Meta Ads consulting: the right people on Instagram and Facebook, with measured sales.

Meta Ads is how you show ads on Facebook, Instagram and Meta’s other surfaces. Today, Meta’s delivery system does most of the targeting itself, leaving advertisers with two strong levers: the event signal sent from their site, and the diversity of their creative. We wrote this guide for ecommerce and lead generation teams that advertise on Meta or are getting ready to. Every rule is based on Meta’s Business Help Center, its developer documentation and its engineering blog; the sources are numbered at the end of the page. We deliberately left out third-party tactic lists and figures we couldn’t verify.

Advertising on Instagram and advertising on Facebook run in the same system, from a single Meta Ads account. Our Meta Ads consulting and ad management service has four parts: making measurement solid, producing creative that is genuinely distinct, setting up campaigns in a simple structure that respects the learning phase, and reading results against your own order and lead records. The ten sections below explain the mechanisms behind this work using Meta’s own documentation.

What stages does Meta take an ad through before showing it?

Meta selects an ad in two stages: first, the ads retrieval stage narrows tens of millions of candidate ads down to a few thousand, then ranking models score those candidates and the ad to show is decided.¹ The system that runs the retrieval stage is called Andromeda; Meta’s engineering team says it delivered a 10,000x increase in model capacity at the retrieval stage.¹ Meta for Business describes the same process from the user’s point of view: when someone opens Instagram or Facebook, Andromeda scans tens of millions of ads, and only the most relevant and distinct creatives move on to the next stage.²

This chain has four links: signal, candidate ads, delivery in the feed and conversion. Each link feeds the next. If the signal is weak, the system can’t learn who to show the ad to; if creatives look alike, only one idea enters the candidate pool; if the landing page is slow, even a good ad won’t produce conversions. That’s why our consulting starts at the beginning of the chain, with measurement.

Which ad wins the auction?

The auction isn’t won by the highest bid but by the ad with the highest total value.³ According to Meta, an auction takes place every time there’s an opportunity to show someone an ad, billions of auctions run every day, and the ads whose target audience includes that person compete.³ Total value combines three main factors: bid, estimated action rate and ad quality.³

Together, estimated action rate and ad quality measure the ad’s relevance; that’s why an ad that’s more relevant to the person can beat ads with higher bids.³ Meta also applies a reserve price and writes that its auction adjustments won’t lead to charging advertisers more than their bid.³ Meta doesn’t publish the formula that combines the three factors; the practical conclusion we draw is this: raising the bid doesn’t make up for a weak signal or low-quality creative.

Why are the Pixel and the Conversions API set up together?

The Pixel and the Conversions API are set up together because Meta recommends a “redundant” setup, where the same events are sent from both the browser and the server.⁴ The Pixel is code added to your site that runs in the visitor’s browser; it can lose some events because of network connectivity issues or page loading errors. The Conversions API (server-side event reporting) sends the same events from your server and fills in what the Pixel misses.⁴

When both channels send the same event, the sale mustn’t be counted twice; for this, Meta asks you to use the same event name (event_name) and the same event ID (event_id) in both events.⁴ Timing affects results too: for sales ads, Meta recommends sending Conversions API events within an hour; the upper limit is seven days, but performance can suffer with events sent outside the recommended window.⁵

1. Event map: page view, product view, add to cart, initiate checkout, purchase or form submission; when each event fires and with which value is written down.
2. Pixel: browser-side events are set up together with the event ID.
3. Conversions API: the same events are sent from the server with the same name and ID; the action_source, event_source_url and client_user_agent fields are added to web events.⁴
4. Verification: deduplication, event coverage and match quality are checked in Events Manager; an end-to-end trial is run with test events.

What does Event Match Quality measure?

Event Match Quality scores, from 0 to 10, how effective the customer information you send with a server event is at matching that event to a Meta account.⁶ The score is calculated from the quality of the information sent and the share of events matched to Meta accounts; because it uses data from the last 48 hours, events need to be sent regularly.⁶ Matched events help show ads to people more likely to convert and attribute conversions to ads.⁶

Contact details such as email and phone number are sent to Meta not as plain text but as hashed values; Meta’s Business Tools Terms require both this and that you have the legal right to collect and share the data.⁶ That’s why raising the match score isn’t a data collection race: we clarify your privacy notice and consent setup with your legal counsel, and send only fields that are consented to and actually in your hands. The fbp and fbc cookie values from the browser are also added to the server event; Meta notes that these values can change and need to be refreshed regularly.⁴

Why does creative diversity decide Instagram ad results?

