Advanced Pay Per Click (PPC) Certification Program

Certification Training
2380 Learners
View Course Now!
42 Chapters +

ROAS, Conversion, & Setting Initial Bids Tutorial

2.2 Introduction

Hello, this is Brad Geddes, the author of advanced Google AdWords, the founder of Certified Knowledge, and the PPC faculty chair for market motive. In this video, we're going to get into bidding and setting our initial bids based upon return ads spent, profit. And looking at some estimated conversion rates to be able to set proper bids when you first start.

2.3 Examining the Metrics

So first, we need to look at some math and getting into the metrics whenever we're doing calculations. So we're just going to build a simple spreadsheet for how these calculations are done for the next few minutes. So first we need to start with our Ad Group and our Average CPC. This is going to give us our cost information as we build things. Now I've chosen the Ad Group in this case for these estimations. But if we wanted a lot more detail, we could use up the keyword level. If you're doing display, we can do this at the placement level or interest level. So whatever data point you really want to estimate from you can put those in the Ad group list and change it from Ad groups or to placements or to keywords. So next we need to get into derived metrics to see how all these numbers play together. So one of the first numbers to really look at is conversion rates. Conversion rate is a simple ratio of how many conversions did you receive divided by the number of clicks you received. So on our ad group chocolate here, we have a 10% conversion rate. So we have 100 clicks and a 10% conversion rate, you multiply 100 times 10% and that leaves you with 10. So conversion rates are really important number to look at because a slight change in conversion rate makes a large difference on a number of conversions you receive, which then can affect your revenue and how much you can bid based upon just one number changing. So next, when we have these conversions what we need to know is how much we're actually paying for a conversion. So based upon our average CPCs and our conversion rates, what does it cost us to actually get a sale? So once again it's a very simple number. It's average CPC divided by conversion rates. Now, this is an easy way to start when you don't have a lot of data yet. If we do have data from our count then all we have to do is take how much money we spent on those clicks divided by the number of conversions and that would also give us our cost per conversion from actual numbers. So then once we know our conversion rates and how much we're paying per conversion the next thing we need to determine is what do we actually make when we get a conversion? So our revenue per conversion's going to be our final sale price or what that conversion value is worth, minus cost of goods. Now, if you're selling digital-based products, you don't have cost of goods, so your final sale price is your revenue per conversion. Sometimes, you have to back into these numbers. So if you're a lead generation site, what you really want to look at is for every ten leads you get, you get a final sale. And therefore, you take the final sale price divided by whatever your ten leads are and that's your revenue per conversion. So if you have multiple steps to a conversion, clicks to website, then leads that occur, then leads get passed to a sales team, sales team calls up and makes a sale. Then you just need to take those final sales divided by the number of leads it takes to get to make a sale, and that's your revenue on average for each one of those leads. So depending on the number of steps in your process, you may need to change this calculation around slightly based upon all the steps that occur before you actually make money.

