Facebook ads for SaaS work, but the playbook changed significantly in late 2025, and most SaaS founders are still running the old one. Meta’s Andromeda system, which completed its global rollout in October 2025, shifted ad delivery from audience-first to creative-first. The campaigns that worked in 2023 and 2024 are now actively fighting the algorithm instead of working with it.
This post covers what actually drives results for B2B SaaS companies running Meta ads in 2026, including what changed with Andromeda, what stayed the same, and how to structure your campaigns for the system that exists today.
Running Facebook ads for your SaaS with inconsistent results?
We build and manage Meta ad campaigns specifically for B2B SaaS companies, including free trial funnels, demo request campaigns, and retargeting systems built around your product, your buyer, and the post-Andromeda algorithm.
The Quick Take: Facebook Ads for SaaS Before and After Andromeda
| Pre-Andromeda Approach | Post-Andromeda Approach |
|---|---|
| Segment campaigns by funnel stage (awareness, consideration, conversion) | Consolidate campaigns and let creative do the funnel work |
| Build detailed interest and demographic targeting stacks | Run broad targeting and let strong, specific creative filter the audience |
| Test 3 to 5 creative variations per ad set | Supply 10 to 20 genuinely distinct creative concepts per campaign |
| Optimize targeting to reach your ICP | Optimize creative to speak to your ICP — creative is now the targeting |
| Advantage+ is optional or for scaling phase | Advantage+ placements are the default starting point, not an experiment |
The Takeaway: Andromeda shifted Meta from a system you configure to a system you feed. The advertisers winning in 2026 focus less on campaign architecture and more on creative volume, creative quality, and clean conversion signals.
💡 Pro Tip: Andromeda completed its global rollout in October 2025 and now powers all Facebook and Instagram ad delivery. If your campaign structure still resembles what you built in 2023, you are not fighting performance problems. You are fighting the algorithm’s architecture. The fix is not in your bids or your budgets. It is in your creative volume and your campaign consolidation.
Table of Contents
→ What Andromeda Changed for Facebook Ads for SaaS
→ Why Facebook Ads for SaaS Are Different from Every Other Vertical
→ Campaign Structure: Andromeda-First vs. Traditional Layered Funnel
→ What Creative Actually Works for B2B SaaS on Facebook in 2026
→ Targeting for Facebook Ads for SaaS After Andromeda
→ The Metrics That Actually Matter for SaaS Facebook Ads
→ The Biggest Facebook Ads Mistakes SaaS Companies Make Now
→ How Much to Spend on Facebook Ads for SaaS
→ The Bottom Line on Facebook Ads for SaaS in 2026
→ FAQ: Common Questions About Facebook Ads for SaaS
What Andromeda Changed for Facebook Ads for SaaS
Meta’s Andromeda system replaced the old ad retrieval engine and changed how Meta decides which ad to show each user before the auction even begins. The previous system evaluated a limited pool of ads based on your targeting inputs. Andromeda processes hundreds of thousands of ad creatives per user impression in real time, interpreting thousands of behavioral signals to decide which creative each individual user should see.
The practical result for SaaS advertisers is significant. The system no longer asks “who should see this ad?” first. It asks “which ad should this person see?” That reframe changes what you need to optimize. Your creative is now the primary audience signal, not a supplement to your targeting configuration.
Andromeda also introduced sequence learning, which enables Meta’s system to understand where a user sits in the purchase journey. The system can detect whether someone is in awareness, consideration, or conversion mode based on their behavior patterns, and serve creative accordingly. This is the capability that makes consolidated campaign structures viable in a way they were not before. Meta can now do the funnel sequencing work that advertisers used to do manually through campaign architecture.
For SaaS specifically, this creates both an opportunity and a risk. The opportunity: a well-stocked creative library with genuinely distinct concepts, formats, and messages can now reach the right buyer at the right stage without hyper-segmented campaign structures. The risk: if your creative library is thin or repetitive, Andromeda has nothing to work with, and your delivery suffers regardless of how clean your campaign setup is.
