Meta's ad relevance diagnostics are not just vanity metrics. They are direct signals from the algorithm about how your ads are competing in the auction. When those rankings slip below average, you are not just seeing a number change in your dashboard. You are watching your delivery shrink, your costs climb, and your results deteriorate in real time.
The good news is that poor relevance scores are fixable. They are not random. Each ranking dimension points to a specific problem, and each problem has a specific solution. The challenge most advertisers face is not knowing where to start or what to fix first.
This guide gives you a sequential, practical process for diagnosing and fixing poor ad relevance scores on Meta. You will work through each step in order: identifying which ranking dimension is failing, auditing your targeting, refreshing your creative, aligning your copy, building a testing system, and scaling what works. Every step builds on the one before it.
Whether you manage a single brand account or run campaigns for multiple clients, this process applies directly to your Meta Ads workflow. The goal is not just to improve a number in a column. It is to lower your cost per result, increase your delivery efficiency, and build a system that keeps your relevance scores healthy over time.
Let's get into it.
Step 1: Diagnose Your Current Relevance Score Breakdown
Before you change anything, you need to know exactly what is broken. Meta replaced its single relevance score in 2019 with three separate diagnostics, and each one tells you something different about why your ad is underperforming.
Here is what each ranking measures:
Quality Ranking: How your ad's perceived quality compares to other ads competing for the same audience. This reflects creative quality, ad experience signals, and whether Meta's system considers your ad valuable to users.
Engagement Rate Ranking: How your ad's expected engagement rate compares to other ads targeting the same audience. Low engagement ranking usually points to an audience-creative mismatch, where the right message is not reaching the right people.
Conversion Rate Ranking: How your ad's expected conversion rate compares to other ads with the same optimization goal and audience. This ranking is influenced not just by your ad but by your post-click experience, including your landing page.
Each ranking uses five labels: Above Average, Average, Below Average (Bottom 35%), Below Average (Bottom 20%), and Below Average (Bottom 10%). The more specific the label, the more urgent the fix.
To add these columns to your reporting view in Ads Manager, navigate to the Columns dropdown, select Customize Columns, and search for each of the three rankings individually. Add all three and save the view. Now you can see every ad's ranking status at a glance.
Once you have the columns visible, scan your active ads and note which specific ranking is lowest for each underperforming ad. This is critical because it determines your next move:
Low Quality Ranking: Your creative is the problem. Focus on Step 3.
Low Engagement Rate Ranking: Your audience targeting is the problem. Focus on Step 2.
Low Conversion Rate Ranking: Your landing page or post-click experience is the problem. Focus on Step 4 and check message match.
Build a prioritized list of the ad sets and creatives with the worst rankings before moving forward. Do not make changes across your entire account at once. Work through your worst performers first and treat this list as your repair queue.
The most common mistake at this stage is jumping straight to creative changes without checking which ranking is actually failing. If your Conversion Rate Ranking is the issue, refreshing your creative will not fix it. Diagnose first, then act. Reviewing your Meta ads performance data systematically before making any changes will save you significant time and budget.
Step 2: Audit Your Audience Targeting for Misalignment
If your Engagement Rate Ranking is the primary issue, your audience setup is where you need to look. Low engagement ranking means Meta's algorithm expects your ad to generate below-average engagement compared to other ads competing for the same audience. That is almost always a sign that your message and your audience are not aligned.
Start by reviewing the audience definitions attached to your lowest-ranking ad sets. Ask yourself whether the audience you have defined actually reflects the type of person who would genuinely respond to this specific creative and offer. If you are running the same broad interest-based audience across five different ad sets with very different creatives, that is a red flag.
Next, check for audience overlap using Meta's Audience Overlap tool, which you can find under the Audiences section in Business Manager. Overlapping audiences mean multiple ad sets are competing against each other in the same auction. This drives up your own costs and fragments your delivery, which negatively affects relevance rankings across all affected ad sets.
Look at your interest targeting structure. Interest stacking, where you layer many unrelated interests into a single ad set, can dilute relevance by broadening your audience beyond the people who actually care about your offer. If you have stacked ten interests together hoping to widen reach, you may be reaching a large pool of loosely connected users rather than a focused group with genuine intent.
Consider which audience type is best suited for each creative. A cold interest-based audience needs different messaging than a warm retargeting audience or a Lookalike Audience built from your best customers. Running the same creative across all three audience types without adjusting the message is a common source of engagement ranking problems.
Questions to ask during your audit:
Is this audience too broad? Broad audiences can work well with the right creative, but if your creative speaks to a specific pain point, a tighter audience will generate better engagement signals.
Is this audience too narrow? Audiences that are too small can limit delivery and prevent the algorithm from optimizing effectively, which can also suppress rankings.
Are multiple ad sets targeting the same audience? Consolidate or differentiate. Either merge the ad sets or make sure each one has a meaningfully different audience definition.
After completing your audit, update your audience definitions to better match the creative and offer in each ad set. Tighter alignment between who you are targeting and what you are showing them is one of the most direct levers you have for improving Engagement Rate Ranking. Exploring AI-based customer targeting solutions can also help you identify higher-intent segments you may be missing with manual methods.