2.4 Calculating Profit and ROAS

So next we need to look at what do we actually make on a profit basis after a conversion. So this is what you get to actually put in your bank account. So profit per conversion is a simple number of your revenue per conversion. So, what did you actually make when you got a sale? Minus your cost per conversion, how much did it cost you to get that sale? So back to our ad group of chocolate, we make $20 every time we get a sale. It takes us $10 before we can actually get a sale, it takes us $10 in clicks to get a $20 sale, so therefore our profit per conversion will be 10. In our flower ad group our cost per conversion is $30, because we have an average CPC of $3, conversion at 10%, so it costs us $30 to get a sale. But, flowers have average higher sale prices than chocolate does. So, for our flower ad group we actually make $50 when we get a sale. So, 50 revenue per conversion minus 30, our cost per conversion, means we make a sale, we make $20. So your profit per conversion let's you know what you actually get to keep after your marketing costs and put in your bank account to grow the company. So the one of the the easiest numbers to work with in setting bids is known as return on adds spent or ROAS. An ROAS is a simple percentage of revenue divided by spent. So if you spend $100, and then that $100 makes you $100, so 100 divided by 100 is a 100% ratio. A 100% return on ad spend is just break even. Now, if for every dollar you spend, you make $2. So on our chocolate ad group, every time we spend a dollar, we make $2 back. That means our return on ad spend is 200%. So ROAS is a very simple ratio, it's a nice percentage to work with with setting bids. It fits very well into formulas. However, we often have to look at the balance between profit per conversion and return on ads spent. And this is where, in our chocolate ad group, we make 200% ROAS. And often, if you're just looking at straight ratios, this looks like a better ad group than flowers, that only has 166% return on ads spent. However, when we look at what we get to keep in our bank account, our flower ad group, we get to keep $20 per conversion. With our chocolate ad group, we only keep $10 per conversion. So while on a return on an ad spent basis, chocolate's more effective. On a bank account basis, our flower ad group is more effective. So you often have to go back and look at profits on ad groups as well as return on ad spend to determine which ones really are best. But return on ad spend is what we're going to use on a lot of our formulas going forward to determine bids. because it's such a simple way of plugging into spreadsheet programs and running formulas to determine new bids based upon what you want to make on a return ad spend basis. Compared to your average CPC's, your conversion rates, and your cost per conversion. Now with these numbers, we need to know conversion rate to be able to determine what our return on ad spend is or what our bids are.

2.5 What Makes a Conversion?

So, therefore, we really need to look at how do we estimate conversions rates. So when you first start, you won't have any data, so you have to make some estimates. Once you have some data, then you can use real information. So when we look at what makes a conversion, first off we have our traffic source, this is Google Ad Words, Facebook, Bing, LinkedIn, so forth. And then we can take our traffic source down to a keyword level or an ad group level even within a traffic source such as AdWords or Bing. So next, then, we have our offer, which is often our ad copy and then what our text on our landing page is. And then our ad gets us clicks, and certain costs associated to getting those clicks. Then we have our landing page. And our landing page then is what convinces someone who clicked on our ad and read our initial offer to want to convert. So what we really need to look at is, based upon your landing page, what are your initial conversion rates? So in this case, we're assuming you're using a good traffic source. AdWords, Bing, Facebook, they're all good traffic sources. We're going to assume you have good ad copy. And this is where ad copy testing is very important, but for now, we need to make some initial assumptions to get into conversion rates. So then based upon your landing page goal, what are some sample conversion rates that are good numbers to work from?

2.6 Simple Email Capture

So if it is doing a simple email capture, name, email address. That's it, really simple form to fill out. Now for not established sites common to see 5% to 10%. For established sites it's pretty common to see 10% to 20%, now these are averages. There are sites that have 35% conversion rates for this. There are sites that have 0.2% conversion rates. But on an average, if you have compelling text around your form, you'll often see 5 to 10 for unknown sites and 10 to 20 for more established sites. So this gets into how conservative do you want to be versus how aggressive in setting bids? If you want to be fairly conservative, well loose at low ad. 5% for unknown or 10% for established. If you want to be a little more aggressive with how you're going to start bidding, then you use the higher end of the scales.

2.7 Simple/Complex Whitepaper Download

Now we get a whitepaper downloads, again simple whitepaper downloads. Often you'll see 10 to 15% conversion rates, in simple whitepaper downloads means the information required. So again, just name and email address. Now, what happens though when you get into complex forms. This could be a complex whitepaper form, a complex lead generation form, but we'll address lead generation separately. What happens is your conversion rates may be 1% to 10% which is a really big range. The question really gets into is how personal are some of these forms you're asking. If you're asking for things like phone numbers, cell phone numbers, lots of required fields, you're going to see your whitepaper conversion rate's lower. If you ask for less important information, you'll see them higher. Now this also gets balanced with how common or uncommon the information is you're providing in a whitepaper. If you have a really complex form, and your're offering information which is widely available on the web, you're going to see 1 or 2%. If you have a very complex form and you have information that very, very few people have on the web you're still going to see 8-10% conversion rates because your information is so rare. So that's why in estimated conversion rates can be difficult because it goes to how much information you're asking for balanced with what's the scarcity of the information you're giving away.

