30
69 Comments

Spent $2,000 on Facebook Ads

Since I started to do Facebook Ads, I've spent $2,000. More than half of that in the month of August alone.

June: $110
July: $794
August: $1,200

The Return of directly attributed purchase via FB ads has been about $890

So why should I continue?
Why did I increase spending in August?
Why am I spending more than 10% of my revenue on FB Ads?

I'm spending more because I'm discovering new ways to use FB. By executing the fb ads myself, even with some advisory help, I'm the one doing each and every step. I've discovered many more angles, audiences, ad types, that I can test.

IF I had spent, let's say $500 and got 0 sales, then I'd stop, re-evalutate, but really really stop and do something else.

But Facebook gave me a little bit of a win. Just enough to keep doing it.

Originally I intended FB ads to be a channel for finding new customers. But I realized only a week or two ago that it's powerful as a reinforcement ad channel. So after a brief downturn in ad spend (less than $10 per day) I upped my ad spend again to include Retargeting campaigns. Have they worked out yet? No, but I only spent $15 so far. Need more traffic to retarget more :)

And when I was just starting on FB ads I came up with a ton of different campaign types (freebies, trip wires, lead magnets, videos) And tried them out. Only when I focused on one, did it work (videos)
But now that that worked, I'm going back to the campaign types I came up with more than 2 months ago and trying them again. With new knowledge of how I can set up campaigns, ad sets, and ads. And how I can test it out.

Here's the ultimate reason I'm going hard on FB Ads now, I'm sick. Like I posted about a week ago, I got sick midway through August and I haven't been able to make videos, do email marketing, etc.

So adding a new source of traffic is perfect for this moment in time. I can check it 10 times a day instead of checking twitter.

Will I consistently spend 10% of my revenue on Ad Spend? No.
My goal will be to get $1k to $2k in Gumroad and Stripe Sales each month. Use that cash to spend on FB Ads. Since AppSumo pays Net 60 terms, I'd rather not rely on that to pay off my FB Ads.

So in the case, this week I'm adding more FB campaigns that go directly to gumroad instead of to my site. This should increase the Gumroad revenue. (I say "should" but we'll see when it happens)

Been paying between $10 and $150 per Lifetime Deal conversion. You can see why that might suck since it's priced at $99 and I'm only making 70% of the revenue (via AppSumo)

For freebies, I'm spending around $1 per conversion. This needs to convert only 1 out of 50 times to a lifetime deal (eventually) to be barely profitable. But can become break even if 1 out of 10 buy something on Gumroad (between $15 and $50 dollars each)

And I think that I have a great baseline to compare new tests against. If I want to test a theory, I can run $10 through a test, instead of $100 if I needed to test something before with no baseline.

My theory here is that now I can iterate much faster. I can create new offers, and try different tactics such as pushing ads to YouTube videos instead of directly to a sales page.

It will be interesting to continue to work a little bit at a time on this marketing channel that works while I sleep.

, Founder of
Better Sheets
on August 30, 2022
  1. 12

    But Facebook gave me a little bit of a win. Just enough to keep doing it.

    Isn't that how Vegas keeps you hooked?!

    I sunk ~$1.5K on FB & ~$1K on Google Ads. Decided never to repeat the mistakes!

    1. 2

      Yeah $1.5k on FB and $1k on Google and I'm guessing with no real serious investment into training ($600+ say into 121 training or courses at least).

      That's like saying I posted 20 mailers in the post / went to one networking event / sent 20 cold emails.

      That's such a tiny amount to try to learn what will and won't work for your brand, even if working with a professional in ads.
      You need to learn what's the right audience / copy / creative / platform / placement / time of day etc.

      Say $1,500 gives you 750 clicks. 750 / 4 audiences = 187 clicks per audience. You likely barely would get 1-5 conversions per audience which are far too little numbers to answer which is best. Same goes for all other factors above.

      And they all need to stack up their learning to really build up to success. Once you learn which audience is best that goes from 0 to 5 conversions p/m. Same process with images gets you 5 to 10. same for copy 10 to 20 etc.

