Updated October 2026 · By Farrukh Moin Ahmed, digital marketing expert
Short answer: For Meta Ads campaign structure on a small or mid-sized budget, one purchase-optimized campaign covering every funnel stage usually beats separate prospecting, retargeting and traffic campaigns, because Meta gets more purchase data in one place. When I consolidated Woda Bags’ three Meta campaigns into one in late April 2026, cost per purchase fell from $49.07 to $25.69 and ROAS rose from 4.34 to 10.01.
Key takeaways
- Woda’s traffic campaign bought 14,587 cheap landing page views and no attributed purchases.
- Retargeting alone was the most expensive way to get a sale, at $87.87 per purchase.
- One all-funnel campaign since late April 2026: 41 purchases, 10.01 ROAS, $25.69 per purchase.
- Consolidate when each campaign gets well under about 50 purchases a week.
How was Woda Bags’ Meta account set up before?
A common way to run Meta Ads for an online store is one campaign per funnel stage: prospecting for new people, retargeting for visitors, sometimes a traffic campaign on top to fill the retargeting pool. It looks organised. On a small budget, it can quietly make every sale more expensive.
Woda Bags is a US bag brand. I took over its Meta Ads in late April 2026. Before that, the account had three campaigns, each with its own budget and its own goal:
| Campaign | Optimized for | Spend | Purchases | Purchase value | ROAS | Cost per purchase |
|---|---|---|---|---|---|---|
| Prospecting traffic | Landing page views | $1,585 | 0 | $0 | n/a | n/a |
| Retargeting | Purchases | $791 | 9 | $1,723 | 2.18 | $87.87 |
| Prospecting + retargeting | Purchases | $6,569 | 141 | $30,184 | 4.59 | $46.59 |
All figures in this article are Meta-reported, using 7-day click and 1-day view attribution, as of October 2026.
What was each campaign actually doing?
The traffic campaign bought cheap clicks that didn’t buy
The traffic campaign looked efficient on the surface. It reached people at a $5.90 CPM, had a 9.21% click-through rate, and bought 14,587 landing page views at about $0.11 each. But it was optimized for landing page views, so Meta found people likely to click, not people likely to buy. Those 16,262 link clicks produced 68 adds to cart and no attributed purchases.
This is the most common trap in ecommerce Meta accounts. Meta does exactly what you ask. Ask for cheap visits and you get cheap visitors.
The retargeting campaign was too small to learn
The retargeting-only campaign spent $791 for 9 purchases. A retargeting audience for a small brand is limited by how many people visit the site, so this campaign had very few chances to convert each week. Few conversions mean slow learning, and it ended up the most expensive way to get a sale, at $87.87 per purchase.
The combined campaign worked best
The campaign that targeted prospecting and retargeting audiences together did best: 141 purchases at $46.59 each and a 4.59 ROAS. It had the most budget and the most purchase data to learn from. That was the clue. The account performed best wherever Meta had the most conversion signal in one place.
What did I change?
I replaced the three campaigns with one. It targets every stage of the funnel at once, optimizes for purchases, and runs on a $9 daily budget. Instead of three campaigns each learning from a slice of the data, one campaign learns from all of it, and Meta can move spend to whichever audience is converting that week.
| Earlier purchase campaigns | All-funnel campaign (since late April 2026) | |
|---|---|---|
| Spend | $7,360 | $1,053 |
| Purchases | 150 | 41 |
| Purchase value | $31,906 | $10,550 |
| ROAS | 4.34 | 10.01 |
| Cost per purchase | $49.07 | $25.69 |
“Earlier purchase campaigns” combines the retargeting and the prospecting-plus-retargeting campaigns. The traffic campaign is left out, because it wasn’t optimized for purchases. Including it would make the earlier results look worse, not better.
What do the funnel numbers show?
