aimClear’s Roadmap for Facebook Ads: Building A Better Channel Together
Being a Facebook Ads marketer is like having a love/hate relationship with a miracle. aimClear has been killing Facebook Ads since 2007, just days after the emergent DIY channel was launched. That December our Hail Mary Facebook Ads campaign literally saved a client’s Christmas. The rest is history. We love FB ads and believe the channel often has a deadeye role in the mainstream marketing mix.
About a trillion impressions and hundreds of clients later, we’ve come to some strong opinions as to where the Facebook team should focus their efforts. The channel, as incredibly beautiful as it is, still needs to play catch up with Google in form and functionality. There is a lot of misunderstanding out there about what FB Ads really is. Paid? Organic? What? Facebook works hard to make things easier and more turnkey to community managers and SMBs. But sometimes the changes make it harder for marketers.
Also, there seems to be a covert PR war against FB Ads. We’ll say this: Anti FB Ads concerns seem to have better PR firms. The mashup of PR bullsh*t prior to and during the IPO was staggering to the informed observer. Let’s just say here: Anyone who thinks Facebook Ads don’t have great ROI doesn’t use FB Ads for the right KPIs.
This post is aimClear’s contribution to where we believe the road should lead. We know that FB is crammed with incredible ideas, awareness of bugs, and at no loss for radical hacks from the wunderkind team. This post shares some of what we’ve learned and our feedback as to what we would do next.
It is said that “if we had more time we would have made it shorter.” In this topic’s case, that axiom does not hold true. The copious content herein belie the granularity and importance of what needs to be discussed. Our suggestions are not prioritized or complete. They’re a serious start.
This article is divided up in the following categories: Account Structure, Targeting, Management & Workflow, Power Editor Specific, Reporting/Metrics, Bugs and Help! Please chime in with any suggestions as to what YOU believe FB should focus on, whilst moving the Ads platform forward.
Current Status: Today, the Facebook Ad account structure is: account, campaigns, and ads. Within each account there are campaigns, and within each campaign are ads. Targeting is at the ad level.
Suggested Improvement: Create structure similar to other PPC platforms such as Google AdWords: account, campaigns, ad groups, ads – where the “ad group” holds the targeting.
Rationale: The current system leads to anarchy and no obvious structure. aimClear has overcome this system by assembling campaigns to mimic ad groups wherein each ad within a campaign is targeted to exactly the same as the others. This way, when looking at ads within a campaign we are comparing variations (creative) to the same constant (targeting).
Build & Save Targeting Segments + Production Tool
Current Status: Must create an ad to build targeting.
Suggested Improvement: Have the ability to create a targeting segment independent of ads and vice versa. Targeting should be held at the “ad group” Level (see first suggestion) and then ad creative should be applied to the targeting segment (or “ad group”).
Rationale: Sophisticated advertisers do not create targeting after creating ads. Instead, they create targeting then match the best ads to the targeting. With a tool such as this, advertisers will be able to spawn and deploy more ads to more targeting segments in rapid-fire fashion.
Social Segment “Keyword” Stemming Tool
Current Status: Within the Precise Interest tool, Facebook drops down suggested terms stemmed from letters entered. After selecting multiple interests, Facebook stops dropping down suggestions even though there are many more related.
Suggested Improvement: Create a keyword-stemming tool similar to Acquisio’s tool.
Rationale: To work around this limitation, aimClear has devised an alpha pattern solution where we start with the root interest, select all related interests then type in “Keyword A,” grab all relevant interests, then “Keyword B,” etc. Often, we have to go further to find all the interests relevant, such as “Keyword Ab…” etc.
Keyword Stemming with Alpha Patterns example – Gridiron Football Interests:
As you can see, if an advertiser is looking to target American gridiron football interests, many of the interests Facebook’s tool drops down (seven interests dropped down in total) are for soccer. There is one relevant interest for gridiron football: “Football Watching Playing” though the Facebook tool will not show additional interests related to “watching Football.” Many advertisers may select that and call it a day without stemming further, potentially missing great opportunities as there are MANY interests related to “watching football.”