      You can get lucky but it generally takes a company months to figure this stuff out and requires someone with actual training in the platform.

      1. 4

        Precisely. Digital ads are for funded companies that can splurge a gazillion $$$s on learning what works. Bootstrapped startups are better of putting in their time and money on content, which will drive organic traffic for years to come.

        1. 3

          Well if by Gazillion $$$'s you mean about £1,500 a month for say 1-6 months. Plus either the cost of a good agency or 121 training for yourself to do it.
          I know a lot of bootstrapped startups with easy cash available for that much anyhow.
          But yeah. It's not free like SEO. But then it takes 10x less grunt work than SEO too.

    2. 1

      Haha that's a good point

    3. 1

      Sure, if it is clearly a gamble, it's a gamble. But from what I can tell the consistent return on investment is attainable in creating better offers, better copy, better ads, better targeting. I think the analogy is, if we're going with "gambling analogies" that it's closer to poker, than slot machines.

      At least the thinking that I'm going with at the moment is that I can improve in all of those areas.

      If I continue to spend and only get paid net 60, then I'll have to curb my spending. But if I can get at least the ad spend back within 7 to 30 days then I'm on a profitable consistent path.

      Here's another way of seeing it: I was running a newsletter a few years ago and I was making roughly $4 per subscriber per year. In my current set up I'm making about $21 per customer LTV selling a product/membership to a video library. And I think I'm still undermoentized since I can add in workshops, and consulting at a higher rate than I'm doing now. Spending most of my time on the video library, and customer service. But can switch to consulting and workshops to increase LTV later. And it would always be great to have more customers when I do run workshops.

      Of course I have a wider amount of products, prices, and customers than I had when I ran a newsletter but it's a good baseline. It makes me realize that no matter what if I can convert FB Ad spend to at least newsletter subscribers I have a monetization channel and I know my COGS vs ARPU.

  2. 5

    I have not used any facebook product for the last 4 years, curios about the types of users fb ads can help reach.

    1. 1

      With 2.9 Billion users, it's majority of internet users. For me it works out well because there are 1.5 Billion Google Sheets users. It's quite a large "niche", right?

      Even if there's maximum 30,000 to 50,000 people you can target, I think FB ads is useful to get them to come back to your site. Retargeting is a no-brainer for reaching people who already know your site.

      Not just retargeting and moving to a paid user, but also to bring back old users, to upgrade.

      I'm using targeting now for upgrades from website visitors to free members, AND from free members to paid members.

      1. 3

        Yes and no, I think.

        Sure, there are a ton of users still active on facebook and its affiliates.

        But how many can you target with $2k budget?

        P&G, Ford and alikes spent billions budget social media ads every year, $2000 probably is like nothing comparing to them.

        As a indie maker, just wanted to call out that FB ads may or may not work for ya.

        If it works for you, great, I am happy for you. But 2.9 billion users? Nowhere your 2k budget will reach 1% of that.

        1. 0

          well I guess I can share with you my own experience. I reached 418,000 people with $2,100 spend.

          I put a screenshot here: https://twitter.com/Kamphey/status/1564854215876702214

          1. 3

            Good for you, if you don't mind me asking (never used FB ads before so forgive me), what do you mean reached 418,000 users? Have they all click on the ads?

            1. 1

              Reach on fb is the amount of people who saw an ad, impressions is the amount of times they saw an ad. Link clicks are link clicks. Link clicks are not in the screenshot I shared on twitter but it's been 9,495 link clicks so far.

              1. 1

                Cool, so less than 10k people saw the ads (according to FB), not 2.9 billion nor 418k users.
                Just curious, how did you compute the ROI and why do you think it's a good idea for other indiehackers to do the same?

                1. 1

                  To get ROI, I add up all the attributed sales and multiple by the price.
                  I think it's a good idea at the very least to retarget people who are likely to buy from you. So retargeting visitors, is the simplest thing. Sometimes people just run out of time and need a reminder or two that you exist.