💡 Pro Tip: Andromeda evaluates creative similarity strictly. Minor variations like swapping a headline word or changing a button color no longer count as distinct creative. The system requires genuinely different concepts, different hooks, different formats, and different messaging angles. If your five “variations” all lead with the same problem statement and the same visual structure, Andromeda treats them as one creative, not five.
Why Facebook Ads for SaaS Are Different from Every Other Vertical
Facebook ads for SaaS operate on a fundamentally different value exchange than ecommerce or traditional lead gen, and most SaaS companies burn significant budget before they figure this out. In e-commerce, the ad drives a purchase. In lead gen, the ad drives a form fill. In SaaS, the ad drives a signup that starts a relationship that may or may not convert to revenue weeks or months later. That extended conversion timeline changes everything about how you structure, optimize, and measure campaigns.
The SaaS buyer journey on Facebook typically looks like this: a potential customer sees a problem-focused ad, clicks through to a trial or demo page, signs up, experiences the product during an onboarding sequence, and either converts to paid or churns. Meta’s algorithm only sees the signup event and has no visibility into what happens inside your product after that. If your trial-to-paid conversion rate is poor, Meta will keep sending you trial signups at your target CPA while your actual revenue per signup deteriorates. Facebook ads for SaaS require you to close that loop manually through your data infrastructure.
The other key difference is the buyer. B2B SaaS buyers are skeptical professionals evaluating tools against specific workflow problems. They do not respond to lifestyle creative, urgency tactics, or generic benefit statements. They respond to ads that name their exact problem, show a credible solution, and make the next step feel low-risk. The creative and copy approach that works for DTC brands actively repels the SaaS buyer you are trying to reach. Andromeda amplifies this dynamic, because the system rewards creative that generates strong engagement signals from the right people, and weak creative simply does not get delivered.
💡 Pro Tip: Pass your trial signup data back to Meta via the Conversions API and tag signups that convert to paid customers as a separate custom conversion event. Once you have 50 or more paid conversion events, create a lookalike audience from that pool and use it as a targeting signal in your prospecting campaigns. This is the single highest-impact technical setup move for Facebook ads for SaaS. You train Meta’s algorithm to find people who pay, not just people who sign up, and that signal becomes more powerful under Andromeda, not less.
Campaign Structure: Andromeda-First vs. Traditional Layered Funnel
This is where the biggest strategic split exists in SaaS Facebook advertising right now, and the right answer depends on your account’s data density. Andromeda changed the optimal campaign structure for most advertisers, but it did not eliminate situations where a traditional layered approach makes sense. Understanding which model fits your account is more important than following a blanket rule.
The Andromeda-First Structure: Consolidated Campaigns with Creative Doing the Funnel Work
The Andromeda-native approach consolidates your campaign structure and relies on creative diversity to reach buyers at every stage of the funnel. Instead of separate campaigns for awareness, consideration, and conversion, you run one or two campaigns per business objective, load each ad set with 10 to 20 genuinely distinct creative concepts spanning multiple funnel stages, and let Andromeda’s sequence learning serve each user the creative that fits their behavioral context.
This approach works because Andromeda can now detect purchase intent signals and funnel stage at the individual level. Including awareness-focused creative, consideration-focused creative, and conversion-focused creative in the same ad set does not create confusion. It creates options. The system matches the right creative to the right user at the right moment better than manual campaign segmentation ever could.