Creative diversity is decisive because Meta treats ads that look alike as variations of the same creative and shares learning across those variations.² Meta defines creative diversification as producing distinct campaign assets built for different customer types or use cases, and recommends ads that genuinely differ in look, feel, story and message.² Uploading the same video with three different headline colors is, in the system’s eyes, a single idea.

Creative fatigue is the other side of the same mechanism: in Meta’s definition, when an audience is overexposed to the same image or video, engagement drops and cost per result rises.² That’s why we build creative not as a one-off production but as a system that brings in new ideas every month.

Which formats do Facebook and Instagram ads need for Reels, Stories and Feed?

Reels and Stories need a 9:16 vertical format; Feed needs a 1:1 square or a 4:5 vertical format.⁷ Meta recommends 4:5 for single-image ads in Facebook Feed and 1:1 in Instagram Feed; for Stories, Status and Reels placements, it reminds you that people mostly hold their phones upright and recommends full-screen 9:16.⁷

In accounts where the feature is available, Ads Manager lets you upload up to 10 images or videos to a single ad and fills in missing ratios itself by cropping, padding or expanding.⁷ Meta recommends reviewing automatic crops before publishing so important content isn’t cut off.⁷

When are Advantage+ campaigns the right choice?

For campaigns with a sales, leads or app promotion objective, Meta recommends Advantage+ campaigns, which apply AI to many steps at once.⁸ The end-to-end solutions in the Advantage+ product family apply AI to a campaign’s audience, placement and budget together; this group includes Advantage+ sales, app and leads campaigns.⁹ The product formerly called Advantage+ shopping campaigns is now named Advantage+ sales campaigns.⁹ The condition we add is this: automation works only as well as the signal it’s given; if measurement isn’t solid, we fix that first.

That’s also the logic behind broad targeting. To find your audience, Advantage+ audience uses information such as past conversions, Pixel data and interactions with previous ads; Meta recommends A/B testing Advantage+ audience in almost every campaign type except retargeting campaigns.⁸ In practice, this means strengthening the signal rather than narrowing the audience with interest lists: the system learns who to show ads to from the conversions coming from your site.

We don’t leave the automation blind either. We write real business constraints such as age, country or delivery area into the settings, and tie audience definitions to reporting so new and existing customers can be read separately. For stores that sell with a catalog, product data quality is a separate line of work; we handle it together with our ecommerce marketing work.

How is the learning phase managed?

The learning phase is managed by giving the ad set enough budget and a period without edits: according to Meta, ad sets usually exit learning once they get about 50 results within a week of their last significant edit.¹⁰ During this time, delivery is less stable and cost per result is usually higher; an ad set that can’t get enough results shows the “Learning limited” status.¹⁰

Not every edit restarts learning, only significant ones. Meta’s distinction is as follows:¹¹

• Always significant: any change to targeting, creative or the optimization event; adding a new ad to the ad set; changing the bid strategy; pausing the ad set for 7 days or longer.
• Significant depending on size: budget amount, spending limit, bid cap, cost per result or ROAS goal. In Meta’s example, raising the budget from $100 to $101 likely won’t restart learning, while raising it to $1,000 may.
• Doesn’t restart learning: Advantage+ campaign budget redistributing budget across ad sets.

Meta’s recommendations for the learning phase point the same way: don’t edit before learning is complete, don’t open too many ads and ad sets, combine similar ad sets to pool learning, and set a realistic budget.¹⁰

How do you read the attribution setting and the measurement dashboard?

The attribution setting decides which conversions are credited to your ad, and results in Ads Manager are counted according to it.¹² Meta now offers standard and incremental attribution models at the ad set level; custom attribution, which lets you bring the attribution logic from your own analytics tool into Meta, is also available.¹² Incremental attribution works with models that estimate whether a conversion was caused by the ad.¹²

Meta’s own warning matters here: results from ad sets using different attribution models can’t be compared in the same table, because each model counts differently.¹² That’s why we build the measurement dashboard in three layers.