                  When I was buying a ruby on rails course last year, I kept putting it off. It's a $49 sale, but I got a ton of fb ads over weeks reminding me it exists. I kept remembering it. And eventually bought it. Great for the course creator and great for me. I went on to make CouponCodeMaker .com after I learned Ruby on Rails. yay!

                  1. 1

                    Sounds good, thanks for sharing and appreciate it.

      2. 0

        That's a great one Andrew.

        1. 2

          Thanks Jossy. anything in particular you're interested in knowing more about?

  3. 3

    What I did at the beginning of the evolution of my startup, (as I knew paid ads would be important for my business) was to provide google ads and Facebook ads for small clients in my country as a freelancer.

    I got 3 small clients, worked for them for 3 months, learned Facebook and Google Ads and then quit.

    A win for both sides, I didn't charge much, delivered a good work, and learned the tools.

    1. 2

      This is an interesting path. I like the hustle. Charge low, get in on the ground floor, get paid to learn and learn by spending other businesses' money.

      I have been using growthmentor.com and reached out to some mentors on there with PPC experience. Had calls with them. One person who gave me a great campaign structure idea I've followed up with almost every two weeks to share what's working and what's not. Each time he has some insight for me that I would not have picked out.

      And each week he can see the results. So he's learning as well what's working and what's not, by my ad spend. While he does this day to day for his work, I'm at least a few more bits of data.

      The experience is helpful: having to report to someone what's working, and what's changing as I do send him a loom video when i get a new idea, and set up a new thing, and if I see surprising results.

      I'm sure there are FB ad masterminds I can join, but I do like specifically interfacing with someone who is an expert, not just someone else flailing around on fb ads.

      And I've been watching a LOT more youtube videos of fb ads, This guy Ben Heath seems good and I like this structure as well as how he presents some ideas.

  4. 2

    Paid ads are definitely a money game! We have a minimum entry point for our clients of a monthly ad spend of £1000, soon increasing to £2,000. The more ad spend, the more ammunition we have to fire through our A/B testing matrix and discover what does and doesn't work for a business.

    We usually tell people to go in with the mindset of you might not break even until 2-6 months in, but once you do it's so worth it and is one of the most scaleable models out there!

    I would definitely be wary of any advice Facebook or Google reps give you...their main priority is making Facebook or Google more money.

    Obviously, it's not right for everyone, and there are SO many elements to paid ads to understand. It really is a specialist skill in digital marketing and takes time to learn all the nuances, but in my biased opinion, it's one of the most powerful ways to scale your business online.

    1. 1

      Looking at the experts, they all have this minimum spend, and I think that quickly correlates, in people's minds, as a minimum spend to be successful. And I myself thought for a while that I had to spend $800 USD or more each month to be successful. But that's not true at all. Simple retargeting campaigns can be extremely profitable at less than USD $10 a day and leave it at that. No need to spend money on new cold customers if you have other channels and a limited budget, right?

      I agree 100% that it's scalable. but this too seems like a myth. The more I read, watch about FB ads and learn by doing, that there is a point at which ads go "stale". Seems like each and every audience, plus ad has an S curve. The cost per conversion ends up going up for each individual pairing. So from this I infer that I need to be vigilant and continually improve the copy, improve/change/test the targeting, as well as improve/change/test the "offer".

      What I like about this is that FB ads seem like one thing to those who don't know much. I hear the phrase "FB Ads" thrown around as a negative in most indiehacker circles. But in actuality it's quite a lot of parts, quite a lot of moving pieces.

      And I'm so glad I'm in the middle of it right now with not a great offer and not a great priced product. As I increase the price over the next 12 - 18 months and change the offer, I think it will only improve. And the base line has been set. I know I can get new audiences on my freebies (in droves now) and I can get existing visitors back to the site with consistency.

      Regarding your mention about reps: I have heard this too from others, so it's no surprise. FB Ad reps are literally trying to get you a "win" as quickly as possible with no connection what so ever to your revenue, or your own goals. For example they will tell you to do awareness, and traffic and conversions campaigns. Meaning they can show "results" as impressions, and link clicks very fast, so you think well I can convert these later! Where the people I've talked to that run FB ads (not FB ad reps) will say you need to set your goal as to what you really want, conversions. sales. money in the bank. Then work backwards from there. There are other goals you can set to test, but first test the sales!

