{"id":21832,"date":"2023-04-26T12:51:30","date_gmt":"2023-04-26T07:21:30","guid":{"rendered":"https:\/\/www.karooya.com\/blog\/?p=21832"},"modified":"2023-04-26T12:51:31","modified_gmt":"2023-04-26T07:21:31","slug":"ppcchat-when-platforms-meltdown","status":"publish","type":"post","link":"https:\/\/www.karooya.com\/blog\/ppcchat-when-platforms-meltdown\/","title":{"rendered":"PPCChat | When Platforms Meltdown | 25th April 2023"},"content":{"rendered":"\n

Hello readers! The PPCChat sessions serve as a valuable resource for staying informed on the latest perspectives and challenges faced by the PPC industry’s experts. This week’s discussion focused on the recent Facebook platform meltdown and its impact on PPC professionals, with Julie F Bacchini<\/a> hosting the session.<\/p>\n\n\n\n\n\n\n\n

<\/div>\n\n\n\n

Q1: Were you impacted by the Facebook platform meltdown over the weekend? If so, how did you respond?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

I don’t have any clients advertising on Facebook at the moment, so I watched this one from the sidelines. @NeptuneMoon<\/a><\/p>\n\n\n\n

We’re still trying to figure that out. a couple of major accounts were unaffected which is a big relief, but we’re still getting to the bottom of any effects across the team. @JuliaVyse<\/a><\/p>\n\n\n\n

Yes, we were impacted. There was just a little overspend, but a very high cost\/landing page view which led to a drop in traffic for that day. We didn’t stop ads, and had conversions on Sunday and Monday, as well. Happily, things look good for now. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

We were but fortunately, FB spend at the moment is low across client accounts so impact wasn’t awful. We noticed that spend was spiking at 2-3 am so altered scheduling to run solely during business hours – not ideal but seems to have reduced any mad 2 am spends. @PPC_Fraser<\/a><\/p>\n\n\n\n

Managing mostly Google Ads accounts, only indirectly. I noticed minor volatility in the “echo” of branded search campaigns. Reached out to a partners & SM buyers, some of whom (surprisingly) we’re unaware. It’s interesting how we’re so “plugged” in, but also not. @teabeeshell<\/a><\/p>\n\n\n\n

From initial data, we saw a 40%-50% increase in CPMs We initially responded by reviewing accounts and prioritising accounts that were hit the hardest. @C_J_Ridley<\/a><\/p>\n\n\n\n

Shockingly no, I wasn’t. @JonKagan<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q2: We talked about this somewhat recently, but what, if anything, do you have in place to alert you of major issues in your accounts?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Rules, alerts, and the best tool in the biz: a second pair of eyes! @JuliaVyse<\/a><\/p>\n\n\n\n

Si, senora. I have automated alerts set to make me aware of any major changes within an account. Be it spend, conversion rate, or a bunch of other KPIs I set alerts to make we aware of BIG changes. @ppcClickShark<\/a><\/p>\n\n\n\n

On Facebook we don’t have any rules set up. It would be good, as this happened on weekend when live monitoring is not possible. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

A few (email) alerts that help one act quicker include: 1) +\/- 50% spend, DoD 2) +\/- 20% CPA, WoW 3) X spend threshold, 0 conv, separately at KW, ad, ad group, & campaign levels (& camp type) 4) CTR thresholds at X imprs (matters for GAds), separate by camp type. @teabeeshell<\/a><\/p>\n\n\n\n

We have a few scripts and automated reports in place to inform us of these changing, and we’re also currently reviewing automated solutions as preventative measures… At the moment we’re not implementing too many due to the list of automation over-reacting. @C_J_Ridley<\/a><\/p>\n\n\n\n

We have analytic alerts in place that trigger every 3-4 hours. If a metric falls well below or above a certain volume, it alerts us. @JonKagan<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q3: When you become aware of a major platform issue or malfunction, what are the first steps you take?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Usually I look at spend. Do we need to pause the campaigns or not. Next, it’s best to try and understand what’s happening, and then alert the client with next steps. @JuliaVyse<\/a><\/p>\n\n\n\n

I try to dig and find the cause to understand the situation. Is it something temporary? Is it only affecting one account or is something larger, that affects others too? Depending on the cause, I take different actions. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

1. Read about & fully understand the problem. 2. Check the impact on my accounts. 3. Solve or await resolution 4. Alert client of situation with a solution if possible. @ppcClickShark<\/a><\/p>\n\n\n\n

Step one is to figure out if my accounts are being impacted, and if so, in what way and at what level of severity. See what, if anything, platform is saying. Might pause everything while it gets sorted by the platform. Let client know as needed. @NeptuneMoon<\/a><\/p>\n\n\n\n

Communication is key, never can be too much. At a minimum let clients know A) We’re aware, and B) Here’s what we’re doing right now, today, & future if\/then steps. @ShopifySupport<\/a> + other CMS support accounts are good ecomm resources for early outage\/issue alerts. @teabeeshell<\/a><\/p>\n\n\n\n

Impact. Always impact. There’s a tendency to get sucked into the detail whereas if the issue isn’t as bad as first thought, your approach to fixing it changes. Without sounding all Tony Robbins, think it’s also crucial to learn and react e.g. adding more data to looker. @PPC_Fraser<\/a><\/p>\n\n\n\n