Manual precise interest keyword stemming can be a very long, cumbersome process, and rarely all-encompassing. Advertisers will find more interests to target ads to, create more targeting segments, target more interests thus increasing the probability they will spend more advertising dollars.
Copy + Paste Feature for Targeting Elements
Current Status: Facebook “precise interests” can be copied from a CSV and pasted directly in to the Ad UI and Power Editor.
Suggested Improvement: Ability to copy + paste into other targeting elements: (geo) Cities, Zip Codes, Countries, Workplaces, Universities, and Education Majors. Also note: this feature only really works after advertisers use the Facebook tool to identify which precise interests actually exist. Taking external CSV lists and pasting them in to the Facebook Precise Interest tool can yield some wacky results.
Below is a list of business conferences and the interests Facebook’s tool populated for this list that includes a radio station and the film Where the Heart Is.
Rationale: The copy + paste function in to the precise interests has been a major time-saver for many advertisers, aimClear included. Manually entering multiple zip codes, cities, countries, universities, education majors, and especially workplaces (for larger B2B clients) still takes a large amount of time.
Negative Interests & GEO
Current Status: Can only target people based on positive variable.
Suggested Improvement: Include negative interest and geo variables.
Rationale: Often clients want to target people who live in communities other than metropolitan areas. Having a better geo tool that would allow advertisers to target where people do not live. There is a lot of competition for advertising within metropolitan areas and opportunity for business to sell products or services to people who live in more rural areas (think major retailers who’d want to only promote their online sales to users who do not have access to their physical store).
Negative interests can be employed for many things, but for a relevant, powerful example: These could be utilized for political parties to influence “swing” voters – people who do not like “democrat,” “liberal,” “Obama,” “conservative,” “GOP,” “Republican,” etc. Arguably, what Facebook users do not like is as revealing as what users do like.
Job Titles as a Separate Targeting Option
Current Status: The only way to target job titles is through the precise interest tool. While it works, it may not be terribly accurate and can’t be matriculated with other Precise Interests.
Suggested Improvement: Siphon job title data to its own targeting option.
Rationale: Having job titles as a separate targeting option, not clumped together with precise interests, will add a deep layer for both B2B and B2C marketers. For example, we would then be able to target “Financial Planners” who like “Fastcompany, or Inc. 500” and are in the broad category SMB owner, and graduated with a degree in finance.
Beef up job titles and work place variables and LinkedIn won’t even be a thought for social advertising, not even for the staunchest B2B brands and products.
If there was a way to marry the data between users’ Facebook accounts and their LinkedIn accounts – while keeping the accounts separate – the B2B sector would go bananas. The LinkedIn target grid and data married with Facebook’s would be more beautiful than a Jolie-Pitt baby.
Targeting Workplaces as Categories
Current Status: Target specific, individual companies.
Suggested Improvement: Add the ability to target companies by category.
Rationale: Take a look at the LinkedIn targeting grid. It’s amazing. B2B advertisers (and some of our clients) are still skeptical of Facebook as a tool to target workplaces and job titles. Buttoning these down will instill confidence in more B2B advertisers.
Excessive Variable Warning Errors
Current Status: Facebook tools do not alert advertisers who are creating targeting segments when they’ve reached the limit of countries targeted, workplaces, and precise interests.
Suggested Improvement: Alerts displayed in the interface when advertisers have reached the allotted amount of variables Facebook will allow.
Rationale: Currently, if advertisers attempt to add more than 25 countries, more than 200 precise interests, and more workplaces than allowed (we don’t have the exact number for that one, and it’s difficult to find in Facebook’s help section) there is no warning. If an advertiser attempts to add more than 200 precise interests the tool will just stop allowing interests to be added. Conversely, Facebook will not alert or stop the ability to enter additional countries or workplaces, however, the estimated reach will no longer increase (or decrease). For an advertiser who is not totally familiar with this anomaly, he or she may be discouraged by the targeting – “Why is the estimated reach not increasing?” – leading to distrust.