      Also, I set up times to speak to FB Ad Reps twice and both times an hour before the call they cancelled. SO WEIRD. it happened twice already so I'm just not even trying to get on a third call. As an Indiehacker dad, I set my calendar for the day and it's done. I do it or do not. I had an hour to go play with my son, but couldn't plan anything else beyond that , last minute, because I thought I was going to be on a call.

      1. 1

        Yes if you want to just stick to just remarketing and have other channels that are funneling large amounts of traffic into your site then yes, that's a way you can spend less. You would still do better at that too with more money but yes. All agreed there.

        Cost per conversion often looks good in the first few days or week as Facebook we think has a sneaky tactic around that we think.

        Generally though the good campaigns where you've really nailed all the factors can keep good for months and in some rare cases even years.

        But yes there's ad fatigue. It's obviously really. If you keep showing the same picture or video to the same users then it gets boring. So that's why part of management is to come up with new angles of that.

        It is scalable just as an obvious fact as there's 10's of thousands of companies making a fortune of it right now today. We have a client who made $75k from $4k in month one, $125k from $8k in month two and $160k+ from $15k in month three. We have a client who has seen over 10x ROAS consistently from their £5k p/m for a year and a half, making many millions in revenue.

        Yeah your own offer is the biggest factor on what will end up working. We have some clients start with us who jump up to 10x and others who struggle to get past 1x and we've not done anything differently, just some companies value proposition is stronger with more powerful brands behind them.

        Yeah just leave the FB reps! Not worth it :) Feel free to DM me if you really get stuck on something.

  5. 2

    More and more I am convinced that paid ads are generally a great approach for indie makers because you have to spend a big chunk of your revenue to get some results.

    Two of the members of the WBE Space explored ads (both FB and Google) and they reached the same conclusion (Google performed a bit better than FB btw).

    If you want to hear their experience I recorded an interview with them

    Anyways, it's important to keep experimenting and I really appreciate you sharing your journey with us

    1. 3

      I can't understand the indie hackers argument advocating the non-use of paid ads.

      Of course, try not to depend on them, work on your digital positioning and SEO, but USE paid ads, and use it strategically.

      1. 1

        When earning income that's seemingly hand to mouth, it's hard to consider doing anything other than spending time on something. It's hard to convert time into money.

    2. 1

      In my particular case, I'm glad I don't have to solely rely on comparisons to other indie makers. Since my core offering is a tutorial library which happens to come with templates and tools and scripts, I can compare nicely with course creators. And other experts who offer workshops as well. These high ticket items are doing great on FB. I looked at a digital agency like Agexa and see they spent $35M on FB ads to generate $70M of revenue. Great they essentially got a 2x ROAS, but look at that revenue number after the ad spend: $35M.. wow.

      I recall someone speaking at a little event once who spent $90,000 in Facebook ads to make $100,000 and I thought . eeke, that's a lot to spend. But then I realized. well at that point I'd love to have made $10,000. And getting customers always leads to more revenue. Because some course creators can sell again and again.

      A person converting at $100 in ad spend to get revenue of $100 sounds terrible. But that same person can buy another course at $100.

      And there are audiences out there that just won't see your tweet, EVER. FB and Goog have audiences that are just not going to see your IH post, or your reddit post. they just won't. But they will see your FB ad. or Google Ad.

      1. 1

        Great point. Thank you Andrew

    3. 1

      This comment was deleted 2 years ago.

  6. 2

    Hi Andrew. Thanks for sharing. Last year I was running FB ad campaigns and I found the book "Ultimate Guide to Facebook Advertising" by Perry Marshall very helpful. They discuss practical ways how to set up different types of campaigns, retargeting etc.

    One of the things i found most helpful was how to set up a "funnel" type campaign e.g. make "Video Ad A". Target users who watched 90% of this video with "Video Ad B". Then target them with a different ad etc. They also included things like "Users who landed on your website but didn't purchase".