Step 3: Refresh Creative Assets That Are Dragging Quality Rankings Down
Creative fatigue is one of the most common and most overlooked drivers of poor relevance scores. When the same audience sees the same creative repeatedly, engagement drops, negative feedback signals accumulate, and Meta's algorithm begins to deprioritize your ad in the auction. Your Quality Ranking reflects this directly.
The first thing to look for is frequency. In Ads Manager, check the frequency metric for ads with below-average Quality Rankings. If frequency is climbing while CTR is declining, creative fatigue is almost certainly a contributing factor. This is your signal to rotate.
Static image ads that have been running for an extended period are typically the first to show fatigue. If you have been running the same image creative for several weeks against the same audience, it is time to replace it. The goal is to introduce fresh creative that enters the auction without the accumulated negative signals attached to the fatigued ad.
This is also the right moment to introduce new creative formats. If you have been running primarily static images, testing video ads or UGC-style content can meaningfully reset audience response. These formats often generate stronger engagement signals, particularly on Instagram placements, which can help lift your Quality Ranking faster than simply swapping one static image for another.
Generating fresh creative at scale used to require designers, video editors, and significant production time. Platforms like AdStellar have changed that. With AdStellar's AI Creative Hub, you can generate new image ads, video ads, and UGC-style avatar creatives directly from a product URL without any design or production resources. You can also clone competitor ads from the Meta Ad Library inside the platform to understand what creative approaches are currently winning in your niche, and then use that intelligence to inform your own fresh creatives.
Once you have generated new creative options, use AdStellar's chat-based editing to refine the messaging. Align the language in your creative to the specific pain points and language patterns of your target audience. A creative that speaks directly to what your audience cares about will generate stronger initial engagement signals, which helps establish a stronger Quality Ranking from the start.
A few practical rules for creative refreshes:
Do not pause the fatigued ad before the new creative has data. Launch the new creative, let it gather initial signals, and then pause the underperformer once you can see the new version is gaining traction.
Keep your best-performing creative formats as a reference point. If a specific format consistently generates strong Quality Rankings, use it as a template for future refreshes rather than starting from scratch each time. Understanding AI ad creation workflows can dramatically speed up how quickly you produce and rotate fresh assets.
Success indicator: New creatives entering the auction with no fatigue signals, fresh engagement data accumulating, and Quality Ranking beginning to stabilize at Average or above within the first week of delivery.
Step 4: Align Your Ad Copy and Headlines to Audience Intent
Creative visuals get attention. Copy and headlines determine whether that attention converts into a click. If your Quality Ranking or Engagement Rate Ranking is suffering despite decent creative, your copy is often the culprit.
Start by reviewing the primary text and headlines attached to your lowest-ranking ads. Read them as if you were the specific person in your target audience seeing this ad for the first time. Does the language match how that person thinks and talks about this problem? Does the headline immediately address something they care about, or does it lead with a product feature that means nothing to someone who does not already know your brand?
The most common copy mistake is writing generic messaging that could apply to any audience. Generic copy consistently underperforms against tailored messaging because Meta's engagement signals reflect whether users find the ad personally relevant. When someone in your audience reads your headline and thinks "this is exactly what I need," that generates a positive engagement signal. When they scroll past because the copy feels like it was written for everyone, that is a negative signal.
Here is how to tighten your copy alignment:
Lead with the outcome, not the product. Instead of "Introducing our new project management tool," write "Stop losing track of deadlines before they cost you a client." The first version talks about you. The second version talks about them.
Match tone to audience temperature. Cold audiences who have never heard of your brand need different copy than warm retargeting audiences who have already visited your site. Cold audiences need context and credibility. Warm audiences need a reason to come back and convert.
Test multiple headline variations. Different people in the same audience are motivated by different things. One person might respond to a headline about saving time. Another responds to a headline about reducing cost. Running multiple headline variations against the same audience lets the algorithm identify which motivation resonates most, which improves both your Engagement Rate Ranking and your overall campaign efficiency. Knowing what to include in ad copy for different audience temperatures is a skill that pays dividends across every campaign you run.
Also check the message match between your ad copy and your landing page. If your ad promises a specific offer, discount, or outcome, your landing page must deliver on that promise immediately. A disconnect between what the ad says and what the landing page shows creates a poor post-click experience that directly suppresses your Conversion Rate Ranking. This is one of the most impactful and most frequently overlooked fixes available to you.
After updating copy, give the revised ads at least several days of delivery before evaluating ranking changes. Copy improvements take time to register in Meta's delivery system as new engagement signals accumulate.
Step 5: Build a Structured Testing System to Find Winners Fast
One-off fixes will improve individual ads, but they will not solve the underlying problem if you do not have a system for continuously finding what works. A structured testing process is what separates advertisers who consistently maintain strong relevance scores from those who are always playing catch-up.
The core principle is straightforward: test systematically rather than randomly. This means deciding in advance what you are testing, how you will measure it, and what threshold determines a winner or a loser.