The structural framework that performs best under Andromeda is a two-campaign split: one testing campaign (ABO, ad set budget) where new creative enters the account and proves itself, and one scaling campaign (CBO, campaign budget optimization) where proven creative gets sustained budget. New ads earn their place in the scaling campaign by generating a threshold of conversions at your target CPA in the testing campaign first.
| Campaign Layer | Purpose and Setup |
|---|---|
| Testing campaign (ABO) | New creative enters here. Ad set budget controls spend per concept. Creative graduates to the scaling campaign after hitting your conversion threshold. |
| Scaling campaign (CBO) | Proven creative runs here. Campaign budget optimization lets Meta shift spend toward the best performers in real time. Broad targeting. Advantage+ placements enabled. |
The Traditional Layered Funnel: When It Still Makes Sense for SaaS
The traditional three-layer structure with separate campaigns for awareness, consideration, and conversion still makes sense in specific SaaS account scenarios, and dismissing it entirely is a mistake. The key variable is data density. Andromeda learns from conversion data. When a SaaS account generates fewer than 50 conversion events per week per campaign, the system does not have enough signal to make reliable creative-matching decisions. In those accounts, some degree of manual funnel segmentation still improves performance by giving the algorithm cleaner, more focused optimization targets.
The traditional layered structure also remains appropriate when you need to control messaging tightly at each funnel stage, specifically for enterprise SaaS with long sales cycles where awareness, consideration, and conversion messaging serve fundamentally different buyer moments that the algorithm cannot reliably distinguish from behavioral signals alone.
| Account Scenario | Recommended Structure |
|---|---|
| Under 50 conversions/week, seed stage, small budget | Traditional layered funnel. Optimize for higher-funnel events to generate enough signal. Keep creative lean and high-quality rather than chasing volume. |
| 50+ conversions/week, product-led growth SaaS | Andromeda-first: two-campaign testing/scaling split, broad targeting, 10 to 20 distinct creatives, CBO on scaling campaign. |
| Enterprise SaaS, long sales cycle, distinct buyer personas | Hybrid: consolidated structure by persona, not by funnel stage. Keep messaging distinct per buyer type, but consolidate the awareness-to-conversion journey within each persona campaign. |
The rule of thumb: use Andromeda-first consolidation when your account generates enough conversion volume for the system to learn. Use traditional segmentation when it does not. The mistake is applying either model dogmatically without checking your account’s actual data density first.
💡 Pro Tip: If your budget generates fewer than 50 conversions per week and you want to move toward a consolidated structure, optimize for a higher-funnel conversion event first, such as trial signup initiation or a key product activation step, rather than the paid conversion itself. This generates more conversion volume per dollar, gives Andromeda enough signal to learn, and positions the account to eventually optimize for the downstream events that actually predict revenue.
What Creative Actually Works for B2B SaaS on Facebook in 2026
Creative is now the most important lever in your entire Meta advertising program, more important than campaign structure, more important than targeting, and more important than budget allocation. Andromeda pre-screens creative before it ever enters the auction. If your creative does not generate strong engagement signals from the right audience, it does not get delivered, regardless of how much you are willing to spend.
For SaaS specifically, creative volume and creative distinctness have become as important as creative quality. The system needs a variety of genuinely different concepts to match the right message to the right buyer at the right stage.
Problem-First Static Ads
A bold problem statement headline with minimal design remains the highest-converting creative format for cold audience Facebook ads in B2B SaaS. Not a product screenshot. Not a feature list. A single sentence that names the specific workflow pain your product solves. This format filters immediately. The people who recognize the problem self-select by clicking, which means your traffic quality is higher before Meta starts optimizing.
Screen Recording Demo Videos
Short screen recording videos showing the product solving one specific problem in 30 to 60 seconds consistently outperform produced brand videos for Facebook ads for SaaS. Show one workflow. One feature. One outcome. A 45-second Loom-style recording of your product eliminating a painful manual process will outperform a polished 90-second brand video on almost every SaaS paid social campaign, because it shows the buyer exactly what they get without making them work for the insight.
Founder and Customer Testimonial Videos
Authentic, unpolished talking-head videos from founders or customers perform exceptionally well because they carry social proof signals that designed creative cannot replicate. A 60-second video of a customer describing a specific before-and-after outcome will stop a relevant buyer mid-scroll. No production value required. The authenticity is the asset, and Andromeda rewards content that generates genuine engagement over content that looks expensive.