    1. 1

      That book looks good. I like the subtitle from an early version: How to Access 600 Million Customers in 10 Minutes. that number is like 2.9B now :)

      I did set up a video campaign to get people to watch many videos and then retarget them when they do watch 75%. then I use a Lookalike audience to add more cold audience to the mix that are higher possibility to convert to the 75% watched :)

  7. 2

    In general, $2,000 for Facebook is just a regular monthly budget.

    In my practice of working with targeting specialists, I noticed that they could not attract leads because of the limited sample of the target audience.

    The thing is that my director did not allow me to spend more than $ 400 a month. Therefore, I came to the conclusion that only large investments will be able to show a profitable and payback channel or not.

    1. 1

      I'm definitely trying to work my way up to $2k a month. More than half way there in August. but will try to pare it down below $1k for September until I can get some cash faster. I think after 3-6 months of consistent ads I can keep going up in budget as the cash from AppSumo comes on net 60 terms. so in 2 months I get this month's payments.

  8. 2

    So when you look at it your ROAS is around negative 60%? You spend ~USD2.40 on Facebook advertising to earn USD1.00?

    That sounds like a horrible deal. But since you are getting something out of it (i.e. knowledge & experience) it might not be a big deal. But keep an eye on those numbers.

    But Facebook gave me a little bit of a win. Just enough to keep doing it.

    This is probably their "game". Almost like a casino, you should feel like getting a bit of a win just enough to continue.

    I am not saying you should not advertise on Facebook. But you should be extremely specific in your targeting.

    But I am very anti-"let's advertise on FB and call it marketing", so take my views with a grain of salt.

    Finally, a question, did you compare FB numbers with your organic numbers? Or with other channels? How do they compare?

    1. 1

      My organic numbers are weird. Because I do PR, Blog, SEO, Marketplace (appsumo).. so generally speaking I have a $0 cost for marketing. Obviously I should figure out my time spent on each of these things and factor that, but as a solo person I just do more of what is working, and what I think genuinely will help more people. Most of the PR I do is podcasts and interviews about my journey so it's not directly affiliated with marketing to the product.

      To balance the cost of FB ads, I am also considering adding more costs like to specific writers for certain blog posts or parts of my site I want to use for SEO. For example also considering applying more dev costs to create SEO-only pages. So there will be a cost there.

      I have written down 17 things I can do consistently and that are all FREE cost. The only real cost is my time. I can scale each and every one of those up with additional spending sure, but with each one I would also spend more time. For example I "scaled" up twitter by writing for 1 month solely on twitter. It worked but I couldn't do it sustainably.

      And I'm basically doing a month hard on each marketing channel

      1 month on Twitter
      1 month on YouTube
      1 month on Facebook ads

      And each one isn't 100% based on ROI. I also base my decision to keep going on factors including "Do I like it?" "Can I test more?" "Can I continue doing this myself?"

      The most important part of the decision ( this year) is that the marketing channel can exist asynchronously in my life. Oddly scheduling a lot of tweets, is async, but the generally scaling on twitter is not async. While I can set up a new FB campaign for 1 week and literally leave for a week, come back and see how it did. I can pick it up and check it any time of day and there's no downgrade of experience.

      To get back to FB Ads and your question about ROAS, I posted my overall ROAS, but I dial in and can see which campaigns and which ad sets and which ads actually drove the revenue. I started off with a bit too many ideas. And that's where I can identify overspend. In campaigns that I thought would work, spent $80 - $120 on them with 0 results or rather they got 1 result and then two weeks later, nothing.

      There's a weird thing that can happen: it's been stated to me that I can "break" the profitability of a campaign by changing the budget too much too quickly. And another weird thing is that I can end up with campaigns that never leave the "learning" phase because I'm testing too low of a budget.

      It's a matter of dialing these in correctly to make the most of my ad spend month to month later.

      for example I tested out lead magnets early on and got 0 results at $1 per day. Turned it off. Then recently (in the past 2 weeks) I just created a new campaign, new ad set, new ads, and spent the right money. ($10 per day across 10 ads) Because I know I can spend $10 in one day to test it out or $70 in a week to test and can cut it off if nothing. But if I spend $1 per day and get nothing... I'd have to wait 70 days to see the same results. Also it might not ever convert because each ad is an auction.