When the advertiser seems to be finished and selects “Place Order” to submit the ad to Facebook’s approval process, then – and only then – is the advertiser alerted that choices must be pared down. This is a waste of time and can be very frustrating.
Email Full & Partial Account Structure Document, Including Creative
Current Status: There is no way to output account structure including creative, for client approval, without scraping together reports by hand. This makes working with clients very clunky and eats up time.
Suggested Improvement: Look to Google AdWords Editor HTML output. Make it better by adding the ability to email clients for approval. Include the ability to output featuring performance metrics, especially at the creative level. We want to send clients targeting and ads that work, including the ad copy and image
Facebook Ads Management
Rotate Ads Evenly Option
Current Status: Ads within a campaign are auto-optimized for CTR.
Suggested Improvement: Create a “Rotate Ads Evenly” or “Optimize for CTR” at a campaign level.
Rationale: Ad fatigue is one of the more serious issues Facebook advertisers must tackle. Facebook users log in multiple times per day and often spend a great deal of time with each login, resulting in over-saturation of ads. To combat this, advertisers deploy many variations of ads for users to see. For advertisers to be sure each ad is seen evenly they must manually rotate ads.
It is understood by the advertising community that it is in Facebook’s best interest to serve ads that have the highest CTR, however, it is in the users and advertiser’s best interest to show many variations of ads, then optimize through CTR.
Current Status: No day-parting.
Suggested Improvement: Add day-parting feature to the ad platform.
Rationale: There are better times during the day to share links on a Facebook page and definitely more optimal times for advertisers to places ads. Currently, if advertisers wanted to only show ads in the afternoon, they would have to manually turn them on when desired and pause them all in the evening. Allowing advertisers to select what times of the day their ads show would allot more control of media spend and allow them to make better decisions when it comes to ad buys.
Frequency Caps on Individual Users
Current Status: No frequency caps.
Suggested Improvement: Add a frequency cap on an individual users.
Rationale: Some users need to see an ad a couple of times before they click, but others will never click… no matter how many times they may see an ad. In the latter situation, said users may become annoyed with the ad and X-out of it, damaging the quality score and tanking the CTR.
Frequency capping will create a happier environment for users, as they will be less annoyed by incessant and, to them, irrelevant ads. By employing a frequency cap, advertisers will have more control, feel more confident in the platform and most likely see their CTRs increase.
Account-Wide Daily Budget
Current Status: Facebook Ad accounts have daily campaign budgets, but no daily account budget.
Suggested Improvement: Employ an account-wide daily budget.
Rationale: Some advertisers have very limited budgets, perhaps $150 per day, this budget divided across many campaigns (where each campaign is like an “ad group” with different targeting), often 10+ campaigns. The only way to ensure we, as advertisers, do not exceed daily budgets, is to make all the campaigns’ daily budgets not exceed the overall daily budget, providing $15/campaign, which we then tweak based on the size of the targeting and response. The current structure forces us to throttle some campaigns that may spend more with budget while giving others too much budget.
NOTE: If a structure similar to AdWords was in place, a campaign daily budget would suffice because there would be “ad groups” within the campaign.
Paused Status Upon Creation of Ads & Campaigns
Current Status: Facebook ads and campaigns become active upon creation (of campaigns) and approval (of ads).
Suggested Improvement: Ability to create campaigns and ads in a paused state.
Rationale: Advertisers need the option. In addition, advertisers should be able to edit an ad and its targeting and keep it paused. Currently, if we alter an ad in any way (aside from its name), it will go through an approval process and become live again. If an ad is edited paused, the ad should return to a paused state once approved.
Power Editor Specific
Power Editor Desktop Application
Current Status: Chrome add-on.
Suggested Improvement: Make Power Editor an off-line desktop application.
Rationale: Power Editor as a desktop application would make it faster, more responsive and more reliable (like Adwords Editor) rather than just a web app. Even when making Power Editor full-screen, it’s difficult to work in, and we are constantly resizing panes to see full data.
Power Editor: Find & Replace & Batch String Manipulation
Current Status: No find and replace option for Power Editor.