Creative Volume: The New Requirement
Under Andromeda, three to five creative variations per campaign is no longer sufficient. The system needs 10 to 20 genuinely distinct creative concepts to optimize effectively. Distinct means different hooks, different formats, different angles, and different problem framings, not the same ad with a different background color. SaaS companies that treat creative production as a one-time launch task rather than an ongoing operational function will see performance degrade as their creative library ages and Andromeda exhausts the options available.
💡 Pro Tip: When writing ad copy for Facebook ads for SaaS, lead with the problem in the first line, not your product name, not a benefit claim, not a question. “Still manually exporting data from three different tools every Monday morning?” will outperform “Introducing [Product]: the all-in-one solution for…” every time with a B2B SaaS audience. And under Andromeda, that first-line engagement signal directly affects whether the ad gets delivered to the next user. Weak hooks do not just underperform. They get deprioritized at the retrieval stage.
Targeting for Facebook Ads for SaaS After Andromeda
The targeting approach that worked for SaaS Facebook ads in 2022 and 2023 is now actively working against you, and the replacement strategy is simpler than most advertisers expect. Interest-based targeting stacks have degraded significantly. Under Andromeda, your creative is your targeting. The specificity of your creative, not the precision of your audience configuration, determines whether the right buyers see your ads.
Broad Audiences with Strong Creative as the Filter
For most B2B SaaS Facebook ad campaigns in 2026, broad targeting with minimal interest filters outperforms granular interest stacking. Meta’s system reads your creative and finds the people most likely to respond based on thousands of behavioral signals. If your creative specifically names the problem your ICP faces, the algorithm finds the ICP more efficiently than you can through manual targeting. The creative becomes your targeting mechanism.
Lookalike Audiences from Paid Customers
Upload your existing customer list to Meta as a custom audience and build a lookalike from it. A 1% lookalike of your paid customer base remains the highest-quality cold audience available for SaaS Facebook ads. If you have fewer than 1,000 customers, combine paid customers with highly activated trial users to build a seed audience large enough for reliable lookalike modeling. This signal becomes more powerful under Andromeda because the system uses it as one of thousands of inputs for creative matching decisions, not just audience filtering.
Retargeting in the Andromeda Era
Retargeting still works, but how you structure it needs to change. Separate retargeting campaigns for website visitors, non-activated trials, and non-converted trials remain a valid tactic, but they work best when fed by a consolidated prospecting campaign that generates enough warm audience volume to make retargeting pools meaningful. The old model of running three to five separate retargeting campaigns each with small budgets fragments your conversion data and deprives Andromeda of the signal density it needs to optimize each layer effectively. Where possible, consolidate retargeting into one or two campaigns with creative addressing different objection states, and let the system match.
💡 Pro Tip: For seed-stage and early Series A SaaS companies with small customer lists, build your retargeting pool faster by running a top-of-funnel video view campaign before launching conversion campaigns. Anyone who watches 50% or more of your awareness video becomes a warm audience you can retarget at lower CPAs than cold traffic. This also generates early engagement data that Andromeda uses to improve your cold audience delivery over time, making the investment in awareness do double duty.
The Metrics That Actually Matter for SaaS Facebook Ads
The metrics most SaaS companies track for their Facebook ads are the wrong ones, and optimizing toward the wrong metrics is how you build a campaign that looks good in Ads Manager while your actual revenue impact stays flat.
