Most small B2B companies I talk to have a lead magnet somewhere on the site. It is usually a PDF guide, written two or three years ago, sitting behind a form in the footer or on a landing page nobody links to. When I ask how many customers came from it, the answer is a pause.
That is the normal result, and the reason is straightforward. The thing was built to collect email addresses. It was never built to help anybody buy.
Why the generic ebook stopped working
For years the trade worked like this. A buyer wanted to understand a topic, you wrote a 15 page summary of that topic, and they handed over an email address to read it. The information had a price because it was hard to find elsewhere.
Information is no longer hard to find. A buyer who wants to understand how commercial HVAC maintenance contracts are structured asks ChatGPT and gets a competent answer in nine seconds, with no form. Your PDF is competing with a free, instant, personalised version of itself.
So the generic guide now fails twice. Few people fill in the form, and the ones who do are often students, competitors, or somebody doing background reading with no project and no budget. Then a sales rep spends a week chasing them, decides the leads are rubbish, and stops calling any of them. I see this constantly.
The piece of information that is still scarce is the part that comes from doing the work. What things cost in this region right now. What goes wrong on month four of a project like theirs. Which of the three options fits a company their size. No general purpose tool can produce that, because it is not published anywhere.
The two-part test for a good lead magnet
Before you build anything, run it through two questions.
Could anybody else produce this? If a smart person with a search engine could write the same document in an afternoon, it has no value to trade. Your lead magnet should contain something that exists only because you have done this work many times: your pricing ranges, your failure patterns, your checklist built from actual site visits, your benchmark numbers from jobs you delivered.
Does it help them decide something? A buyer with a real project has a specific question blocking them. Should we replace or repair. Do we need one supplier or three. Is our current setup costing us money. A lead magnet that answers a blocking question gets used. A lead magnet that explains a category gets filed.
If the answer to both is yes, the people who download it are people with projects. That is the whole point. A smaller list of buyers beats a larger list of readers, and it changes how your sales team treats the leads.
Calculators, checklists and assessments beat guides
The format matters more than owners expect, because the format decides whether the output is about the reader or about the topic.
A calculator takes their numbers and returns a number that belongs to them. A freight company enters shipment volume and gets a cost per shipment they can compare to what they pay now. That output is personal, so it gets screenshotted and sent to a colleague, which is free distribution.
An assessment asks eight to twelve questions and returns a position: here is where you sit, here are the two things to fix first. It works because the buyer learns something about their own company that they could not have looked up. This site runs one, the growth readiness score, and the reason it is built that way is that the result is immediate and specific to the person answering.
A checklist works when the buyer is about to do something risky and wants to avoid a known mistake. A list of the 14 questions to ask a supplier before signing, written by somebody who has seen those contracts, gets printed and carried into a meeting.
A guide is the weakest of the four, and it is the one everybody builds, because it is the easiest to produce. If you do build one, make it narrow. Not a guide to warehouse automation. A guide to what warehouse automation costs for a 30,000 square foot facility in the Lower Mainland, with the numbers in it.
Gating or not gating: how to actually decide
Owners argue about this for weeks. The rule I use is simple: gate the thing that produces a personal answer, publish the thing that builds trust.
Publish openly anything whose job is to get you found and believed. Articles, pricing explanations, comparisons of approaches, answers to common questions. These need to be read by search engines and AI assistants, and a form blocks both. A page behind a form does not get cited in an AI answer, because nothing can read it. If getting found in those answers matters to you, that alone settles it, and the foundations of getting found in AI answers covers why.
Gate anything that requires the buyer to give you their situation in order to produce a result. A calculator, an assessment, a custom report. Here the form is not a toll booth. It is the input the tool needs to work, and buyers accept that trade without complaint.
One middle option that works well: let the tool run and show the result on screen with no email required, then offer to send the full version or a comparison by email. The people who give you their address at that point have already seen the value, so they are worth more than a blind form fill.
What happens after the download
This is where most small companies stop, and it is where most of the value was sitting.
The typical setup: the form submits, a PDF arrives, a record appears in the system, and nothing else happens. Maybe a sales rep sees a notification and sends a template message asking if they would like to book a call. The buyer downloaded a document about a problem they have not yet decided to solve, and the first thing your company does is ask for a meeting. The timing is wrong, so they ignore it, and the lead is now cold forever.
What should happen instead depends on what they downloaded, and that is the useful part. The download tells you which problem they have. A buyer who ran the cost calculator is at a different stage than one who took the supplier checklist.
Build three or four follow-up messages that go deeper on the exact problem the download addressed. Not a newsletter. Not company news. One genuinely useful thing each time, spaced over two or three weeks, each ending with a small, low-cost next step rather than a sales meeting. There is a full explanation of how to structure that sequence in the post on email nurture for small B2B companies.
The other half is human. If the download came from a company that matches your ideal customer, somebody should look at it and decide whether to reach out personally. Not everybody. The ones who fit. A sales team that gets 40 downloads a month can look at 40 company names in ten minutes.
Measure it by pipeline, not downloads
Download count is the number everybody reports and the number that tells you the least. A lead magnet with 200 downloads and no deals is worse than one with 22 downloads and three proposals, because the first one is consuming sales attention for nothing.
Track three things instead.
- How many downloads turned into a real conversation. A conversation means a call or a meaningful reply, not an opened email. If this is under one in twenty, the offer is attracting the wrong people.
- How many turned into a quote or proposal. This is the number that connects the marketing work to money. It will lag by months on a long sales cycle, which is why you need the date recorded against every lead.
- How long it takes. If downloads turn into deals nine months later, you need to know that before you cancel the programme at month five. Long cycles kill good programmes because nobody was patient enough to see the result.
None of this works unless the download is recorded against the company record and stays there through to the deal. That is an ordinary data discipline problem, and it is the most common reason owners cannot answer the question at all. If your records are messy, fix that before you build another download, because otherwise you will have no way to judge it.
The takeaway
Stop building documents that explain a topic. A free assistant does that better, instantly, without a form. Build the thing that only you can produce: your numbers, your failure patterns, your judgment applied to their situation.
Pick one format where the buyer puts something in and gets a personal answer out. Publish the trust-building material openly so it can be found and cited, and put a form only in front of the tool that needs their input to run.
Then plan the two weeks after the download before you launch the download itself. In my experience that is the difference between a lead magnet that fills a list and one that fills a pipeline. If you only have time for one of those two jobs, do the follow-up and keep the old PDF.