      1. 1

        My organic numbers are weird. Because I do PR, Blog, SEO, Marketplace (appsumo).. so generally speaking I have a $0 cost for marketing.

        Yeah, when you compare organic(owned+earned) vs paid, organic is always $0. When I asked about the comparison of numbers, it was more in the line of funnel(or flywheel, or whatever) metrics vs FB. How many arrive ➡ how many interact ➡ how many take action ➡ how many convert. Are they wildly different? What would you consider your best channel in that sense?

  9. 2

    Facebook Ads and Google Ads both work great all the time for 100's of thousands of businesses around the world, if managed correctly. I can give a long list of examples.
    Two reasons it doesn't work:

    1. Your business hasn't achieved product-market fit or basically, you don't have a value proposition that actually appeals and converts.
    2. You don't execute on the platform correctly.

    People are right below to say if you're not yet delivering sales elsewhere in a meaningful way and feel you have product market fit, then paid ads are an expensive way to learn. So best to get some initial sales from other 'grunt work' style channels.

    Very often though it's just bad execution that's the issue. There are countless gotcha's now built into Facebook and Google where they tell you to do things a certain way that isn't in your best interests. Google reps are the worst. They destroy accounts.

    You need to hire 121 regular coaching or outsource it to professionals. After practising a lot of other marketing channels I specialised in paid ads as it's what I found to be most reliable at scaling businesses in a predictable way, as long as they have an offer that appeals and the margins are good.

    When I run audits on accounts run by self-learners, it's a huge gap between them and full-time paid ad professionals.

    The last point I'd make is you need to just spend a lot more than you think or at least leave tests running longer. Your budgets you talk about are small and you're likely often selecting winners from your AB testing with too little data.

    And ultimately getting up to $2k+ ad spend a month is when you should have a much better chance of enough conversions flowing to have enough data for the platform to really jump into gear. That's when they're at their best. Conversion focused campaigns obviously are more effective when Facebook/Google/LinkedIn can actually see conversions.

    Anyhow, good luck on the journey and well done for getting into it!!

    1. 1

      I would hope people are reading my product update and don't think of it as a single post. I wrote this as an update to my product that's been around for 2 years already. I don't want to misrepresent this fb ad spend.

      I've already gotten a great sales channel with AppSumo Marketplace. That's not going away any time soon. But I do wish to get out of that particularly small market (2-3 million users) and into a larger market of 20-30 million google sheet users who pay for courses/tools/tutorials.

      1. 2

        Yeah no just some others have said about when Facebook doesn't work, that would be a time. It sounds like you're far past that :)

  10. 2

    I think your theory about iteration is right on. Paid Marketers specialize in targeting different sets of customers and directing them to the products that are most valuable to them.

    As you mention you have different plans so if you can find differences in people who buy cheap plans vs expensive plans then you can create campaigns to target those cohorts and increase the return on your spend.

    Good Luck!

    1. 1

      Also consider just going hard on one specific audience that is more likely to buy than anyone else on the planet. IF that group is 30,000 to 50,000 large, then you can also target them on FB. But I would consider doing a 360 type campaign. Targeting them on each and every platform. FB, IG, Goog, Newsletters, TikTok, Twitter. It can't be much money per month to target the same people over and over again wherever they go, for 5-10 days.

      I'm working on a campaign format right now that is 10 different pieces of ad creative, but targeting one single audience and capping the frequency, so they only see 2 ads per day. This is diametrically opposed to the normal campaign theory of a few ads that work well to as many people as possible. If you want more on this, I found it on YouTube on a channel by Ben Heath. It's called "Omnipresent"

  11. 2

    This is a good promotion experience, My website may be able to try this method. I hope to push the traffic of my ai logo maker website up.

    1. 1

      I'd say, first put a link in your IH profile. I tried taking a look at your site and have no way to get to it. I don't even know the name of it. IF it's there, in your profile, then I missed it.