| Metric | Why It Matters for Facebook Ads for SaaS |
|---|---|
| Cost per trial signup | Your primary optimization metric. The cost of acquiring someone who enters your product experience. |
| Trial activation rate | The percentage of trials that reach your activation milestone. Tells you whether your ad attracts qualified users or just signups. |
| Trial-to-paid conversion rate | The ultimate quality signal for your Facebook ads for SaaS. A low trial-to-paid rate despite acceptable CPA means your ad attracts the wrong buyer. |
| Cost per acquired customer (CAC) | Cost per trial divided by trial-to-paid rate. The true cost of acquiring a paying customer from Meta ads. |
| CAC-to-LTV ratio | Whether your Facebook ad spend is sustainable long-term. A 3:1 LTV-to-CAC ratio is the minimum viable benchmark for SaaS paid acquisition. |
Cost per lead and CTR are not irrelevant, but they are inputs, not outcomes. A SaaS company that optimizes its Facebook ads for CPA without connecting that CPA to trial activation and trial-to-paid conversion will consistently mistake a cheap traffic problem for a strong campaign. Track the full funnel from ad click to paid customer, or your optimization decisions will rest on incomplete data.
💡 Pro Tip: Set up a simple weekly tracking sheet that maps Meta ad spend to trial signups, trial activations, and paid conversions. Even a manual Google Sheet updated weekly gives you the funnel visibility that Meta Ads Manager cannot provide on its own. Most SaaS companies that say “Facebook ads do not work for us” are actually saying “we do not know where in our funnel Facebook ads break down,” which is a data problem, not a platform problem.
The Biggest Facebook Ads Mistakes SaaS Companies Make Now
Most Facebook ads for SaaS underperform for predictable reasons that have nothing to do with the platform and everything to do with how campaigns are structured and stocked. These are the mistakes worth fixing before touching your bids or budgets.
- Running too little creative volume. Under Andromeda, three to five creative variations is not a creative strategy. It is a creative shortage. Andromeda needs 10 to 20 genuinely distinct concepts to match the right message to the right buyer. If your creative library is thin, the system has nowhere to optimize, and your delivery suffers.
- Treating minor variations as new creative. Changing a headline word or swapping a background color does not produce a distinct creative in Andromeda’s evaluation. The system requires different hooks, different formats, different problem framings, and different visual approaches. Lazy variation is not variety.
- Sending traffic to the homepage. Your homepage serves multiple audiences and multiple goals. Your Facebook ad traffic has one job: sign up for the trial or request a demo. A dedicated landing page with a single CTA converts significantly better than a homepage for paid SaaS traffic.
- Optimizing for the wrong conversion event. If Meta is optimizing toward form fills but your revenue comes from activated trials, you train the algorithm to find form-fillers, not buyers. Define your conversion event as the action closest to revenue that you can reliably track and pass back to Meta via the Conversions API.
- Fragmenting budget across too many campaigns. Under Andromeda, splitting budget across five or more campaigns chasing the same objective deprives the system of data density. Consolidated structures with CBO let Meta shift spend toward the best-performing creative in real time. Fragmented structures slow learning and produce less consistent delivery.
- Judging campaigns too early. Facebook ads for SaaS need 30 to 60 days of data before drawing meaningful conclusions, especially when your conversion event is a trial signup that takes weeks to convert to paid. Pausing campaigns at two weeks based on early CPA data is almost always premature and resets the optimization the algorithm has already built.
How Much to Spend on Facebook Ads for SaaS
The minimum effective budget for Facebook ads for SaaS depends on your cost per trial and how many trials you need per month to generate meaningful optimization data. Andromeda’s algorithm needs approximately 50 conversion events per week per campaign to optimize reliably. For most SaaS companies with a cost per trial between $30 and $80, that means a minimum of $1,500 to $4,000 per month before your campaigns have enough data to improve on their own.
For seed-stage companies testing Facebook ads for SaaS for the first time, a practical starting budget is $2,000 to $3,000 per month. That gives you enough to run a meaningful conversion campaign and a basic retargeting layer without spreading spend too thin. The biggest budget mistake early-stage SaaS companies make is allocating $500 to $800 per month, generating 10 to 15 trials, and concluding that Facebook ads do not work for SaaS. The real issue is that the budget never gave the algorithm enough signal to optimize.
For Series A companies scaling a proven SaaS paid social program, the budget conversation shifts from minimum effective spend to CAC-to-LTV sustainability. If your CAC from Facebook ads for SaaS is below one-third of your average customer LTV, scaling budget aggressively makes financial sense. If your CAC-to-LTV ratio is marginal, fix the funnel before scaling spend.