The first layer is Ads Manager: results counted by Meta’s own attribution rules, cost per result and ROAS. The second layer is Events Manager: whether events are arriving, deduplication and match quality. The third layer is your own records: the order dashboard, the CRM and GA4. Decisions are made by reading all three together; sales in the Meta dashboard aren’t expected to match sales in your store dashboard exactly, but if the gap suddenly changes, the measurement setup is the first thing to check.

How does our Meta Ads service work in the first 30 days?

The first month is planned to make measurement solid and take the campaign cleanly through the learning phase. The sequence below is what we apply in a newly opened or taken-over account.

1. Week 1 · Audit: the Pixel, Conversions API, deduplication and match quality are checked; account structure, attribution settings and past creatives are reviewed.
2. Week 2 · Setup: missing events and server-side sending are completed; creative concepts are prepared in three ratios.
3. Week 3 · Launch: campaigns go live in a simple structure; the budget is planned at a level where the ad set can reach about 50 results a week.¹⁰
4. Week 4 · First review: learning status, costs and the comparison with order records are reported; next month’s creative and budget plan comes out of it.

What should you look for in a Meta ads agency?

A “Facebook ads agency”, an “Instagram ads agency” and a “social media ads agency” usually describe the same job: building and running campaigns in Meta’s single ad system. What separates them isn’t the label but who takes on measurement and creative. We see three common models:

Ask for concrete answers to these questions: who sets up the Conversions API and how, how is deduplication verified, how many new creative concepts come each month, which independent source are results compared against, and who keeps the ad account?

1. Business portfolio and ad account: the ad account, Pixel and event dataset sit in your brand’s business portfolio, and the agency works with partner access. If an agency opens the account in its own portfolio, campaign history and event data can stay behind when you leave.
2. Budget kept apart from the fee: ad spend is paid to Meta from your own payment method, and the management fee is invoiced separately. When the two are bundled, you can’t see how much actually reached Meta.
3. Measurement: a Pixel-only setup leaves sales lost in the browser out of learning; check in Events Manager that the Pixel and Conversions API run together and are deduplicated.
4. Creative capacity: in an account without new concepts every month, creative fatigue pushes costs up; get in writing who produces creative, how often and in which ratios.
5. Meta Business Partner badge: ask which solution type the badge was earned in; it reflects the volume and quality of a company’s work on Meta, not a guaranteed result in your account.
6. Guarantee language: competitors’ bids and Meta’s formula aren’t visible in the auction; an offer that guarantees a sales count or a fixed cost doesn’t fit how the system works.

What is a Meta Business Partner, and what does the badge tell you?

Meta Business Partners is Meta’s program for companies it recognizes for their ability to help businesses grow. According to Meta, eligibility varies by solution type and is typically a combination of the quantity and quality of a company’s work across Meta technologies such as Facebook, Instagram and Messenger; those who qualify are recognized with a Meta Business Partner badge (Meta for Business: Meta Business Partners). Meta doesn’t publish numeric thresholds on that page, so we don’t repeat the spend figures that circulate on third-party sites.

The badge says a company runs a certain volume and quality of work on Meta; it doesn’t say how your account will be built. Even with a badged partner, the decision should rest on the answers to the questions above and on the first month’s report.

What drives Meta Ads management pricing and Instagram ad costs?

Advertising on Instagram or Facebook costs two separate things: the ad spend paid to Meta, and the fee of the partner who manages the account. Ad spend is set by your budget and the auction; cost per result varies with the industry, competition, creative and the quality of measurement. The factors and common models behind the management fee:

We work out the scope with you in the first meeting, then send a written proposal. Ad spend is separate from our fee and is paid to Meta directly from your own ad account.

Meta Ads consulting pricing: estimated ranges

Estimated range, excluding VAT; rows marked “/ month” are monthly, the others are one-off. Scope is confirmed in a written proposal after the first meeting. As of October 2026. Ad spend is not included in the service fee.

If you also advertise on search, we explain reading both channels in the same dashboard on our Google Ads consulting page. To see how AI assistants describe your brand outside the feed, you can run our free AI visibility analysis. If you’d like to talk about your own account, pick a meeting day below.