      What kind of traffic do you want? to what end? IF possible the best is to make money directly, but there are so many other ways to collect users, like with lead magnets, and with newsletter signups. And that converts faster/easier so you can actually get some data into FB if you've never done ads before. advertising a direct offer can sometimes just result in 0 sales. But it's always worth a shot, then add in other conversions.

  12. 2

    Thanks, Andrew, good to learn about your experiences as I consider the available channels for attracting customers.

    1. 1

      I would not recommend spending so much on paid ads with a new product. Seed the userbase, visitors, with existing marketplaces, and directories (PH, IH, AppSumo) then once you get even 100 visitors a month I think retargeting is the way to go first for FB ads. I was scared for so long to do any paid ads and didn't have any budget for it. So I'm far behind. Hence why I'm spending so much now.

      1. 2

        Great insight, thanks Andrew.

  13. 2

    So whats your CAC right now and is it growing? Thank you so much for sharing

    1. 1

      ranges from under a dollar to over $100. I have different levels of customers. freebies, lead magnets, trip wires, free sign ups, monthly members, lifetime members.

      But also technically most of my customers are absolutely free via AppSumo marketplace except i pay 30% after the sale. And I've raised prices recently so that's actually a substantial number now. Which incentivizes Appsumo to promote me more, because it earns them more. Thus my personal CAC goes down even if /when I spend money.

      1. 2

        interesting.!! Are you ROI positive?

        1. 1

          Nope. $890 Revenue from $2,000 Ad Spend. ROAS of 0.45:1

          1. 1

            dang, i had similar experiences

  14. 2

    Hi Andrew, thank you for all the information you shared. I wish you all thee best 👌🏿

    1. 1

      Thanks. anything in particular you liked? Something else I can answer, to go deeper for you?

  15. 1

    Since I have no idea what your product or your target group is, I have no idea if what you did was right or wrong.

    From my experience: August is very bad month for ads on Facebook, since lots of people (well, Europeans at least) are on vacation, and there is a (yearly recurring) seasonal dip. At least in eCommerce ads on Facebook.

    1. 1

      I sell Google Sheets tutorials, templates, and tools at BetterSheets.co

      1. 1

        Hm. My suggestion would be to give the campaigns some time. Erratic (daily) changes are usually counterproductive because FB needs time to learn what works and works not.

        Since there are like a million things that can be customized, the only thing I can further recommend is to continue learning and experimenting. For me it took weeks for the first small successes and months of continous spend to see acceptable success with ads.

        1. 1

          What was your metric for "acceptable success with ads" ?

          1. 1

            Primarily seeing a positive trend. When I started my ROAS was horrible. My CAC was insanely high. Over time however a trend developed and my costs went down.

            I can see the end the tunnel and after a few months of data I now have a much better understanding of the channel, what ads and messaging works in which location and which target group on which device. These are very important insights for me.

  16. 1

    Hey Andrew, this is excellent! Thanks for posting.

    I am in a similar segment. I have an add-on for Google Sheets for applicant tracking aimed at recruiters.
    I was wondering what kind of audience segment should I create? Any tips on the targeting for this segment?

    1. 1

      While there is no interest targeting for "Google Sheets" there is "List of Google Products" and other "Google" related interests like google workspace. I would start there.

      But I would also make at least 2 different interest based ad sets, 1 that is broad and 1 that is extremly defined. I just did this recently and had success on day for a freebie. I used "Define further" as much as I could to get down under 100,000 extremely defined audience. And I added other broad ad sets too. The trick I discovered is to use CBO to let FB test these audiences itself.

      And then I created another campaign for retargeting. I do think this is an easy mistake to make because I didnt do CBO yet. I think I actually should be making another ad set with the retargeting a warm audience, not a seperate campaign.

      Here is something that might not be intuituve for targeting: I sell a Google Sheets course and one of my best ad sets is targeting Udemy users. Everyone and their mother would have suggested "Google Sheets" as an interest.

      Look for similar price point products that your customers also buy. I even experimented with Excel users. While this wont work for an addon, it shows I think just how much breadth you can test.

