I was handed a small LinkedIn Ads budget, under $3,000 a month, and told to make it work anyway.
No funnel. No retargeting. Just one lead generation campaign, with the objective of getting demo requests through form fills.
LinkedIn practitioners usually give two pieces of advice here. A small budget is better suited to retargeting, and a fully cold audience needs around $5,000 a month to start.
This campaign had both problems at once: a fully cold audience on a small budget.
The conventional answer would have been to build a full funnel with top of funnel awareness, middle of funnel retargeting, bottom of funnel conversion campaigns.
But a funnel needs enough budget and volume to keep every stage running. At this budget, spreading the spend across multiple stages would leave too little behind to make any one of them work well.
So the question wasn’t whether we could build a perfect LinkedIn funnel.
It was whether we could make one cold campaign work without one.
That meant accepting one important trade-off; we wouldn’t get the awareness that a multi-touch funnel normally creates. Several touches before a conversion can make a lead more familiar with the brand and the problem you’re solving.
But that awareness has a cost, and I wouldn’t recommend asking LinkedIn to carry all of it, even with a bigger budget. Email, events, and content can build awareness across channels without making one paid campaign responsible for the entire journey.
Intent also works differently on LinkedIn. Someone searching on Google usually arrives with a problem already in mind. A person who sees a LinkedIn ad is often encountering the idea for the first time.
Even when that person fills out a form, it doesn’t mean they are sales ready.
So whether you run a funnel or a single campaign, a LinkedIn lead generally needs more qualification than a high-intent Google Search lead.
A funnel can build familiarity along the way, but it doesn’t guarantee stronger intent or a faster conversion.
Skipping the funnel doesn’t remove that difference in intent. It means you’re choosing not to pay for multiple stages of the journey on LinkedIn. The trade-off is that more of the qualification work has to happen after the form fill.
We wanted to see how far a single cold campaign could go under those constraints.
The campaign closed out with a 0.61 percent click-through rate and a 10.28 percent form completion rate. Cost per lead landed at $247.65.
Metadata.io’s 2025 analysis of $57.6 million in B2B LinkedIn ad spend put the median cost per lead at $202, across 153 advertisers.
Coming in above that median made sense given the conditions. This campaign was fully cold, with no warm or retargeted traffic mixed in to pull the number down. And the conversion wasn’t a content download. We were asking someone who had just encountered the brand to book a call.
The question then became, what made the campaign perform as well as it did?
None of it came down to one clever targeting trick or one winning ad. It came down to getting the basics right, the things LinkedIn specialists talk about all the time, but that are much harder to get right when every wasted dollar matters.
They came down to four things:
- Getting the right people. The audience has to match the companies you actually want, and the budget has to reach them when they are online.
- Getting the message right. The ask has to match the trust that exists, and the ads need a fair test and honest reporting.
- Removing friction from the form fill. The click has to land somewhere that makes conversion as easy as possible.
- Following up properly. A form fill is only the start of the qualification process. Follow up quickly and across more than one channel.
Getting the right people
The audience has to actually be the audience you think it is
LinkedIn gives you plenty of first-party data to work with from industry to company size, job function and more. Having that data doesn’t guarantee an accurate audience.
Some of these fields depend on how companies classify themselves. Industry, for example, starts with the category a company selects for its LinkedIn page. LinkedIn can also layer inferred categories on top of that.
With a limited set of categories to choose from, similar companies can end up classified differently. We saw that firsthand in this campaign.
We started with IT services and consulting firms, using LinkedIn’s native filters. When we looked at the companies actually entering the audience, close to 60% were product companies, the exact ones we were trying to avoid.
Excluding them one by one wasn’t a practical long-term fix. So we excluded the Software Development industry altogether.
That created a second problem. A major chunk of the companies it removed turned out to be legitimate targets, and product companies were still showing up under IT Services and Consulting.
We only caught that by going through the excluded list manually, company by company. The industry labels weren’t reliably telling us what these companies did.
So instead of relying on LinkedIn’s industry filters, we built a list of the companies we actually wanted to reach. Nothing else about the campaign changed. That change alone reduced cost per lead by 50%.
First-party data is a good starting point. It works best when you check it against the actual companies behind the filter, whether you’re adding them or excluding them.
A small LinkedIn ads budget has to reach people when they are online
A monthly budget divided evenly across every day of the month assumes the audience and the auction behave the same way every day. They don’t.
We saw this clearly in the same campaign. When we broke down the daily performance by weekday and weekend, average daily spend on weekdays was roughly 3.5x higher than on weekends. Average daily impressions were also close to 2.5x higher.
Several weekend days spent almost nothing, even though the daily budget cap was available. The issue wasn’t that people who saw the ads were less likely to click as the CTR remained broadly consistent across weekdays and weekends. The difference was how much delivery LinkedIn could find in the auction.
That matters when you’re managing a fixed monthly budget. A flat daily allocation can leave part of that budget effectively unused on days when there is less opportunity to deliver, while limiting how much you can spend when the audience is more active.
For this campaign, shifting more of the daily budget toward weekdays made more sense than treating every day equally. The monthly budget didn’t change. We simply gave LinkedIn more room to spend when there was more opportunity to reach the audience.
