The headline: manual 100, maximum delivery 203
| Bid approach | Cost per click, indexed |
|---|---|
| Manual bid | 100 |
| Maximum delivery (the platform sets the bid) | 203 |
Same accounts, same period. Handing the bid to the platform roughly doubles what a click costs. On a US account at a $46.98 median cost per landing page click, that is the difference between a $23 click and a $47 one.
This is not an argument that maximum delivery never works. It is an argument that it is priced like a convenience and most accounts are paying for a convenience they did not know they had bought. Maximum delivery is the default in Campaign Manager.
Talk to Phil, who runs this research
He will benchmark your ad account against the 1,071 B2B accounts and $79M of LinkedIn ad spend behind these figures, and build you a business case from it — including what your bidding method is costing you per click. Free. Read-only access. No pitch deck.
Book a call with Phil →Why the platform's default costs you more
Maximum delivery instructs LinkedIn to spend the full daily budget. The bid is then whatever clearing that budget requires. On a tight B2B audience — and tight is the whole point of LinkedIn — there often is not enough cheap inventory to absorb the budget, so the system bids up until there is.
The mechanism shows up clearly in the frequency data. When a campaign runs on a heavier day than its own normal, costs rise with it:
| That day's frequency vs the campaign's normal | Cost per landing page click, indexed | CPM, indexed |
|---|---|---|
| 0.5 to 0.8x | 83 | 92 |
| 0.8 to 1.25x (a normal day) | 101 | 101 |
| 1.25 to 2x | 114 | 115 |
| 2 to 3x | 131 | 135 |
| 3x+ | 166 | 156 |
A campaign pushing three times its usual frequency pays 66% more per landing page click than the same campaign on a quiet day. Maximum delivery is the setting that pushes it there, because it has been told to spend the money regardless.
Does bidding high help? Mostly no
The received wisdom on conversation ads was to bid maximum, on the theory that inventory is scarce — a member can receive one message ad every 45 days — so you must win the auction. The panel does not support it. Paired within 49 accounts running both:
| Outcome of a higher bid on conversation ads | Change | 95% interval | Real? |
|---|---|---|---|
| Cost per send | +6% | −1% to +17% | No (ns) |
The interval includes zero, so the honest reading is that the higher bid bought nothing you can measure. We held the opposite view until September 2026 and changed it when this cut came in. If you are still running conversation ads at maximum bid on the old advice, that is budget you can reclaim today. See conversation ads bidding.
Video is the exception. In the video panel a high bid cost 90% more per view — which is a real effect, not noise, and means the bid does buy delivery there. Whether you want to pay it is a different question; the video views objective already beats brand awareness by 18% on cost per view and engagement by 25% on cost per completion without touching the bid.
What to bid, by format
| Format | What the panel says | Do |
|---|---|---|
| Sponsored content, thought leader ads | Maximum delivery costs 103% more per click | Manual bid, start below LinkedIn's suggested range |
| Conversation ads | A higher bid moves cost per send +6%, not significant | Bid low. The old bid-maximum advice is wrong |
| Video | High bid costs +90% per view; auto max-views beats manual CPV | Use the video views objective, automatic |
| Lead gen forms | Enhanced conversion bidding ran 62% worse than maximum lead on webinar registrations | Avoid enhanced conversion on thin conversion volume |
The objective moves the click more than the bid
Before you tune a bid, check you are buying the right thing. From the UK cut, where the objective comparison is cleanest:
| Objective | Cost per landing page click | Landing page click rate |
|---|---|---|
| Website visits | $10.03 | — |
| Conversions | $10.26 | 0.45%, best of any objective |
| Engagement | $18.18 | — |
| Brand awareness | $53.03 | — |
| Video views | $82.48 | — |
A 5x spread, from objective choice alone. If you are running brand awareness and judging it on landing page clicks, the bid is not your problem.
The order to fix things in
The post-click panel measured which lever actually moves cost per lead. The relationship between landing page conversion rate and cost per lead was 72 times stronger than the relationship between bid and cost per lead (72 against 1.5). That ratio should decide where your next hour goes.
- The landing page. 72x. Going from 1% to 2% halves your cost per lead and no bid change available to you can do that.
- The bidding method. Manual instead of maximum delivery. Worth roughly half your click cost.
- The objective. Up to a 5x spread on cost per landing page click.
- The bid number itself. Last, and on conversation ads, not at all.
All figures are from Kiin Labs: a panel of 1,071 B2B LinkedIn advertising accounts, $79M of spend, 22,942 campaigns, 1.34 billion impressions, 2.4 million landing page clicks and 139,231 form submissions, for the 12 months to 7 September 2026. LinkedIn Audience Network is excluded throughout. Figures are medians, not averages, because the top 10% of accounts spend 64% of the money. Where a comparison says "paired", both options ran inside the same account, so account quality cannot explain the gap.