2.8 Lead Generation

So when we get into lead generation. Your simple forms 1 to 10%, again very large range, with that gets into, are you asking for a phone number? Are you asking for specific addresses, zip codes? What are you asking very personal information? This often gets into credit score industry, insurance forms, property evaluations, mortgage applications, where you need some very specific information to be able to process the question, but the more personal you ask for questions, the lower the conversion rate. So, you may see 0.5-5% conversion rates when you get into a very personal questions. Now again there are mortgage companies who have 10, 15, 20% conversion rates asking personal questions. That's for landing page testing, very important. But often you want to start out, especially in some of these more expensive industries. Slightly conservative and then build up when you get data. Now if your lead gen is multiple pages, often you will see slightly lower conversion rates with your forms.

2.9 Expensive/Low Cost Ecommence

Now we look at e-commerce. We really have to types of e-commerce. We have expensive, and less expensive items. You get expensive items; big screen tv's, high end stereos systems and often your average order values may be more than $1000; 0.05's pretty common, 1% is really good. Now, when you look at low cost e-commerce, used books and items that are $10, $20, $30, 2% is pretty common industry standard. Now, there are sites, again, that have ten, 15, 17 percent conversion rates on e-commerce. So it's not that you can't have much higher than these, it's that we need some number to start with, and until we have actual data, it's usually better to use more industry-standard information. So that we're making some good initial assumptions, we're not losing a lot of money when we start, and then when we see the actual data, we can go ahead and use it to increase our numbers.

2.10 Local Service Companies

Then we get into local service companies. You're often looking at call through information. So electricians and plumbers and accountants and so forth. 3 to 6% is a pretty common range to see when calling local companies. Now, that is off of a desktop. Often on mobile devices, those numbers are much higher. It's not uncommon to see 15 to 20% call through rates for a locksmith off of a mobile phone. So that's really siloing our data, though, to do some of those estimates. So off when we first start, you want to use, again, industry averages, and then we can layer over information, or as you learn more, you can make better estimates for your company or for your industry as a whole. Now lead into local business calls. Often with a store, call through rates are really low, because most people just want driving directions to get to a store. However, a restaurant's also a local business, but if you have a high-end restaurant, often people are calling to make reservations and get their table for that evening. So often restaurants do have higher conversion rates than a local store does. So if you have a company like a local store which you really want to measure, it's not just calls, but also intents to convert such as driving directions or coupon prints compared to coupon redemption in-store to look at all the conversions that are truly happening from your PPC accounts.

2.11 Estimating Initial Conversion Rates

Now we get an info marketing, and often these are pages of hundreds of paragraphs of scroll and text, and so forth. 1% is a fairly common industry average for information-based marketers. If you have a low price point, you're offering a 9.99 product, you may see five, 7% conversion rates. The very high price product. Often you'll see .5% conversion rates. So, with info marketers, often get conversion rates, again change based upon price points.

2.12 Affiliates

It only happens like affiliates. Affiliates are companies who get paid referral fees based upon sales or leads made to a merchant. Now, often their conversion rates are similar to the merchant usually a touch less. In this case, we really want to use, 1% is a good starting number for affiliates. But if you're an affiliate for a large insurance company, and you're collecting the leads yourself and then re-selling the leads. So you're in control of the entire funnel, then you can use the lead gen conversion rates. These are more based upon affiliates who are doing pre-sales before e-commerce or before sending them to a completely different site and that other sites responsible for converting the traffic. So, then, by looking at the conversion rates, and estimated conversion rates you can get an idea for what you should be able to accomplish. Again, many companies do much higher than these numbers. We need something to start with. So, if you are a more aggressive marketer, use the high end of the ranges, or look at your current data, your analytics information. If you're a more conservative marketer, then use the lower end numbers. So it's very hard to truly estimate a conversion rate for every company. Before they actually advertise and get traffic. However, we can start with some overall averages to determine what we should be ballpark wise before we get into setting our initial bids. So now that we have an idea about how some of the formulas are run, and an idea about initial conversion rates, we can start getting into how we set initial bids based upon these calculations.