Getting the message right
On a small LinkedIn ads budget, the ask has to match the trust that exists
Cold traffic rarely converts well when the first ask is a demo or a 30-minute commitment. The person may never have heard of you.
Benchmarks for cold paid-social campaigns generally show that direct demo requests convert at lower rates than lighter offers. The exact difference depends on the audience, offer, and how familiar people are with the brand.
In this campaign, we tested a few versions of the ask.
One asked people to book a demo. Another offered a consultative session. A third offered a customized plan.
The smallest ask won: a short 15-minute call, with nothing bigger implied behind it.
The pattern was clear in our tests. The meaningful jumps in leads followed the smaller ask, not a bigger commitment.
The test comparing your ads has to actually be fair
LinkedIn recommends running multiple ads within an ad set so you can identify and replace underperformers. But simply putting ads together doesn’t guarantee a fair comparison.
To understand which ads work, each needs enough impressions to generate a meaningful amount of data. The problem is that LinkedIn’s delivery system predicts which ad is likely to perform better and can skew delivery toward it before there’s enough evidence to make that call.
On one account, one ad received about 95% of a day’s impressions while another received about 5%. The ad getting almost no delivery had the better click-through rate.
We tried LinkedIn’s Rotate Ads Evenly setting. It gives each ad a fair chance to enter the auction, but it doesn’t guarantee an even split of impressions. Actual delivery still depends on performance and how the auction plays out.
When impressions and reach continued to drop under that setting, we moved the underperforming ad into its own ad set. With each ad in a separate ad set, they were no longer competing with each other for delivery, making the comparison more meaningful.
Within a few weeks, we were seeing about two form fills a week consistently.
The lesson isn’t that every campaign should use separate ad sets. It’s that you need to understand how delivery affects your tests before deciding which ad is actually performing better.
The reporting has to be granular enough to show you the truth
A single blended number across multiple ads can hide a lot. One message might be doing all the work while another contributes almost nothing.
Every ad needs its own line in the report.
In a funnel, a weak result at one stage might be offset by stronger performance elsewhere. In a single-campaign setup, the blended number can make the campaign look healthy even when one ad is wasting spend.
Granular reporting helps you see what’s working, what isn’t, and where to make the next change.
Remove friction from the form fill
The click has to land on a real conversion point
Experienced LinkedIn practitioners often recommend native Lead Gen Forms for cold campaigns instead of sending people straight to a website or landing page.
LinkedIn reports an average Lead Gen Form conversion rate of 13%, compared with 4.02% for an average landing page, citing an Unbounce study for the landing-page figure. These are broad benchmarks, not a like-for-like comparison of cold demo-request campaigns. Actual conversion rates vary by audience, offer, industry and conversion goal.
The reason is simple. A native lead form can autofill from a person’s LinkedIn profile, making submission quick. A landing page asks a cold visitor to leave the platform, load an unfamiliar page, and enter their details manually.
In my own experience, I’ve rarely seen more than a handful of form fills come from a landing page or website, even for content downloads or event sign-ups. Cost per click has consistently run higher too, adding to the cost of weaker conversion.
For a cold campaign without a funnel behind it, the conversion point can make a substantial difference.
Lead follow up for LinkedIn ads on a small budget
Someone has to actually follow up, and follow up the right way
LinkedIn leads often land with an inbound BDR or SDR team that treats them like Google Search leads. That’s a mistake.
A Google Search lead may have actively searched for a specific solution. A LinkedIn lead may have been interrupted mid-scroll and is encountering your brand for the first time.
Treating the two the same way means writing off leads as bad when they only need a different approach.
Follow-up needs to be fast and use more than one channel: a call, an email, and a LinkedIn message, rather than one attempt followed by silence.
If the first email goes unanswered, the sequence should continue. Share a relevant case study or a short piece of thought leadership to build credibility and give the prospect a reason to continue the conversation.
One way to reduce friction is to place a Calendly or Chili Piper link directly on the thank-you page after form submission. That gives interested prospects a way to book a call without waiting for someone to chase them first.
Where this leaves you
Optimising LinkedIn campaigns comes down to small things done consistently. Understanding how each one affects the result is what makes the difference.
A full funnel makes sense when you have the budget and volume to keep every stage running. It gives you more opportunities to build familiarity before asking for a conversion, though it doesn’t guarantee stronger purchase intent.
A direct, cold, bottom-of-funnel campaign can reduce the need to fund multiple paid stages. But it shifts more of the work to qualification and follow-up after the form fill.
A budget of $5,000 a month can give you more room to build and sustain that structure. It’s a legitimate approach, and often the easier one to operate.
A budget below $3,000 leaves less room for error. A single campaign can still generate leads, but its performance depends heavily on the quality of the audience, the offer, the conversion experience, the ad testing, the reporting, and the follow-up.
Every one of those details needs to be checked, not assumed. If one goes wrong, the campaign absorbs the mistake directly. There’s no other stage behind it to soften the blow.
The real trade-off is room for error. A bigger budget buys you some. A budget this size asks you to earn every bit of it yourself.