The campaign-level numbers tell an interesting story about what each setup was buying:
| Campaign | CPM | CTR | Link clicks | Adds to cart | Purchases |
|---|---|---|---|---|---|
| Prospecting traffic | $5.90 | 9.21% | 16,262 | 68 | 0 |
| Retargeting | $27.61 | 3.06% | 517 | 41 | 9 |
| Prospecting + retargeting | $21.86 | 2.84% | 4,876 | 458 | 141 |
| All-funnel (mine) | $25.98 | 3.45% | 813 | 60 | 41 |
The all-funnel campaign pays more per thousand impressions than the traffic campaign, about four times as much. That’s expected. When you optimize for purchases, Meta bids for people more likely to buy, and those people cost more to reach. Cheap impressions and high click-through rates are not the goal. Cheap purchases are.
Why did consolidating work?
- More signal in one place. Meta’s delivery system learns from optimization events. Its help documentation describes the learning phase as needing about 50 optimization events within seven days. Three small campaigns each saw a fraction of that. One campaign sees all of it.
- Budget follows performance. Fixed budgets per stage force spend into a stage even in a week when it isn’t converting. One campaign lets spend flow to whichever audience is buying.
- The goal matches the business. Optimizing for purchases, rather than landing page views, points delivery at people likely to order.
- No self-competition. Two of the three earlier campaigns targeted prospecting audiences, so they were likely bidding for some of the same people. One campaign can’t compete with itself.
What doesn’t this prove?
It’s worth being clear about the limits, because case studies that oversell are easy to spot:
- Different time periods. The campaigns didn’t run at the same time, so seasonality and other changes may play a part.
- Different budgets. A $9 daily budget is small. Small budgets often show higher ROAS because Meta reaches the most likely buyers first. Results usually come down as spend scales.
- Platform attribution. Meta’s figures include view-through conversions, and Meta and Google can both claim the same order. Woda also runs Google Ads, so neither platform’s ROAS should be read alone.
So read this as strong evidence for small and mid-sized budgets, not a law. Once each funnel stage has enough purchases of its own, separate campaigns can make sense again.
Should your account use one campaign or several?
A simple test: estimate how many purchases your Meta budget produces each week, then divide by the number of campaigns or ad sets you’re running. If each one gets well under the roughly 50 optimization events a week the learning phase needs, consolidate. If each comfortably clears it, separate campaigns can give you more control.
| Weekly purchases per campaign | What I’d do |
|---|---|
| Well under 50 | One all-funnel purchase campaign |
| Around 50 | One campaign, possibly a separate one for a key product line |
| Comfortably over 50 per stage | Separate prospecting and retargeting, judged on their own jobs |
The budget side of that decision, including how to judge each stage once they are separate, is in Prospecting vs retargeting: how to split your Meta Ads budget.
How do you set up an all-funnel purchase campaign?
- Fix tracking first. Meta Pixel plus the Conversions API, with purchase values checked against store orders. Meta processes server events like Pixel events, so it sees purchases browsers miss.
- Use a sales objective optimized for purchases. Not traffic, not landing page views, and not add to cart unless purchase volume is genuinely too low.
- Keep targeting broad. Let the campaign reach new and returning people together, rather than splitting them into separate budgets.
- Mix creative for both stages. Some ads introduce the brand to new people; others answer the objections of people who’ve already visited.
- Test creative, not structure. Once the structure is stable, improvements come from testing hooks, formats, products and offers one at a time.
On a small budget, give Meta one campaign with enough purchases to learn from.
For the full approach, see Meta Ads for ecommerce and fashion and accessories brands. Woda’s Google Ads results are on the case study.
What creative should an all-funnel campaign run?
Once prospecting and retargeting share one campaign, creative has to do the job that separate audiences used to do. Meta decides who sees which ad, so the ad set needs a mix of messages that work for someone who has never heard of the brand and for someone who left a product page yesterday.
A practical way to cover both is to brief creative by the question it answers, not by audience:
- What is this? Product-led ads that show the bag, how it is carried and what fits inside. These introduce the brand to cold audiences.
- Why this one? Ads built around a single reason to choose the product: material, size, price point, a specific use such as travel or work.
