Confirm it when a new client signs up, while the memory is fresh.
- Record everything: Put every answer in one place, your CRM or even a simple spreadsheet, so patterns appear over time.
In our own work, the answer to that one question regularly reveals sources our analytics never credited, a referral from a past client, a comment in a Facebook group, a mention on someone else's podcast. None of it showed up in the dashboard. All of it was driving enquiries.
If you want to turn those answers into action, our conversion rate optimisation work helps you capture and use this data, and our guide on why traffic does not turn into enquiries covers the next step once the leads arrive.
How to use the answers alongside analytics
The goal is not to throw analytics out. It is to use each tool for what it does well.
- Use analytics for the measurable: On-site behaviour, trends over time, and channels it can track properly such as paid ads with tagged links.
- Use the question for the hidden: Word of mouth, dark social, offline and podcast mentions that analytics buckets as direct.
- Compare the two: When a source keeps coming up in answers but never in analytics, you have found something worth investing in.
- Watch the trend, not single answers: One reply is an anecdote. Fifty replies are a pattern you can act on.
For a fuller framework, our guide on how to track marketing ROI for UK SMEs ties this together, and if budget allocation is the real question, how much a UK small business should spend on marketing is a sensible next read.
Frequently Asked Questions
What does 'direct' traffic actually mean in analytics? Direct traffic is what analytics records when it cannot work out where a visitor came from. Some of it is people typing your address in, but a lot is links shared privately in WhatsApp, email and messaging apps, plus visits from offline sources like a podcast or a recommendation. It is best read as a bucket of unknowns, not a real channel.
What is dark social? Dark social is sharing that happens through private channels analytics cannot see, such as WhatsApp, Messenger, email, Slack and direct messages. When someone copies your link and sends it to a friend, the visit usually lands in your direct bucket with no source attached. It is one of the biggest reasons attribution is harder than it looks.
Why is last-click attribution misleading? Last-click gives all the credit to the final touch before someone converts, usually a branded search or a direct visit. That ignores the podcast, recommendation or post that started the journey weeks earlier. It makes the cheap, late touchpoints look like heroes and your real demand sources look worthless.
How do I add a 'how did you hear about us?' question? Add it as a simple optional field on your enquiry form, ask it on sales calls, and include it in onboarding. Keep it open or use a short list with an 'other' box. The key is to record every answer in one place, such as your CRM or a spreadsheet, so you can spot patterns over time.
Should I stop using analytics then? No. Analytics is still useful for trends, behaviour on your site and channels it can track properly. The point is to pair it with the answers people give you. Use analytics for the measurable, and the 'how did you hear about us?' question for everything analytics quietly hides.
The Bottom Line
Attribution is genuinely broken for small businesses, mostly because so much of your best marketing happens in private. Your analytics is not lying on purpose, it just cannot see word of mouth, dark social and offline mentions, so it files them as direct. The cheapest, most reliable fix is to ask every lead how they found you, record it, and read it next to your analytics. Do that for a few months and you will know more about what works than most companies learn from expensive tools.
If you want help setting this up properly, get in touch. We can connect the dots with practical analytics and reporting and a wider digital marketing plan built on what is actually driving your enquiries.
