Leave a comment
Get the GH Bookmarklet

Ask GH

Hi members,

Need help for FB ads. What is the best way to, measure ROI for any Facebook ads?

  • DD

    Daniel Daines-Hutt

    over 3 years ago #

    Hey David, great question and one that most people don't understand/miss

    I've yet to have a coffee but let me see if I can help you...

    So the ROI on a campaign is subjective to the business and the product being sold

    Lead costs and traffic can greatly vary depending on how expensive the product is

    Likewise how much you can spend

    A more expensive product and you can expect much higher lead costs, higher CPC etc but its fine as long as its profitable-easy so far right?

    It also depends on how deep you go with your metrics

    I like to measure ad cpc > traffic generated > traffic to subscriber >Subscriber to sale

    If I know the numbers this far back, then I can have a much better idea of how much a sale is worth to me AND how much I can afford to spend on ads (Or any other channel)

    You can get really deep with this so I want to keep it as simple as possible

    I don't know your sales cycle but mine is pretty much as I mapped out before:

    -People visit and read content
    -They become a subscriber
    -They buy via an email automation

    That's why its important for me to know those metrics at each point as its unlikely for me that a paid ad today results in a sale in less than 30 days

    But by knowing how long that whole process takes, and measuring those conversions I can estimate how effective an advert is


    At the start you need to know how much a customer is worth to the business

    It doesn't hurt to also know the Lifetime Value of the customer-how many times will they buy from you during their lifetime?

    Once you know these things, you can start to calculate an acceptable customer acquisition cost i.e how much you are prepared to pay for a customer

    This can also give you a cap on the MAX you are willing to pay so that you can adjust campaigns etc to suit

    Again, it also doesn't hurt to get a better idea of your sales ramp up time (I forget the correct name for this)

    Basically how long it takes once they go into your sales funnel to eventually become a customer


    Because you don't want to drop 10k a month in paid ads if it takes 7 months to see a sale

    1: You may go out of budget and put yourself in a bad position cash wise
    2: You might think ads are not working at all, when in fact they are driving traffic > subs > eventual sales etc

    So once you know how much you can afford to pay, you can then figure out how much ROI your campaign is creating

    You can even take this much deeper and start to measure how lifetime value is affected by channel acquired etc and see if one channel greatly affects overall profit

    (But if you are running ads this is not really your task, this comes down the to the business tbh)

    For Example (And this is very rough)

    I have a $297 offer

    I know that 4% of my subscribers convert to a sale (Approx) and that the sales cycle is around 45 days (For ease lets say 1 month)

    And I also know that the repeat value per customer is around $3,000 in total

    So that's how much a customer is worth to me over their Lifetime, not including new offers (And again i'm grossly simplifying this)

    So if I run an advert and get 100 subscribers, I can roughly say that will make 4 sales

    4x$297 = $1,188
    I also know that this will eventually become $12,000 on average

    But I also know that I wont see that first sale for around 30 days from running the ad AND I know I won't see the rest of those sales until around 6 months or so later

    This gives me an idea of how to budget during those 6 months etc and also helps me start to measure the campaign

    So from that I can start to approximate both how much I can afford to spend on a customer

    But like I said- I wont get a sale for 30 days so how do I measure it?...

    By knowing my traffic to email conversion AND my email to sale conversion I can get rough idea to how a campaign is running, months before the sale comes in

    If I know each average conversion through that process then I can understand roughly how much:

    A customer is worth
    A subscriber is worth
    A page visitor is worth

    All of this will greatly help you measure and give peace of mind for your CPC etc and also how much you can scale it up

    As a rough number my MAX CAC I would be willing to pay is around $180 per customer (and break even on the initial sale knowing that over the lifetime this will grow)

    Do I want to pay $180?
    Heck no, not really, but if the market is SUUUUPER competitive, then I know I could spend this if I really had to

    So any sale below that CAC of $180 is already a positive ROI

    In the SAAS space they will happily pay 20-30% of CLV to acquire a customer-but in fairness their running costs and CLV can run for much longer

