Meta reference

Meta bid strategies for ecommerce

Which bid strategy to choose in Meta Ads Manager, what number to set, and what to check before you change it.

Checked against Meta's Business Help Centre on 19 October 2026. A Meta or Google badge links to the help page that says it. Onward guidance marks our own rules from running client accounts. They are how we work, not platform policy.

The five bid strategies at a glance

Meta has three types of bid strategy: spend-based (Highest volume, Highest value), goal-based (Cost per result goal, ROAS goal) and manual (Bid cap). Meta

Strategy	What you set	Cost control	Delivery risk	Best for
Highest volume	Budget only	None	Very low	Testing, building a baseline, scaling a proven campaign
Highest value	Budget only	None	Very low	Ecommerce that sends purchase values, before a ROAS goal
Cost per result goal	A target average cost per result	Soft average	Moderate	Proven campaigns with a cost per sale or lead the business can't go above
ROAS goal	A target average return, such as 2.50	Soft average, on revenue	Moderate to high	Ecommerce with clean purchase values and order values that vary widely
Bid cap	The most Meta can bid in each auction	Hard ceiling on the bid	High	Specific cases only, covered below

"What you set" is Meta's definition. The other columns are Onward guidance.

Where to find them. With Advantage+ campaign budget you pick the bid strategy at campaign level. In that menu, Highest volume and Highest value share one option, Highest volume or value. You get Highest value when the ad set's performance goal is Maximise value of conversions. Meta With ad set budgets there is no strategy menu: enter a cost control amount in the ad set's Performance goal section and Meta switches to Cost per result goal. Leave it blank and you get Highest volume. Bid cap sits under Show more options. Meta

Old names you'll still see

Meta renamed three bid strategies and is still rolling the new names out, so older guides, reports and some accounts use the old ones. Meta

Name now	Old name
Highest volume	Lowest cost
Cost per result goal	Cost cap
ROAS goal	Minimum ROAS
Highest value	No change
Bid cap	No change
Advantage+ campaign budget	Campaign budget optimisation (CBO) Meta
Advantage+ sales campaigns	Advantage+ shopping campaigns (ASC) Meta

Choosing a strategy

These are Onward's rules. Work down the list and stop at the first one that fits. Onward guidance Meta's own advice points the same way: build a baseline on a spend-based strategy for at least two weeks before you add a goal. Meta

No proven baseline on this optimisation event?Fewer than 50 conversions in the learning window, or the campaign is less than 2–4 weeks old.

→

Highest volume or valueNo cost control yet. Stop here.

Sending purchase values through the pixel or Conversions API, with several hundred valued purchases and order values that vary widely?A store that pays the same to win a £30 order and a £300 order is the case ROAS goal is built for.

→

ROAS goalSet it below the ROAS you need, then tighten

Would a spike in cost per result damage profit at your current spend, not just muddy a test?

→

Cost per result goalStart 10–15% above your actual average

Do you know the market rate for this auction, need to protect a small and expensive retargeting audience, or want to buy cheaper inventory and accept less reach?Each of these needs real knowledge of the auction.

→

Bid capCaps the bid, not the cost

None of the above, or not sure?

→

Highest volume or valueThe safe default

The most common mistake: adding a cost control too early, before Meta has enough conversions to work with. It starves delivery and gives you test results you can't trust. Onward guidance

Using incremental attribution? Meta doesn't allow bid controls with it, so plan on Highest volume or Highest value. Meta

Each strategy in detail

Highest volume

What it does: Spends your budget to get as many results as possible. You don't set a cost or return target, and cost per result rises and falls with competition in the auction. Meta
Use it when: You're testing creative, audiences or offers and need data more than cost control. You're building a baseline before adding a goal. Or you're scaling a proven campaign and want volume at whatever the market costs. Onward guidance
Don't use it when: The business has a cost per sale it can't go above at your current spend. Onward guidance
Watch: Meta spends the full budget whatever each result costs, so on a bad day cost per result can run at 3–4x your average. That's how the strategy works, not a fault. Judge it over the week. Onward guidance

Highest value

What it does: Spends your budget and bids for the highest purchase values. It needs the Maximise value of conversions performance goal, a working pixel or SDK, and a good spread of values across your products. Meta
Use it when: You send purchase values and basket sizes vary. Meta suggests testing it first without a ROAS goal, for at least three weeks, at 50 or more conversions a week. Meta
Don't use it when: Purchase values are missing, patchy or all much the same. Onward guidance
Watch: Expect a higher cost per result and fewer conversions than on Highest volume, because Meta is paying for more valuable buyers. Meta says learning usually takes at least a week and performance typically peaks after two, so don't judge it early. Meta