      I tried education, learning, spreadsheet, and way more. Try many, try many deep cuts, then make Lookalike audiences when you get any traffic.

      Try worldwide, cut out obvious black holes of FB low quality clicka. I had to exclude Bangladesh and Nepal. FB spent a large percentage of early cash there. And I know those two countries are far down my list of customer locations. Data I get from Google Analytics, and Gumroad.

      Hope that helps.

      1. 1

        Thanks a lot for this comment! Really detailed explanation. I will be trying things out.
        Few doubts:

        But I would also make at least 2 different interest based ad sets, 1 that is broad and 1 that is extremly defined.

        Do you know how the above helps?

        Try many, try many deep cuts

        You mean narrow the audience?

        1. 2

          re: 2 ad sets
          By targeting two ad sets, you'll learn which one works better. Generally speaking many experts are saying to go broader now, but they have a large amount of cash to spend to test more, sooner. I'd prefer to get results fast, AND test. Better to test between two ad sets, each week, than 5-10 at a time. I had very slow learnings early on because I tried too many things. Try two things, see which one performs best, then that is your baseline, try one more the next week and if it performs better its' your new baseline. if it doesn't do better, kill it. try a new one.

          re: Deepcuts
          Some fb videos I've seen pretty much think that FB knows more than we do, and that is true. But we need to start FB somewhere.

          Yes, narrow as much as you can, in as many ways as you can. Use the "define further" part of interest targeting.

    2. 1

      A simple answer for you is you can stack interests. So for example interests in recruitment maybe is one and interests in google sheets. That'd be the dream haha. You can go into the interests and look around for what are the best options for you. Come up with 2 or 3 and then test them against each other and make sure to give it decent budget and enough time.

      1. 1

        Thanks Max! The interests bit is simpler but was thinking if one can make it more specific. I guess the way is to experiment around a bit there.

        1. 2

          Yes actually some of our best performing audiences are just like the entire of the USA for example. Facebook's algorithm is powerful stuff especially when you have conversions flowing. So most important thing is just come up with a selection and test against them to find the strongest for your situation which could be small or big stacks of interests or lookalikes or whatever else.

  17. 1

    For reasons (cough, cough) I'm very knowledgeable of Fb's ads delivery system.

    How did you spend? Ex autobid, manual bid with cap, what events were you optimizing for, how did you select a target audience?

    Most people have medium->good success with Autobid. CBO is probably worth turning on; this allows liquidity between campaigns, eg if one is doing really well and the other is not, your funds will be reallocated to dial the successful one up.

    1. 1

      Autobid.

      I'm testing all types of events to optimize for: mainly sales but recently have added in lead magnets, so the conversion event is the download of that lead magnet.

      I've been turning on CBO recently to new campaigns because I overthink the ad sets and ads and try to test too many theories at once. So what's happened is that too little ad spend on each ad gets no results. I turned on CBO to one of these campaigns recently and it quickly identified what worked. 1 ad in 1 ad set, and then it seems like it's trying each one each day, with good results so far. So I like CBO so far.

      But I'm not doing CBO or the ad set level one on all campaigns. There are definitely some ad sets I want to test vs each other on a day to day basis.

      What I haven't tested yet with CBO, and I'd like to is to mix Cold and Warm audiences. Instead I have just different campaigns for these and what I saw recently struck me as not intuitive but totally logical.

      what's your take on this?
      Set up 4 cold audience ad sets, and 1 warm audience ad set, set up CBO so that FB itself allocates the appropriate spend across those 5 ad sets. While there will be far more people in the 4 cold audience ad sets, the price per conversion across the whole campaign will be lower because you include the warm audience in there.

  18. -5

    This comment has been voted down. Click to show.

Trending on Indie Hackers
I've built a 2300$ a month SaaS out of a simple problem. 19 comments 🔥 Roast My Landing Page 12 comments Where can I buy newsletter ad promos? 11 comments How would you monetize my project colorsandfonts? 8 comments Key takeaways growing MRR from $6.5k to $20k for my design studio 6 comments How I built my SaaS in 2 weeks using NextJS and Supabase 5 comments