The lead quality lesson that changed how I plan every campaign
By Hukum Yadav. Published October 2026. 5 minute read.
Let me ask you something before we start. When your Meta Ads report shows 400 leads this month, do you celebrate? When your client calls to ask how the campaign is going, do you lead with the lead count because it is the easiest number to be proud of?
I used to. Every single month. Until a client called me and asked a question I did not have a good answer to.
I want to walk you through the lesson that changed how I think about every lead generation campaign I plan today. Not a theory I picked up from a course. Not a framework someone else packaged. Something I learned the hard way, in a real account, with a real client who deserved better than what I was giving him.
The month the sales did not match the leads.
A couple of years into my career, I was running Meta Ads for a mid-sized real estate project in Delhi NCR. The brief was simple. Generate leads for a new residential launch. The budget was healthy. The creative was tight. The targeting was built properly.
The first month, I delivered 412 leads. Cost per lead was under 150 rupees, which in real estate is a dream number. I was proud of myself. I built the monthly report, highlighted the big number, and sent it over to the client with a confident email about how we were going to scale next month.
Three days later, the client called me. He was polite, as most good clients are when they are frustrated. He asked me one question. “Hukum, we got 400 leads. How many of them actually came to the site visit?”
I did not know. I guessed. I said probably 30 or 40. The real number, when his sales team pulled it, was 11. Of those 11, four were serious. Of the four, one eventually booked. One sale, from 412 leads, from an entire month of spend.
I remember sitting at my desk after that call feeling like something was wrong with my whole approach. The campaign had done exactly what I asked it to do. It generated cheap leads. The problem was that cheap leads were the wrong thing to ask for. And I had been asking for the wrong thing on almost every lead gen account I had ever run.
If you have been in this field for any serious length of time, you have probably had a version of this call. The dashboard says you are winning. The business says you are not. And you are sitting there wondering which number is actually true.
What was actually happening in that account.
Here is what I eventually figured out, and what I want you to understand clearly.
Meta’s algorithm does exactly what you tell it to do. If you tell it to optimise for “leads”, it will find you people who fill out forms. Not people who buy. Not people who show up to a site visit. Not people who have the budget to actually purchase a 1.2 crore apartment. It will find you the people most likely to click a button and type their phone number into a field.
Those two groups, people who fill out forms and people who actually buy, overlap. But they are not the same people. And if your campaign is only measuring the first group, the algorithm will quietly stop finding you the second group entirely. It is cheaper, faster, and easier to find form-fillers than it is to find buyers. The algorithm takes the easy path. Every time.
My 412 leads from that real estate client were a textbook example. The form was two fields. Name and phone number. No qualifying questions. No budget filter. No intent signal. I had made it so easy to submit the form that the people submitting were students, curious browsers, people who were just filling it to see what the project was, competitors doing recon, agents hunting for commission opportunities, and a handful of real buyers lost in the noise.
The platform was not broken. My brief to the platform was broken. I was telling the system “find me people who fill forms”. The system did exactly that, brilliantly, cheaply, at scale. The client was telling me “find me people who will buy an apartment”. Those two sentences are not the same, and I had been treating them like they were.
That is the moment I understood something I now believe is one of the most important truths in performance marketing. You get exactly what you optimise for. Not what you hope for. Not what the client wants. Not what feels like it should be the goal. You get what the account is literally set up to find. And if those two things are different, you have a problem no amount of spend will fix.
What I do differently now, and what I want you to do too.
After that lesson, I rebuilt how I plan every lead generation campaign. If you take nothing else from this post, take these four shifts.
1. Define what a quality lead actually looks like, in writing, before you launch anything.
On day one of any lead gen account, I sit down with the client and ask them to describe their ideal customer in uncomfortable detail. Not “someone interested in our product”. That is useless. I want to know what budget range, what location, what timeline to purchase, what questions they typically ask, what makes them a yes versus a maybe. If the client cannot answer this clearly, we are not launching ads yet. We are figuring this out first. Because if you do not know what quality looks like, you cannot ask the platform to find it, and you certainly cannot measure whether you found it. Most agencies skip this step because it is awkward and slow. The ones who do it well never go back.
2. Qualify at the form, not after.
I used to let every form be as short as possible because “shorter forms get more leads”. That is true. It is also beside the point. Shorter forms get more leads of lower quality. For most of my clients now, I deliberately add two or three qualifying questions. Budget range. Timeline. One intent-level question specific to the business. Yes, lead volume drops. Yes, cost per lead goes up. But lead-to-sale ratio goes up by five or ten times, which is the number that actually matters. The client cares about customers, not forms. Design the form for the customer, not the dashboard.
3. Report two numbers every month, not one.
Every monthly report I send now shows platform-reported leads on one side, and qualified leads on the other. Not “leads generated”. That number is a vanity metric. “Leads that matched our quality definition” is the real number. The gap between them tells you whether the campaign is actually working or just looking like it is. Showing clients both numbers does something interesting. It builds trust fast. The client stops wondering what you are hiding, because you are clearly not hiding anything. And it forces me, every month, to actually look at the quality side of the account, not just the volume side.
4. Feed the quality signal back to the platform.
This is the part most people skip, and it is the one that compounds the hardest. If you have a CRM or a sales system, send the “qualified lead” status back to Meta and Google as a conversion event. Use offline conversions. Use CAPI. Use whatever your stack allows. Because the platform is only as smart as the data you give it. If you only tell it who filled a form, it will find you more form fillers. If you tell it who became a real opportunity, it will slowly, then quickly, start finding you more of those. This single setup has, in my experience, done more for lead quality than any targeting change I have ever made.
The real shift was not technical.
I want to end with the part that matters most, because the tactics above are easy to copy. The mindset behind them is harder.
When I was celebrating lead counts, I was working for the dashboard. I was proud of a number that had almost nothing to do with my client’s actual business. I was reporting on something that felt impressive but was not creating value. And in doing so, I was quietly training the algorithm to deliver more of the wrong thing, faster, cheaper, forever.
The moment I started measuring what the client actually cared about, which was real customers walking in the door, everything changed. My campaigns got sharper. My spend got leaner. My clients stayed longer. The platform numbers looked less impressive on the surface. But the client’s bank account started telling a different story. Which, honestly, is the only story that ever mattered.
If you are running lead gen campaigns right now, I want you to try something this week. Before your next launch, write down one sentence. If this campaign produces 100 leads next month, how many of them need to be real customers for me to consider this a win?
If you cannot answer that in one line, you are not running a lead generation campaign. You are running a form fill campaign and hoping customers show up by accident. Those are two very different things. The platforms do not know the difference. Your job is to teach them.
That is the lesson that changed how I plan every campaign. I hope it changes how you plan yours too.
If this one resonated, I would love to hear your own version of this story. The month you realised the lead number was lying to you. We have all had one. The sooner we talk about these openly, the better our whole industry gets.