Cost per result goal

What it does: Aims to keep your weekly average cost per result near the amount you set while getting as many conversions as it can. It bids above and below your amount auction by auction, and may spend less than your budget when auctions get expensive. Meta Meta doesn't guarantee results will match the goal. Meta
Use it when: You have a proven cost per result from 2–4 weeks of history, 50 or more conversions in the learning window, and a spike in cost would hurt profit at your current spend. Onward guidance Meta says it works best at 50–100 or more conversions a week. Meta
Don't use it when: The campaign is new, has no conversion history, or gets few conversions a week. With nothing to aim from, Meta either under-delivers or overshoots. Onward guidance
Watch: Judge it on weekly averages, not daily results, and wait at least 7 days after a change before you judge it. Meta Keep the daily budget at least 5x the goal. Meta

ROAS goal

What it does: Aims to keep your weekly average return on ad spend near the number you set while maximising conversion value. It may spend less than your budget to protect the target. Meta You enter it as a number between 0.001 and 1,000.00. 1.00 is breakeven, and 2.50 means around £250 of purchase value for every £100 spent. Meta
Use it when: You send purchase values through the pixel or Conversions API, have several hundred valued purchases, and order values vary widely. Onward guidance It only works with the Maximise value of conversions performance goal. Meta
Don't use it when: You optimise for leads, or for purchases without values, or your value data is new or patchy. Onward guidance
Watch: The same weekly rhythm as Cost per result goal: weekly averages, at least 7 days between changes. Meta If ROAS goal is greyed out, the ad set isn't optimising for value. That isn't a bug. Onward guidance

Bid cap

What it does: Sets the most Meta will bid for each estimated result in every auction. It caps the bid, not what you pay. Meta In Meta's example, a USD 2 bid cap gives a USD 5 cost per result, because not every estimated conversion happens. Meta
Use it when: Meta says it's for advertisers who can predict conversion rates well enough to work out the right bid. Meta We only use it for a specific reason: a known market rate in a crowded category, a small and expensive retargeting audience to protect from high CPMs, or a deliberate choice to buy cheaper inventory and accept less reach. Onward guidance
Don't use it when: You want to control cost per result. Meta says Cost per result goal does that better. Meta
Watch: A cap that's too low shows Bid limited in the Delivery column and can stop spend. Meta Set it below what the auction needs and spend can drop to zero for the day. Onward guidance

What each strategy needs

Strategy	Performance goal	Data Meta recommends	Spends the full budget?
Highest volume	None stated	None. This is how you build a baseline	Aims to Meta
Highest value	Maximise value of conversions	100 conversion events with at least 5 distinct values in the past 14 days (recommended) Meta	Aims to Meta
Cost per result goal	Maximise number of conversions. Some Meta pages call it Maximise volume of conversions Meta	50–100 or more conversions a week	Not always Meta
ROAS goal	Maximise value of conversions Meta	As Highest value, plus 50–100 or more conversions a week	Not always Meta
Bid cap	None stated	None stated. You need to be able to predict conversion rates	Not if the cap is too low Meta

Value optimisation needs a conversion event that sends a value and currency through an active integration, such as the Conversions API or the Meta pixel. Meta Another Meta page says the Conversions API is required. Meta We run both. Onward guidance It can take a few weeks after you meet the requirements before Maximise value of conversions becomes available. Meta
iOS 14.5 and later: if a Cost per result goal or ROAS goal ad set will reach these devices, schedule it to run for at least three full days. Meta

Setting the number

The right strategy with the wrong number still stalls a campaign. Meta's starting rules and ours are close. Where they differ, both are shown. Measure your averages on the attribution setting the ad set optimises on. Meta

Situation	Meta's rule	Onward's rule
Cost per result goal	Start 10–20% above your recent average cost per result. A USD 20 average over 30 days gives a USD 22 goal. Meta	Start 10–15% above your actual average, never at the cost you'd like to pay. Below your real average, Meta can't find enough inventory at that price and under-delivers. Onward guidance
ROAS goal	Start 10–20% below your recent average ROAS: an average of 1.00 gives 0.80–0.90. Meta Meta also describes the goal as the lowest ROAS you can accept. Meta	Start below the ROAS the business needs, then tighten in steps. If you need 3.0, start at 2.5. A goal at or above what you need, set too early, makes Meta so selective that spend stalls. Onward guidance
No history	Run Highest volume or Highest value for at least two weeks and use the result as your starting point. Meta Meta	2–4 weeks on Highest volume or value, and 50 or more conversions in the learning window, before any goal. Onward guidance
Switching from bid cap	Start from your historical cost per result, not the cap value. The cap is a bid, not an average cost. Meta	Same.
Daily budget	At least 5x your Cost per result goal. A USD 5 goal needs at least USD 25 a day. Meta	Same.

