Lowest cost is the default on Meta for a reason — it scales the fastest and the auction system optimises CPA on its own as long as signal is clean. Cost cap wins only when CPA control matters more than scale, which is rare in healthy accounts. Bid cap wins only when the buyer has account-level pricing data (LTV per cohort, margin per unit) sharper than Meta's auction model. Most media buyers reach for cost cap when they should be fixing creative or audience fit, and reach for bid cap when they do not have the pricing data the strategy requires. The right strategy depends on account stage and data maturity.
Lowest cost is the right default for most accounts. Cost cap wins only when CPA constraint is a real business need. Bid cap wins only when the account has pricing data sharper than Meta's auction model. Most buyers pick cost cap to feel in control and reach bid cap without the data it needs. Match strategy to account stage and data maturity, not preference.
What does each bid strategy actually do differently?
Lowest cost gives Meta full control of the auction: the system bids whatever it takes to spend the daily budget as efficiently as possible, with no cap on CPA or bid. Cost cap tells Meta a target CPA and asks it to deliver as much volume as possible without exceeding the cap on average — individual conversions can go above, but the ad set's running average must stay at or below the target. Bid cap tells Meta the maximum per-auction bid, which hard-limits what the system will pay for any single impression and completely restricts delivery when the auction price exceeds the cap. Lowest cost is the default. Cost cap and bid cap are constraint-based alternatives for specific use cases.
Why is lowest cost the right default for most accounts?
Lowest cost is the right default because Meta's auction system has more data than any media buyer about what each impression is worth and how to allocate budget across audiences. In accounts with clean signal (healthy CAPI match rate, enough conversions to clear learning phase, non-saturated audience), lowest cost scales faster and often lands at a better CPA than cost cap because Meta is willing to pay higher per-impression bids on the right audiences, which cost cap would restrict. The auction optimisation does the CPA management work the buyer thinks cost cap is doing — just better. In accounts above the roughly €5k monthly Advantage+ threshold, lowest cost combined with automation is the default winning configuration.
When does cost cap actually beat lowest cost?
Cost cap beats lowest cost in two specific cases. First, when the account has a real CPA ceiling driven by business constraints — margin thresholds, LTV-to-CAC ratios, or client-mandated performance floors — and the ceiling is non-negotiable. In those cases, constraining the auction prevents the campaign from spending into unprofitable territory even when Meta would scale further at a worse CPA. Second, when the account is in a budget-stable scaling phase where predictability of unit economics matters more than growth velocity. For most accounts in growth mode, cost cap underperforms lowest cost by 10-20 percent on volume and often by 5-10 percent on blended CPA once the holdout is measured.
Why does bid cap require data most accounts do not have?
Bid cap requires the buyer to know the maximum they can afford to pay for a single impression — which requires knowing the expected conversion rate from that impression, the expected revenue from that conversion, the margin on that revenue, and the LTV multiple that justifies spending above the first-order margin. Most mid-market accounts do not have this data at the per-impression level and are guessing when they set a bid cap. The guess is almost always too conservative, which starves the auction of delivery, or too aggressive, which produces the same result as lowest cost but worse because the cap occasionally blocks winning auctions. Bid cap works for sophisticated DTC and B2B accounts with per-cohort LTV models and real unit economics. For everyone else, it loses to lowest cost in every realistic test.
What is the common mistake buyers make with cost cap?
The common mistake is reaching for cost cap when the actual problem is creative fatigue, audience saturation, or signal quality. A campaign running at a CPA 30 percent above target does not need a cost cap at target — it needs a creative refresh or a wider audience. Adding a cost cap to a campaign that cannot deliver below the cap just restricts delivery, which produces zero conversions at the target CPA, which is worse than the original 30-percent-above-target delivery. Cost cap is not a diagnostic tool. It is a constraint for campaigns that could deliver below the cap but need to be forced to. Audience saturation shows up in the frequency-to-CPA curve before it shows up in CPA, and the right fix is almost always upstream of the bid strategy.
What is the right bid strategy by account stage?
1) New account, first 90 days: lowest cost regardless of category. The account needs learning signal before any constraint strategy can work. 2) Growth-mode scaling account: lowest cost continues to win unless a hard business CPA ceiling exists. 3) Mature account with stable unit economics and real CPA ceiling: cost cap for the campaigns bound by the ceiling, lowest cost for everything else. 4) Sophisticated DTC or B2B with per-cohort LTV data: bid cap on specific audiences where LTV is quantified, lowest cost on exploration audiences. 5) Any account below 50 conversions per week on an ad set: lowest cost. Constraint strategies starve low-volume ad sets of the delivery they need to clear learning.
How do you audit whether your bid strategy is right for the account?
1) List every active campaign and the bid strategy it uses. 2) For every cost cap campaign, ask: is there a specific business reason the CPA is capped at this number? If no, test switching to lowest cost. 3) For every bid cap campaign, ask: do we have per-impression economic data justifying this bid? If no, test switching to lowest cost. 4) For every lowest cost campaign that is scaling well, leave it alone. 5) Run paired tests (28 days, matched budget, same creative and audience) when a switch is proposed. 6) If lowest cost wins or ties on blended CPA, adopt it. If the constraint strategy wins, keep it. In Advize's experience, roughly four in ten cost cap campaigns are running cost cap without a real business need and would scale better on lowest cost.
Conclusion
Bid strategy is a function of account stage and data maturity, not preference. Lowest cost wins at scale and in the first 90 days of any new account. Cost cap wins only when CPA constraint is a real business need, not a comfort preference. Bid cap wins only when the account has unit-economics data sharp enough to beat Meta's auction model, which is rarer than most teams admit. The practical next step is to audit which strategy each campaign is actually running and ask whether the business need matches the strategy choice.