    I've yet to find a good rule of thumb in a physical product space but I would estimate around 5% of CLV based on costs/profit margin etc

    (This is assuming you do actually upsell, crosssell and repeat sale to customers and you would be crazy not to as these are the easiest people to make sales to
    Otherwise go for around 5% of the initial sale)

    So for me I use around 5% of CLV as my target to aim for:

    5% of 3,000 = $150 per new sale

    But again I have a 30 day sales cycle

    I wont get a sale from the ad today BUT I will get traffic

    So if I know the value of a subscriber I can then calculate the value per visitor and see if my CPC is in a range I can afford to keep paying

    Still with me?

    That's why rough estimates of 'How much is a lead worth' or 'how much should I be spending on ads' is totally subjective and its why you need to know this info

    If more business owners knew these details they could sleep much easier at night #makestatssexy

    So lets look at what a profitable CPC would be for me

    My current conversion is:

    Subscriber to customer 4%
    Traffic to subscriber 10%

    These are not average numbers

    So to get that 1 Sale I would need:

    25 new subs

    Which is around 250 visitors at current traffic to email subscriber conversion rate

    For most sites this would be waaaaaay more and thats why improving conversion rates at each point is so valuable

    (Assume I had half the results we do right now i.e 2% sales conversion rate and 5% traffic to sub rate, I would need 50 subscribers and 1,000 visitors- a HUGE difference in long term results)

    So with those current stats I know that 25 subs are worth $3,000 after 6 months so...
    Each sub is worth around $120 each

    (Again these numbers are greatly different to what most people would see- my product is much higher than most AND our conversion rates are way above average)

    And from that I can estimate that a single site visitor is worth $12 each

    $3000/250 = $12
    250 visitors = 25 subs
    25 subs = 1 sale etc


    As long as my CPC is lower then $12 then the campaign is profitable right?

    Ecom version

    2% of traffic becomes a sale
    Product cost is $100 (Lets say a pair of shoes)

    Top range CAC would be $5 per sale (5% of total initial sale value)
    100 visitors = 2 sales = $200
    1 visitor = worth $2 each

    So as long as cpc is <$2 per click, at 100 visitors the campaign breaks even
    Also this is 1/2 what acceptable top CAC would be...

    Here are some good articles on this:






  • YK

    Yannis Karagiannidis

    over 3 years ago #

    You tag the traffic at a campaign level and you monitor the activity of the users that medium brings you in. Can't give exact instructions since I don't know your marketing stack

  • JA

    Justin Adelson

    over 3 years ago #

    There are two great answers to this, but I figured I'd do a quick Cliff's Note version.

    The simplest way to do this is to avoid vanity metrics like engagements and "All Clicks" (i.e. not link clicks). Yes, these are nice to see, but unless you can prove that they result in a conversion, they are just pretty to look at.

    Next, establish your goal - I'm going to use paid conversions for this example.

    Next, what is the LTV of your customer? My fake product is going to be Justin's Waffle Maker 5000 - it is an amazing waffle maker that lasts a long time and costs $149.99. To make it super simple for me, I am going to make the LTV of each customer at $125.00 (this post is not about LTV but about calculating ROI).

    My goal in paid advertising is to get my COCAs at 1/3rd of the LTV, which means I am looking to capture each acquisition at around $42.00. So, every $1,000 I spend on FB ads, I am looking to get around 24 new sales, which gets me about $3600.00 in revenue or a 260% return on my marketing investment (note: this does not factor in overhead costs, shipping, etc.).

    Say I didn't do enough market analysis and people do not want the JWM 5000. I spend $1,000 on ads but only get 5 sales at $200/sale - the total revenue is $750.00; I've lost $250.00, my ROI is -25%.

    (return - cost)/cost = ROI

    In summary, pick a mid- to bottom-funnel metric as a goal, use your LTV as a baseline for success, and use the formula above to determine how much of a return you received from your campaign.

    Again, this is the Cliff Notes version. I am sure there are days of notes that can be shared on this subject, but I like to use the above as a guide.