Suggested Improvement: Find and replace feature for Power Editor within selected campaigns, find and replace based on pre and post variable rules.
Rationale: Currently, the easiest way to spawn and duplicate many ads for different targeting segments is to duplicate the ads, place them in another campaign and edit each ad manually and individually. For sophisticated advertisers, this includes ad naming conventions, URL tagging, and potentially elements within in the headline and body copy. A find and replace feature would save advertisers significant time.
Reporting & Metrics
Measure Percentage Of Targets “In-Community”
Suggested Improvement: Allow advertisers to reference multiple ads’ demographic targeting (Interest Group) from within Insights analytics to calculate A) % of demographic in first degree of separation against community size and interest group size (liked) and B) % of second degree of separation (friends of friends) against community size and interest group size (liked).
Current Status: We need to calculate this crucial metric by hand, undertaking demographic research on friends and friends-of-friends.
Rationale: When building a community of users from certain targeted demographics, important questions answered need to include, “How many users that have liked my brand’s page have interests we are after” and “How many of their friends” interests are also congruent?” In coming days, we’ll write more about this emergent metric and its importance to marketers. Understanding community focus and density has everything to do with FB organic ROI in the subscription model, after the “Like.”
Campaign Metrics – Expanded Look
Current Status: Advertisers can only look at 28 days of data within the UI, and it defaults to “Past 7 Days.”
Suggested Improvement: Advertisers should be able to look back on as many days of data they wish and set a default they wish to see.
Rationale: Advertisers know there’s a quality score that is applied to Facebook ads, though Facebook will not reveal what impacts it. Defaulting ad metrics to only the past 7 days is misleading – advertisers may think, “Wow, we’ve got a great CTR!” But looking back at the past 28 days (preferably lifetime of the ad) would reveal a vastly different CTR. This leads to a distrust of metrics, additional legwork, and does not provide an accurate or relevant “snap-shot” of ad performance.
Advertisers should also be able to choose the number of days data is displayed for. Some advertisers want to look at day snap-shots, some a week, some 28 days, and some lifetime. Make everyone happy and make it a choice.
Strength Of Message Reports
Current Status: Hire a mathematician. Tag the titles of ads with string delimitation. Do the math. Only smart people with time on their hands or a budget can cull these reports as it stands.
Suggested Improvement: Offer marketers the ability to pull top images, top headlines, top body copy as indicated by CPC, CTR, Conversion (someday…), CPM, etc. Phase two should offer pivot table options for viewing top creative elements.
Alter All Campaigns Metrics
Current Status: Metrics at the account-level (all campaigns page layout) are not action-worthy for advertisers. Currently, the metrics are: campaign, Status, Start Date, End Date, Budget, Remaining, and Spent.
Great. So we have three metrics that have to do with how much money advertisers are budgeted to ship to Facebook that day, how much is left to ship them, and how much advertisers have already sent. Additionally, as an account manager who never schedules campaigns (things happen with clients, product launches get delayed, media spend is late, and they want to pause campaigns at will) we have two more columns that are completely useless.
Suggested Improvement: Metrics at the “all campaigns” page or at the account level should be customized, or at least feature metrics advertisers can take action from: Status, Budget, Spend (all good), Clicks, Impressions, and CTR should at minimum be there as well.
Rationale: A more robust dashboard at a high-level will empower advertisers to take action in campaigns.
Desired Data in Reports: Negative Feedback, Ad Position, and Quality Score
Current Status: Ad reports provide basic data.
Suggested Improvement: Additional data such as: negative feedback (when users X-out of ads), ad position, and quality score for each ad would be fantastic additional data Facebook could provide that would lead to more sophisticated insight for advertisers.
Rationale: As stated before, often what users do not like (X-out) can be as telling as what users do like (click on). If advertisers can identify more quickly what ads users are find offensive, misleading, uninteresting, etc. they can then optimize the ad more efficiently, effectively leading to more ad spend.