Did anyone else find that reach campaigns were largely unaffected? it seemed to be more traffic\/conv campaigns. @JuliaVyse<\/a><\/p>\n\n\n\n

First reaction is always to analyse the data to see if we can resolve it ourselves. If we can’t, we next aim ro minimise damage, whether this is reducing budgets or pausing elements altogether. Finally, we reach out to see if others are experiencing the same. @PPC_Fraser<\/a><\/p>\n\n\n\n

Check our platform and real-time analytics to see if the issue is hitting us or not. @JonKagan<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q4: In the event of a platform issue, how proactive are you in communicating with your clients or stakeholders? At what point(s) in the process do you communicate with them?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

I am very proactive, and would rather have multiple communications as I know more than wait until I have a resolution. First communication is often – hey this is happening and we are actively on the case. More detail later about impact, our actions, etc. @NeptuneMoon<\/a><\/p>\n\n\n\n

Always if there is a big change with a big impact, and as fast as possible. Minor things that don’t affect performance can be said later, as an information. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

I would also add that this type of situation is one where it really helps to know your client\/stakeholder well so you know what to tell them and when. Some want play-by-plays, others just an acknowledgement and summary of actions taken. @NeptuneMoon<\/a><\/p>\n\n\n\n

I like to get Comms out to client as soon as I’m aware of a plaform-wide issue or if I spot an issue in their account that has had a significant impact on the account’s performance Even if it’s a “Hi, we’re currently monitoring …” Email @C_J_Ridley<\/a><\/p>\n\n\n\n

If the issue appears to persist more than 2 hours, we communicate it to the client proactively. If it is under 2 hours, it is reactively. @JonKagan<\/a><\/p>\n\n\n\n

Depends on severity of the issue & client comm preferences, but I err on the side of over-communication (builds trust). For quicker FYIs, most clients share a Slack Connection, so that’s instant. For longer explanations, email is best with a Slack TL;DR alert. @teabeeshell<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q5: In a case like the Facebook issue over the weekend, how do you handle unexpected and\/or significant overspend in your accounts?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

In a case like the Facebook issue over the weekend, how do you handle unexpected and\/or significant overspend in your accounts?” 1. Gauge impact on performance 2. Figure out how to budget for the remainder of month. @ppcClickShark<\/a><\/p>\n\n\n\n

Own it. Whether it’s your fault or completely unavoidable like an ad platform bug, be transparent and honest with the client\/stakeholder(s) @NeptuneMoon<\/a><\/p>\n\n\n\n

Happily, we didn’t have a high overspend, just few dollars. I guess discussing with the client for a solution can help. Maybe they are not very budget restrained. Also, the platform may refund that amount. Lastly, you can cut budgets on underperforming campaigns. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

First, figure out just how much overspend there was. In the case of FB over the weekend, it sounds like for some it was quite significant. Provide suggested courses of action for remainder of budget period and let client weigh in on what they’d like to do. @NeptuneMoon<\/a><\/p>\n\n\n\n

Once you’ve owned it, get it under control. Review how the issue occurred and put measures in place to minimise\/prevent it in the future. Do additional alerts\/automated rules need to be set up? Do you need to tighten targeting or set an account monthly budget? @C_J_Ridley<\/a><\/p>\n\n\n\n

I go to meta like this. @JonKagan<\/a><\/p>\n\n\n\n

Or have a lower ad budget next month. Communicating with the client will help you feel better and come to the best solution. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

I don’t count on a refund happening. It really should when the platform was at fault\/broken\/malfunctioning. But that is not what happens a lot of the time. So I treat it as a bonus part of any action plan. @NeptuneMoon<\/a><\/p>\n\n\n\n

I think it’s important to point out that you can be transparent & “own” it without accepting blame. If it’s the platform’s fault, communicate that and work on the path forward. Don’t beat yourself up over something out of your control. @robert_brady<\/a><\/p>\n\n\n\n

With smaller accounts, an egregious overspend may blow the budget for a time & “resting” the account may be the only option if no refund. Better path is, reduce testing, consolidate via proven campaign efforts, & weather the storm until budget constraints loosen. @teabeeshell<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q6: Have you successfully gotten a refund or credit for any of your accounts when there has been a major platform issue? Was the refund sufficient in your opinion? How were you able to achieve that?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

So few times I can barely remember. Usually on Meta is just some cents. On Google, it happened to receive more tens of $. But nothing significant. Both by communicating with the support or by default. @DianaAlinaAldea<\/a><\/p>\n\n\n\n

The lack of reps and quality support make getting a deserved refund way harder than it should be! If the platform goes wonky, it should absolutely refund at a minimum for overspend or other account parameter ignoring actions it took that cost $$. @NeptuneMoon<\/a><\/p>\n\n\n\n

None worth writing home about. @C_J_Ridley<\/a><\/p>\n\n\n\n

Yes. My highest-ever payout was for $375K. I do however know of a bank that took a certain engine to court over an algorithm bug causing increased CPC and got a payout of $3M @JonKagan<\/a><\/p>\n\n\n\n

It’s tough to extract spend once it’s gone. Depending on the severity (+ who’s at fault), there’s a time opportunity cost to consider. Do we bunker down for a while, or do we split time (managing vs. platform follow-up) attempting to secure a refund…months later. @teabeeshell<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

PPCChat Participants<\/strong><\/p>\n\n\n\n