💡 Pro Tip: Before scaling your Facebook ads for SaaS budget, calculate your current CAC from the channel and compare it to your 12-month LTV. If you do not know your LTV yet because your product is too new, use your best current estimate and commit to updating it quarterly. Scaling paid acquisition without a reliable LTV estimate is how early-stage SaaS companies burn through runway on a channel that looks efficient in Ads Manager but is not sustainable at the unit economics level.
The Bottom Line on Facebook Ads for SaaS in 2026
Facebook ads for SaaS still work. But the system they run on changed fundamentally in late 2025, and campaigns built on the old playbook are now fighting the algorithm instead of working with it. Andromeda shifted Meta from a platform you configure to a system you feed. The advertisers generating consistent SaaS pipeline from Meta in 2026 spend more time on creative production and less time on audience architecture.
The structural decision most SaaS advertisers need to make right now is whether their account has enough conversion volume to support an Andromeda-first consolidated structure or whether they still need manual funnel segmentation to compensate for thin data. That answer depends on one number: do your campaigns generate 50 or more conversion events per week? If yes, consolidate and invest in creative volume. If no, stay layered, optimize for higher-funnel events to build signal density, and plan the migration toward consolidation as your account matures.
The fundamentals that drove SaaS Facebook ad performance before Andromeda still drive it now: problem-first creative, the right conversion event, full-funnel measurement, and enough budget to give the algorithm meaningful data. Andromeda amplifies good inputs and punishes weak ones faster than the old system ever did. Build the creative engine, send clean conversion signals, and give the system what it needs to find your buyer.
🎯 Want a Meta Ad Strategy Built for How Andromeda Actually Works?
We audit, restructure, and manage Meta ad campaigns for B2B SaaS companies, including Andromeda-native campaign architecture, creative volume planning, and full-funnel conversion tracking. Book a free 30-minute strategy call.
→ Book Your Free Strategy Call
No pressure. Just a straight conversation about what your SaaS paid social program needs to perform in 2026.
Frequently Asked Questions About Facebook Ads for SaaS
What is Meta Andromeda and how does it affect Facebook ads for SaaS?
Meta Andromeda is an AI-powered ad retrieval engine that completed its global rollout in October 2025. It replaced Meta’s old retrieval system and now processes hundreds of thousands of ad creatives per user impression in real time, matching the most relevant creative to each individual based on thousands of behavioral signals. For SaaS advertisers, the most significant change is the shift from audience-first to creative-first delivery. Your creative is now your primary targeting mechanism. If your creative does not generate strong engagement signals from your target buyer, it does not get delivered regardless of how precise your campaign configuration is.
Should SaaS companies use a consolidated campaign structure or a traditional layered funnel after Andromeda?
The right structure depends on your account’s data density. If your campaigns generate 50 or more conversion events per week, an Andromeda-first consolidated structure with a testing and scaling campaign split will outperform a traditional layered funnel. Andromeda’s sequence learning can handle funnel-stage matching automatically when it has enough signal. If your account generates fewer than 50 conversions per week, a traditional layered structure with separate campaigns for awareness, consideration, and conversion still makes sense because the algorithm does not have enough data to make reliable creative-matching decisions on its own. The rule is simple: consolidate when data density supports it, and stay layered when it does not.
How many creative variations do Facebook ads for SaaS need under Andromeda?
Under Andromeda, 10 to 20 genuinely distinct creative concepts per campaign is the new benchmark, up from the 3 to 5 variations that were standard before. Distinct means different hooks, different formats, different problem framings, and different visual approaches. Andromeda evaluates creative similarity strictly, and minor variations like headline swaps or color changes do not count as distinct creative. SaaS companies that treat creative production as a one-time launch activity rather than an ongoing operational function will see performance degrade as their creative library ages and the system exhausts its options.
