A year ago I started tagging every lead with where it came from. Not the rough “I think most of our leads come from word of mouth” gut feel — actual, recorded, never-skipped tracking. Whenever a lead landed, the source got logged before anything else happened.

Twelve months later, I sat down with the data. Some of what I found I expected. A surprising amount, I didn’t.

The setup

I tracked five sources, broadly:

Word of mouth / referrals (existing client introduces someone)
Google search (organic, found us via a search)
Paid ads (we ran a small Google Ads budget for the year)
Social (LinkedIn primarily, some Instagram)
Networking events (in-person, one-to-one)

For each lead I logged the source, whether it converted, and the eventual deal value. Took thirty seconds per lead. Over a year, less than a working day in total.

The headline numbers

Total leads received: 312.

Conversion rates by source:

Referrals: 68%
Networking: 41%
Google search: 22%
Social: 12%
Paid ads: 8%

That last number stopped me cold. We’d spent meaningful money on paid ads. The conversion rate was so low I genuinely wondered if I’d miscounted. I hadn’t.

But conversion rate isn’t the whole story. The real picture comes from combining conversion with deal size.

Average deal value by source

Networking: £8,200
Referrals: £6,400
Google search: £3,100
Social: £2,400
Paid ads: £1,600

Now the picture is even sharper. The sources with the lowest conversion rates were also producing the smallest deals. People who came in via paid ads were both less likely to buy and spending less when they did.

People who came via networking were the highest-value clients I had. Just under half of them closed, and when they did, they paid an average of £8,200.

What this told me about my marketing

I was spending money in roughly inverse proportion to what was working.

The biggest slice of marketing spend was paid ads, because they felt the most “professional marketing” thing to do. The smallest slice — basically zero — was networking, because networking is uncomfortable and hard to budget.

After seeing the data, I made three changes:

Killed paid ads entirely. Saved a decent monthly amount. Lost the 8% of leads that converted, which were also the smallest deals. Net positive immediately.

Doubled down on networking. Bought more event tickets. Said yes to more coffee chats. Made it a calendar habit, not an afterthought. Because the data said the second-highest-value source was the one I was barely using.

Invested in referrals deliberately. This was the biggest change. Referrals had been the highest-converting source by a long way, but I’d never actively asked for them. I started a simple practice: at the end of every successful project, asked the client if they knew anyone else who might benefit. Not pushy. Once, at a natural moment. Result: referral volume nearly doubled in six months.

What was less surprising

Two findings confirmed what I’d suspected:

Google search produced the most predictable, “fine” leads. Not the best, not the worst. People who’d searched for a specific service, found us, and were ready to buy. Mid-conversion, mid-value, low-effort. Worth keeping the website healthy for.

Social produced lots of attention, few leads. Posts that did well got hundreds of likes. The leads they generated were minimal. This is the part that always confuses people who’ve fallen in love with their own engagement metrics: vanity numbers don’t pay rent.

The unexpected pattern

The most surprising finding was about timing. When I plotted when leads converted vs. when they came in, I noticed something:

Referrals closed fast. Average time from lead to deal: 11 days.
Google search closed medium-fast. 19 days.
Networking closed slow. 47 days.
Paid ads closed fast or never. Either within a week, or not at all.

This changed how I forecasted. If I had a quiet pipeline and needed work in three weeks, paid ads (when I still ran them) would respond faster than networking. But that “faster” was usually low-quality. Networking required patience but produced larger deals on a longer arc.

The right strategy isn’t one source — it’s a portfolio with different time-to-revenue profiles.

If you’re not currently tracking lead source, start tomorrow. Don’t wait until you have a “system” — even a tag in a CRM or a column in a spreadsheet is enough. The discipline of always recording it matters more than the sophistication of the tracking.

After three months, look at the numbers. Don’t trust your gut feel about which sources work. The gut feel is almost always wrong.

After six months, you’ll have enough data to make budget decisions. After twelve, you’ll have enough to make strategic decisions about how you grow.

The data is in your business right now. You’re just not capturing it.

stedd.io tags lead source automatically and shows the conversion data on a dashboard — so you don’t have to build the tracking yourself.