Work out the ROAS your business needs in the ROAS Target Calculator.

Starting goal calculator

Enter your recent average and weekly conversions. It applies Meta's starting rule, and Onward's range for Cost per result goal.

Strategy

Recent average cost per result (£)

Conversions per week in this ad set

Enter your numbers.

Meta's rules used: start 10–20% looser than your recent average, aim for 50–100 or more conversions a week, keep the daily budget at least 5x the Cost per result goal. Meta Meta Onward's Cost per result range is 10–15% above average. Onward guidance

Campaign not spending?

When a campaign with a cost control under-delivers, the control is almost always the cause, not the creative or the audience. Check in this order. Onward guidance

Bid cap below what the auction needs. Meta stays out of those auctions, so spend is close to zero rather than just expensive. The Delivery column may show Bid limited. Meta
Cost per result goal at or below your real average cost per result.
ROAS goal at or above the ROAS you need, set before the campaign could support it.
A goal on a campaign with no conversion history, so Meta has nothing to aim from.

The fix is the same shape each time: loosen the control, or remove it and go back to Highest volume to rebuild a baseline. Onward guidance Changing bid strategy is a significant edit and restarts learning, so time it carefully if you're mid-scale. Meta

Two quick checks in Ads Manager: hover over a Learning limited status, and if a low bid or cost control is the cause, a tooltip says so. Meta If more budget doesn't bring more results, the goal is the limit. Meta suggests loosening it, refreshing creative or widening the audience. Meta

Common mistakes

1Adding a goal before there's a baseline

A Cost per result goal or ROAS goal needs history to aim from. Meta says run Highest volume or Highest value for at least two weeks first. Meta Put a goal on a new or thin ad set and you get a campaign that won't spend and a test result you can't use. Onward guidance

Do

Run spend-based until you have 2–4 weeks of data and 50 or more conversions in the learning window. Then add the goal.

Don't

Launch a new offer on a Cost per result goal copied from a different campaign's history.

2Treating a goal as a cap

Cost per result goal and ROAS goal are weekly averages. Meta bids above and below your number and doesn't guarantee to hit it. Meta The old names, Cost cap and Minimum ROAS, made them sound like hard limits. The only hard ceiling Meta offers is Bid cap, and that caps the bid, not the cost. Meta

How we explain it to clients: a goal is a steering wheel, not a brake. Onward guidance

3Setting the goal at the number you want

Start from the number you actually get. Meta says start 10–20% looser than your recent average. Meta Set it at your ideal cost, or at a ROAS you don't yet achieve, and Meta becomes so selective that spend stalls. Onward guidance

Fix: start from history, then move in 5–10% steps and change bids no more than once a week. Meta Use the starting goal calculator.

4Judging goal-based campaigns day by day

Meta judges Cost per result goal and ROAS goal on weekly averages and says wait at least 7 days after a change. Meta A large change to the goal amount can also restart learning. Meta

Fix: read results weekly, over a window with at least 50–100 conversions. Meta Make one change, then wait a week.

5Reading Bid cap as a cost per result

Meta's example: a USD 2 bid cap and a USD 10 daily budget buy five impressions at USD 2 each. Two convert, so the cost per result is USD 5, even though no bid went over the cap. Meta Set the cap too low and the ad set shows Bid limited and may stop spending. Meta

Fix: moving from Bid cap to Cost per result goal? Start from your historical cost per result, not the cap. Meta

6Pausing the "expensive" ad set under Advantage+ campaign budget

With campaign budget on, Meta spends where the next result is cheapest, not by each ad set's average. In Meta's worked example the campaign gets 12 results at an average of USD 2.50. Turn off the ad set with the highest average cost and it gets 11 results at USD 2.73. Meta Meta also advises against pausing ad sets with low delivery, and against stacking ad set spend limits on top of a cost control. Meta

Fix: judge results at campaign level: total results and average cost per result for the whole campaign. Meta

7Comparing numbers across attribution settings

Meta says base your goal on results within the ad set's attribution setting Meta, and that results can't be compared in one table across ad sets on different attribution models. Meta Since March 2026 Meta has been changing click-through attribution to count link clicks only, so click-through results can drop for that reason alone. Meta