Ad position is another metric from which advertisers gain telling insight. What exactly is Facebook’s first ad position? No one really knows, but smart advertisers have noted the ad at the bottom of the column gets carried along for the ride while the Facebook user scrolls. Is this the bottom spot? Top spot? That’s not as important as identifying in what position the ads are being placed and if that spot is most profitable, or if bids should be raised or lowered to be optimized for prime positions.
Quality score… at least we can say we tried : )
Most Desired Data: Identifying the Interest that Got the Click
Current Status: Ad reports provide basic data.
Suggested Improvement: Provide that data that identifies which precise interest received the click from a targeting segment. Ad Parlor’s pulse self-serve tool does this… but it will cost you 10% of media spend. Ouch.
Rationale: This is the most important piece of insight that is missing from Facebook ads reporting data. If Facebook advertisers can identify precisely which precise interest got a click from a targeting segment, they can then isolate it, refine their advertising to that precise interest, and allocate more media spend to a target they know is working and converting.
Native Conversion Tracking
Current Status: None.
Suggested Improvement: Add conversion metrics to external sites just as much as internal Facebook properties. This is key because competitive platforms have native conversion tracking and FB needs this feature to be a serious player.
Rationale: Advertisers across all online marketing channels are attempting to figure out attribution and precisely how social advertising impacts sales/leads etc. If Facebook can supply advertisers with accurate conversion through the ad platform this will take out much of the mystery. We had a client spending a great deal of money on search PPC and social PPC (specifically Facebook), but ended up turning off Facebook ads because they “didn’t work.” The conversion numbers weren’t there and the attribution links were still being figured out. When we did pull the plug on Facebook total leads, sales fell, but there was little way to prove Facebook had such or any impact.
Power Editor Download Ads Account Interface Issue
Current Status: Cannot Scroll through the entire “Download Facebook Ads Account.” Scrolling feature only allowed within “Selected accounts” – if an advertiser is managing multiple ad accounts this pushes the Download button below the fold.
Suggested Improvement: Scroll feature within the entire box. Move the box up.
Rationale: The work-around for this is to make the page smaller (“command/control” + “-”)
BUG – Editing Multiple Ads in UI Errors
Current Status: The Facebook Ad UI has check boxes next to ads and campaigns within the interface so that advertisers may edit multiple ads or campaigns at once. Unfortunately, the UI throws errors if advertisers attempt to edit too many (at times, it’s been as little as four ads or campaigns that cannot be edited at the same time).
Suggested Improvement: Fix this bug please . We’ve been experiencing this error since mid-March of ’12 and it’s a huge hassle.
BUG – Power Editor Metrics
Current Status: Power Editor has many options for additional columns in its interface, but performance metrics do not display.
BUG – Power Editor Selections
Current Status: When an advertiser mouses over a desired selection, say an ad in the center pane, often, Power Editor will select an ad three (sometimes more or less) places below (sometimes above). I’ve seen many Facebook advertisers have this issue.
A Voice – Facebook Advertising News Blog
Current Status: Help section. Not very helpful.
Suggested Improvement: Establish an true blog for Facebook advertising taking the help tips further than “have multiple variations of an ad.” Highlight new features. Use it as a sound board for what’s working and what’s not.
Rationale: Facebook is famous for making wholesale changes to the UI and capabilities without nary a mention to advertisers that spend billions of dollars on FB ads. That’s not how customers should be treated. There is really no solid, advanced voice for Facebook advertisers.
Use the new blog to let advertisers know what features are coming and why, what fixes are being employed, and what new targeting options there are. Facebook unrolled handfuls of broad categories and told no one about them!
The broad category “Auto Intender” just showed up one day… with no notice a la “Hey, we have a new cool broad category!” or insight as to what it was, something like: “Users in this category are likely to be in the market for a car!” Without an announcement, the Chicago car dealership, who had tried to use Facebook before but didn’t quite figure it out or think it was a right fit for them, isn’t going to know about this new option. That dealership has already written Facebook ads off and does not know there is a reason to check it out again.
These blog posts don’t need to be the Magna Carta, but it is news when Facebook provides a broad category for ethnicities. Whoa. I swear y’all read our posts .
Heart/Stethoscope © lenets_tan – Fotolia