When diagnosing a specific issue, testing one variable at a time gives you clean data. If you change the creative, the headline, and the audience simultaneously, you will not know which change drove the improvement. However, when you are trying to find winning combinations quickly rather than isolate a single variable, bulk variation testing is far more efficient. Understanding multivariate testing principles can help you design experiments that surface meaningful insights without wasting budget on inconclusive results.
AdStellar's Bulk Ad Launch feature is built for exactly this. You can mix multiple creatives, headlines, audiences, and copy variations at both the ad set and ad level. AdStellar generates every combination and launches them to Meta in minutes rather than hours. What would take a media buyer a full day of manual setup can be done in a fraction of the time, which means you get to meaningful data faster.
Before launching any test, let AdStellar's AI Campaign Builder analyze your historical campaign data. The AI ranks every creative element, audience, and headline by past performance and builds complete campaigns based on what has actually worked for your account. Every decision comes with a transparent rationale, so you understand the strategy behind the structure, not just the output. This historical intelligence dramatically improves the baseline quality of your tests because you are not starting from zero each time.
Once your tests are live, give each ad set sufficient budget and time to exit the learning phase before drawing conclusions. Evaluating performance too early, before Meta's algorithm has gathered enough data to optimize delivery, leads to premature decisions that waste both budget and time.
Use AdStellar's AI Insights leaderboards to rank your creatives, headlines, copy variations, and audiences by real metrics: ROAS, CPA, CTR, and whatever benchmarks matter most to your specific goals. Set your target goals inside the platform and AdStellar scores every element against those benchmarks automatically. This means you can instantly identify which combinations are performing above threshold and which ones to cut. Pairing this with strong performance analytics for ads ensures your decisions are grounded in statistically reliable data rather than early fluctuations.
Success indicator: A repeatable process where new tests are regularly launching, winners are being identified based on real performance data, and your account is consistently surfacing high-relevance combinations rather than relying on guesswork or instinct.
Step 6: Scale Winners and Retire Underperformers Systematically
Finding a winning ad combination is only half the job. The other half is acting on it quickly and deliberately. Relevance scores tend to be strongest when you are running your best-performing creative against your best-matched audience at a budget level the algorithm can optimize efficiently.
Once your testing surfaces high-relevance combinations, move them into your scaling campaigns. Do not leave winners sitting in test ad sets with limited budgets. Promote them into your active campaigns where they can generate real volume and continue accumulating positive engagement signals.
AdStellar's Winners Hub makes this process systematic. Your top-performing creatives, headlines, audiences, and more are organized in one place with real performance data attached. When you are ready to build your next campaign, you can pull any winner directly into it without rebuilding from scratch. This eliminates the common problem of institutional knowledge getting lost between campaigns, where a winning creative from three months ago gets forgotten because it is buried in a past ad set.
On the other side of this equation, be disciplined about retiring underperformers. Ads that consistently rank below average across multiple ranking dimensions are not just underperforming individually. They are consuming budget that could be going to your winners, and in some cases they can drag overall account health signals downward. Pausing them is not a defeat. It is good account hygiene. If you manage campaigns across multiple clients, building this retirement discipline into your workflow is one of the most impactful habits covered in guides on how to manage Facebook ads for clients effectively.
Establish a regular cadence for reviewing your relevance rankings. A weekly check-in is a reasonable starting point for most accounts. During each review, look for creatives where frequency is rising and CTR is falling, ad sets where rankings have slipped since the previous week, and any new winners from your testing pipeline that are ready to be promoted.
One important caution: Do not scale budgets on ads before their relevance rankings have stabilized. Pushing budget into an ad that is still in the learning phase or showing inconsistent rankings can lead to wasted spend during a period when the algorithm has not yet optimized delivery. Wait for rankings to reach at least Average before scaling, and look for Above Average before making significant budget increases.
The goal of this step is not just to scale a single winner. It is to build a flywheel where every campaign generates winners that inform the next one, and your overall account quality improves over time.
Putting It All Together
Fixing poor ad relevance scores on Meta is not a one-time task. It is an ongoing discipline that requires consistent diagnosis, creative refreshes, audience alignment, and systematic testing. The steps in this guide give you a structured path to follow every time your scores start to slip.
Before you move on, run through this quick checklist:
Diagnosis: Have you identified which specific ranking dimension is below average for your underperforming ads?
Audience: Have you audited your targeting for overlap, misalignment, and interest stacking issues?
Creative: Have you refreshed assets showing fatigue signals and introduced new formats where needed?
Copy: Have you aligned your headlines and primary text to the specific intent and language of your target audience?
Testing: Have you built a structured testing system that generates winners consistently rather than relying on one-off changes?
Scaling: Have you moved winners into active campaigns and retired persistent underperformers?
Every step in this process can be accelerated with the right tools. AdStellar is built to handle exactly this workflow: generating fresh creatives from a product URL, building campaigns from historical performance data, launching hundreds of variations at once, and surfacing your top performers automatically through AI Insights leaderboards and the Winners Hub. No designers, no video editors, no guesswork.
If you are ready to stop manually chasing relevance scores and start building a system that scales what works, Start Free Trial With AdStellar and see how fast you can move from diagnosis to winning campaigns with a platform designed to do the heavy lifting for you.