Fix: keep one attribution setting per comparison, and note it on every report. Onward guidance

Changing and scaling a goal

Meta says run a goal-based campaign for at least three weeks before scaling. Once it's meeting expectations, raise the budget gradually and change the goal only after a week. Meta Meta

Situation	Cost per result goal	ROAS goal
Budget left over each week (the goal is limiting spend)	Raise the goal 5–10%	Lower the goal 5–10%
Underspending and missing the target	Combine ad sets until they reach 50–100 or more conversions a week	Combine ad sets until they reach 50–100 or more conversions a week
Budget fully spent, results on target	Raise the budget. Adjust the goal after a week if needed	Raise the budget. Adjust the goal after a week if needed
Budget fully spent, results well off target	Lower the goal 5–10%	Raise the goal 5–10%

Source: Meta's best practices for each goal. Meta Meta

Budget steps: we raise budgets 15–20% at a time, at least 72 hours apart. Onward guidance Meta publishes no step size. It says a small budget change is unlikely to restart learning and a large one may. Meta

Budgets and pacing

Rule	On a £100 daily budget
A daily budget is an average over a Sunday-to-Saturday week. Meta	Judge spend by the week, not the day
Meta can spend up to 75% over the daily budget on a single day. Meta	Up to £175 in a day
Spend never goes over 7x the daily budget in a Sunday-to-Saturday week. Meta	No more than £700 in that week
With ad set budget sharing on, up to 20% of the budget can also go to other ad sets. Meta	Up to £210 in a day and £840 in a week
If you start or change a budget mid-week, Meta prorates the rest of the week and allows up to 25% of the daily budget on top. Meta	Up to £25 a day of extra flexibility for the rest of that week
A lifetime budget is a hard cap, not an average. Meta	Use one when a client needs a fixed ceiling for a period Onward guidance

Compare with Google: Google allows up to 2x the average daily budget on a day and 30.4x in a month. Google Same idea, different maths. Run both side by side in the Budget Pacing Calculator.

Advantage+ campaign budget and Advantage+ sales campaigns

One bid strategy per campaign. With Advantage+ campaign budget you choose the bid strategy at campaign level and set the amount in each ad set. You can't mix strategies in one campaign. Use a separate campaign instead. Meta Meta
It's on by default for Sales, Leads and App promotion campaigns. Once you switch it on or off, you have to wait two hours before switching it again. Meta
Judge it at campaign level: total results and average cost per result for the campaign, not each ad set. Meta
Leave it room to work. Don't pause ad sets with low delivery, use ad set spend limits sparingly (especially with a cost control), and make changes in one go so the campaign re-enters learning fewer times. Meta
Campaign-level changes reach every ad set. Switching the campaign's bid strategy, or a large change to the campaign budget, can send several ad sets back into learning at once. Meta

Advantage+ sales campaigns aren't a bid strategy. A Sales campaign counts as Advantage+ on when Advantage+ campaign budget, Advantage+ audience and Advantage+ placements are all in use. Meta Any of the bid strategies on this page can run inside one. Meta recommends Highest volume or value. Meta

The learning phase

An ad set leaves learning once its delivery is stable, which Meta says usually takes about 50 results in the week after its last significant edit. If that looks unlikely, the Delivery column shows Learning limited. Meta says that isn't a penalty. Meta Meta To check progress, add the Last significant edit and Results columns in Ads Manager. Meta

Restarts learning	May restart, depending on size	Doesn't restart learning
Any targeting change. Any creative change. Changing the optimisation event. Adding a new ad. Pausing for 7 days or more. Changing bid strategy.	The Cost per result goal, ROAS goal or bid cap amount. The budget. The ad set spending limit. Meta's example: USD 100 to USD 101 is unlikely to matter, USD 100 to USD 1,000 may.	Advantage+ campaign budget moving spend between ad sets. An edit to one ad set, for the other ad sets in the campaign. Adding a new ad set, for the others.

Source: significant edits Meta. Check a specific change against Google in the Learning Reset Checker.