- Why now? Ads that answer the hesitation of a returning visitor: shipping, returns, sizing, or a new colour.
- Who else uses it? Real customer photos and reviews, where the brand has them. Never invented ones.
Keep three to six live ads at a time and replace the weakest one when you add a new one, rather than piling ads into the ad set. Too many ads split the same small budget the way too many campaigns did. Judge each ad on cost per purchase and purchase volume, not click-through rate alone. The traffic campaign above showed what a high click-through rate is worth when nobody buys.
For fashion and accessories brands in particular, catalog ads (Advantage+ catalog ads) are worth a slot in the mix. They show the exact products a returning visitor looked at, which covers most of what a separate retargeting campaign used to do. More on that in Meta Ads for fashion and accessories brands.
How do you scale a small all-funnel campaign?
A $9 daily budget is a starting point, not a ceiling. The risk when scaling is that every large budget change sends delivery back into a period of re-learning, and results wobble while it settles. A careful approach looks like this:
- Wait for stable results first. Two to three weeks of consistent cost per purchase is a better signal than one great week.
- Raise budgets in steps. Increases of around 15 to 20 percent every few days are a common rule of thumb among practitioners. It is not a Meta rule, but it keeps changes small enough that delivery adjusts gradually.
- Expect ROAS to fall as you scale. A small budget reaches the most likely buyers first. As spend grows, Meta has to reach less certain buyers, so the cost per purchase rises. Decide the lowest ROAS that is still profitable for the business before you start, so a dip does not cause panic.
- Add creative as you add budget. More spend means each ad is shown more often. Fresh creative keeps frequency from wearing out the audience.
- Revisit structure only when volume justifies it. When the campaign comfortably clears around 50 purchases a week, separating a stage or a product line becomes an option again, not before.
How do Meta and Google Ads work together for a store like Woda?
Woda Bags runs Google Ads as well as Meta Ads, and the two do different jobs. Google Ads captures demand that already exists: people searching for a bag or comparing products on Shopping. Meta Ads creates demand by putting products in front of people who were not searching yet. Some of those people later search on Google, which is why the two platforms often both claim the same order.
That overlap is why I read results in three layers rather than trusting either platform alone:
- Platform view. Each platform’s own ROAS, used to optimize inside that platform.
- Store view. Total orders and revenue from the store, compared with total ad spend across both platforms.
- Trend view. Whether total orders move when one platform’s spend changes. If Meta spend goes up and branded Google searches rise too, Meta is likely feeding Google.
How to split budget between the two, and when each should lead, is covered in Google Ads vs Meta Ads for ecommerce. The Google side of Woda’s account, including Shopping and Performance Max, is in the Woda Bags case study and on Google Ads for ecommerce. If your numbers disagree across platforms, start with analytics and tracking before changing any campaign.
If your Meta account is split across too many campaigns, I work as a Meta Ads expert on structure, creative and tracking.
Frequently asked questions
Is one Meta campaign better than separate prospecting and retargeting?
On small and mid-sized budgets, often yes, because one campaign gets enough purchases to learn from. Once each stage can reach about 50 purchases a week on its own, separate campaigns can give more control.
What happened when Woda Bags consolidated its Meta campaigns?
Its all-funnel campaign, running since late April 2026, produced 41 purchases and $10,550 in purchase value at a 10.01 ROAS and $25.69 per purchase, compared with $49.07 per purchase across the earlier purchase campaigns.
Why did the traffic campaign get no purchases?
It was optimized for landing page views, so Meta found people likely to click, not people likely to buy. It bought 14,587 landing page views at about $0.11 each, with no attributed purchases.
How many conversions does Meta need to exit the learning phase?
Meta’s help documentation describes about 50 optimization events within seven days of the last significant edit.
Will a 10 ROAS hold as the budget grows?
Usually not entirely. Small budgets reach the most likely buyers first, so return tends to come down as spend scales. Budgets should rise in steps, watching cost per purchase against breakeven.
