Two kinds of lead and the mistake of averaging them

A business runs two things: a download offer that generates volume, and a contact form that generates a handful of enquiries.

Both land in the same place, both get counted as leads, and the average tells you nothing true about either.

They are not the same person

A demo or quote request is in-market. They have a problem now, they have decided to act, and they are choosing between options. Small in number, high in value, short in cycle.

A download is researching. Something is on their mind. They may buy in three months, or next year, or never. Large in number, low in immediate value, long in cycle.

Treating them as one pool produces two specific failures.

The first failure: chasing the wrong number

Cost per lead drops when you add a download offer, because downloads are cheap. The blended number looks like an improvement.

Meanwhile the in-market enquiries have not increased, and revenue has not moved.

So the business optimises toward the cheaper lead, shifts budget to the top of the funnel, and the pipeline gets worse while the report gets better.

Separate the two and the picture is honest immediately.

The second failure: treating them the same on contact

Ringing a research-stage download within five minutes annoys them. They wanted a document, not a salesperson, and the call teaches them not to download anything again.

Putting a demo request into a six-week email sequence loses them, because they are choosing a supplier this fortnight and somebody else answered the phone.

Same handling, opposite outcomes.

What each one needs

In-market enquiries need speed above everything. Fast response is the strongest single predictor of winning a considered-purchase enquiry, and it is almost entirely within your control.

Then they need the things that let a buyer decide: price signal, scope, proof, and what happens first.

Research-stage contacts need patience and usefulness. A sequence that teaches rather than sells, at a sustainable cadence, until they raise their hand.

The goal is not to convert them now. It is to be the obvious call when the problem becomes urgent.

Report them separately, always

Two rows, never averaged.

For in-market: volume, response time, conversion to a conversation, conversion to sale, value.

For research: volume, cost, and the rate at which they later become in-market. That last number is the only one that justifies the spend, and almost nobody measures it, because it requires tracking someone across months.

If the research pool never converts to in-market, the offer is attracting the wrong people and the channel should be cut. That is a finding you cannot reach from a blended average.

Should you run both

If your sales cycle is short and demand is immediate, you may not need a research offer at all. Spend everything on in-market capture and response speed.

If the purchase is considered and infrequent, the research layer is where the future pipeline comes from, and skipping it means competing entirely on the moment of intent, which is the most expensive moment to compete in.

Most services businesses need both. They just need to stop adding them together.