Fixing Learning limited

Meta's fixes: combine ad sets and campaigns, widen the audience, raise the budget, raise the bid or cost control, or optimise for an event that happens more often, such as add to basket instead of purchase. Meta

How we work: Meta advises making changes in one go so learning restarts fewer times. Meta We still change one thing at a time when we need to know what caused a result. Batch the housekeeping, isolate the change you want to read. Onward guidance

Value rules and profit optimisation

Value rules raise bids by up to 1,000% or lower them by up to 90% for chosen ages, genders, locations, operating systems, device platforms, placements and conversion locations. Meta Meta's pages disagree on whether they work with ROAS goal: the value rules page says they aren't available with ROAS goal or Bid cap, while the bid strategies page says you can add them on top of goal-based bidding. Check in the ad set before you rely on it. Meta Meta
Bid multipliers have to be replaced by value rules by 2027. Meta
Purchase profit: Meta can optimise for profit instead of revenue. Choose Maximise value of conversions with Purchase profit as the conversion value, then send order-level profit with the Purchase event or product-level margin in your catalogue. Meta asks for at least three distinct positive profit values, profit that isn't a fixed share of revenue, and a catalogue match rate above 90% if you use the catalogue. Send Purchase events in real time: events held back while profit is calculated may not count. Meta This is Meta's built-in version of the profit bidding described in the POAS section for Google.

Meta and Google compared

If you also run Google Ads, each Meta strategy has a close Google equivalent. The differences are where mistakes happen.

Meta	Nearest Google	What's the same	What's different
Highest volume	Maximise conversions	Both try to spend the budget for the most conversions, with no cost target. Meta Google	Budget flexibility: Meta allows up to 175% of the daily budget in a day and 7x in a Sunday-to-Saturday week. Google allows 2x in a day and 30.4x in a month. Meta Google
Highest value	Maximise conversion value	Both spend the budget chasing value. Meta Google	Meta recommends 100 conversion events with 5 distinct values in 14 days before value optimisation. Meta Google's page lists no minimum for Search and Shopping, only conversion tracking with values. Google
Cost per result goal	Target CPA	Both aim at an average cost, with some conversions above it and some below. Meta Google	Meta averages weekly and recommends 50–100 or more conversions a week. Google says you can start Target CPA with no conversion history and recommends your 30-day average as the target, where Meta says start 10–20% above it. Google Meta
ROAS goal	Target ROAS	Both aim at an average return. Meta Google	Meta takes a decimal (4.00), Google a percentage (400%). Meta needs the Maximise value of conversions goal. Google needs at least 15 conversions in 30 days on Search and Shopping. Google
Bid cap	Maximum CPC bid limit (on Maximise clicks)	Both cap the bid. Meta Google	Meta caps the bid per estimated result, so per purchase if you optimise for purchases. Google's limit is per click, and bid adjustments apply on top of it. Google
Learning phase	Learning period	Both use about 50 conversions as a yardstick. Meta Google	Meta counts per ad set, in the week after the last significant edit, and a large goal change can restart it. Google counts per bid strategy (up to about 50 conversions or 3 conversion cycles) and says changing a target doesn't trigger Learning. Google Google
Attribution setting	Conversion window	Both decide which conversions count.	Meta's default is 7-day click, 1-day view and 1-day engagement Meta, and the ad set's setting also steers delivery. Meta Google's click-through window defaults to 30 days. Google
Purchase profit	Profit as the conversion value	Both can bid on profit instead of revenue.	Meta has a built-in Purchase profit value. Meta On Google you send profit as the conversion value yourself (see POAS).

Recent changes

When	Change	What to do
From March 2026, rolling out	Click-through attribution counts link clicks only. Shares, saves and other taps moved to engage-through attribution, and a video engaged view dropped from 10 seconds to 5. Meta	Click-through results can fall with no real change in performance. Re-base any goal set on the old definition. Onward guidance
Rolling out, undated	Bid strategies renamed: Highest volume, Cost per result goal, ROAS goal. Meta	Use the old names table when reading older guides.
From May 2025	Advantage+ shopping campaigns became Advantage+ sales campaigns: an ordinary Sales campaign with Advantage+ campaign budget, audience and placements on. Meta Meta	Any bid strategy can run inside one.
By 2027	Bid multipliers must be replaced by value rules. Meta	If you use bid multipliers, plan the move.
Available now	Incremental attribution optimises for conversions Meta predicts the ad caused. It allows no bid controls, and you can't change the attribution model after you publish. Meta	Test it in its own campaign. Onward guidance

Sources

Every page this reference relies on, checked on 19 October 2026.

Meta Business Help Centre

Other Meta pages

Simplifying ad measurement for a social-first world (Meta for Business, 3 March 2026)
Advantage+ shopping campaigns (rename notice)
Advantage+ campaign experience for sales and app (Meta for Developers blog)
Advantage+ campaigns (Meta for Developers)

Google Ads Help

Anything marked Onward guidance is our own working practice from managing client accounts. If you'd like to go through it against your own account, get in touch.