Do Facebook ads work for SaaS companies?
Yes. Facebook ads work for SaaS companies when campaigns are structured around the specific dynamics of software acquisition and aligned with how Andromeda now evaluates and delivers ads. The key requirements are problem-first creative that speaks to a specific workflow pain, the right conversion event optimized through the Conversions API, enough budget to generate 50 or more conversion events per week, a creative library with 10 to 20 genuinely distinct concepts, and full-funnel tracking from ad click to paid customer. SaaS companies running the pre-Andromeda playbook with heavy targeting segmentation and thin creative libraries are fighting the algorithm and seeing performance decline.
What is the best conversion event to optimize for with Facebook ads for SaaS?
The best conversion event is the action closest to revenue that you can reliably track and pass back to Meta via the Conversions API. For product-led growth SaaS, that is typically a trial signup or product activation event. For sales-led SaaS, that is a demo request or qualified meeting booked. If your budget does not generate enough of these downstream events to hit 50 per week per campaign, optimize for a higher-funnel event such as trial signup initiation to build sufficient data density for Andromeda to learn from.
How much should a SaaS company spend on Facebook ads?
The minimum effective budget for Facebook ads for SaaS is $1,500 to $4,000 per month, depending on your cost per trial. Meta’s algorithm needs approximately 50 conversion events per week per campaign to optimize reliably. For seed-stage SaaS testing the channel for the first time, $2,000 to $3,000 per month gives enough data to run a meaningful conversion campaign and basic retargeting layer. Budgets under $1,000 per month rarely generate enough trial volume to produce reliable optimization signal and are the most common reason SaaS companies conclude incorrectly that Facebook ads do not work for their business.
What creative works best for Facebook ads for SaaS in 2026?
The three creative formats that consistently outperform for B2B SaaS Facebook ads are problem-first static image ads with a bold headline that names a specific workflow pain, short screen recording demo videos showing the product solving one specific problem in 30 to 60 seconds, and authentic talking-head testimonial videos from founders or customers with specific outcome data. Under Andromeda, creative volume matters as much as creative quality. You need 10 to 20 genuinely distinct concepts with different hooks, formats, and angles. Polished brand videos and feature-list creative consistently underperform with B2B SaaS audiences, and thin creative libraries now limit delivery at the Andromeda retrieval stage before ads even enter the auction.
How do you target B2B SaaS buyers on Facebook after Andromeda?
The most effective targeting approach for Facebook ads for SaaS in 2026 is broad audiences with strong problem-specific creative as the primary filter, supported by lookalike audiences built from paid customers or highly activated trial users. Under Andromeda, your creative does the targeting work that interest stacks used to do. The system reads your creative and finds users most likely to respond based on thousands of behavioral signals. Interest-based targeting stacks have degraded significantly and now underperform compared to broad delivery with high-quality, specific creative. Retargeting still works but performs best when consolidated rather than fragmented across multiple small-budget campaigns.
What metrics should SaaS companies track for Facebook ads?
The primary metrics for Facebook ads for SaaS are cost per trial signup, trial activation rate, trial-to-paid conversion rate, cost per acquired customer (CAC), and CAC-to-LTV ratio. Cost per lead and CTR are secondary inputs. Optimizing toward them without connecting to downstream trial quality and paid conversion rate produces campaigns that look efficient in Ads Manager while generating poor revenue impact. Track the full funnel from ad click to paid customer on a weekly basis to make reliable optimization decisions.
How long does it take for Facebook ads to work for SaaS?
Facebook ads for SaaS typically require 30 to 60 days before drawing meaningful performance conclusions. The algorithm needs time to exit the learning phase and accumulate enough conversion events to optimize delivery. SaaS specifically requires additional time because the conversion event, a trial that converts to paid, takes weeks to complete after the initial signup. Under Andromeda, making structural changes or pausing campaigns during the learning phase resets optimization progress more severely than it did under the old system, making patience during the initial learning period more important than ever.