2.13 Ad Group & Keyword Bids

So first off, often you're going to set initial bids based upon return ads bidding or projected return ad bidding. Now you can do bidding at the ad group level or the keyword level. It's very common and not a big deal to do ad group level bidding when you first start. Because your ad groups are already well created, the keywords are very similar to each other. So they should have similar conversion rates and similar CPCs in most cases. And therefore, you're going to start with ad group level bidding. However, what's going to happen over time is as you get your own metrics, if you have a higher volume account, you'll do keyword bidding later on. So these same formulas are used, not just to set initial bids, but also maintain bids going forward. So we'll start to how we do initial bids, we're using estimated conversion rates. However, you can use the exact same ones with actual data, same formulas. To set your bids based upon real data, going forward once you have metrics inside of your account.

2.14 Calculating Initial Bids by ROAS

So what we usually want to do is set our initial bids based upon return on ad spent, so we have our ad group, our estimated conversion rate, again if you've been advertising, you have your actual conversion rate, our revenue per conversion, then what do we want to make a return on ad spend basis? Now often when starting out, our goal is really just to break even. Let's break even, because break even gives us, A, the most traffic possible, while getting actual data and actual metrics. So then we may change or over time 125 or even 200%, 300% once we get real data. What we need initially is information tilled to look at our true conversion rates, our true revenue numbers, our true cost per actions so starting with a break even number is often a good idea to get lots of traffic initially and then once we have traffic and real data we can impose that on the same formulas based upon our true desire to return ads spent. So to calculate your bid based upon ad spend, it's a very simple formula. It's your conversion rate, in this case estimated conversion rate, times your revenue per conversion, so what do you make when you got a conversion, divided by your desire to return on ads spend. So in this case we have a desired return on ad spent of 100%. So we take our estimated conversion rate of 10%, we multiply it by our revenue per conversion. So for the chocolate ad group it's $20. Then we take that number, and we divide it by 100%, our desired return ad spent. That gives us an initial bid of $2. We do the same thing for flower, same calculations. We end up with a $5 bid. Now, with this information when you get into, we're trying to ad spend of 2 and 300%. It's not easy to just glance at the numbers and see your final bid. So when we want to do easy looks into our metrics based upon bid, there is another metric we can use in our overall calculations to quickly see ratios between return on ad spend and our desired bids.

2.15 Revenue Per Click

So, revenue per click is just a simple calculation you can add to your spreadsheet, that lets you know how much money, on average, you make every time you receive a click. So revenue per click, it's a simple formula of total revenue for a keyword or ad group, wherever you're setting your bids, divided by number of clicks you received. So in our chocolate ad group we made $200 in revenue, it took us 100 clicks to make those $200 in revenue. So then, 200 divided by 100 is two, so every time we get a click from chocolate, we make $2 in revenue. Now, this makes it very simple then to look at your revenue per click, compared to your desire to return on ad spend to be able to visualize your bids for how most people can do math in their heads.

2.16 Calculating Initial Bids by Revenue Per Click

So for instance if you want to calculate your bid based upon revenue per click. It's just your revenue per click divided by your desired return on ads spent. So with return on ads spent 100% is break even. So therefore if our revenue per click is $2 our desired return on ad spent is break even so it's 100%, that means our initial bid is our revenue per click. It's $2, it's exact same number. This is where ratios become very easy then. So now, our revenue per click, again, still $2, but we changed our desired ROAS. We want to start doubling our money now. We've got some real metrics, we want to actually start building up our bank account, so we decide we really want to make $2 every time we spend one, which means our return ad spin is 200%. Now revenue per click of $2, divided by return ads of 200% means our initial bids a dollar. So now most people can do this math in their head of looking at, here's my current revenue per click, here's my desired return on ad spend and that ratio is essentially an easy one to see for your initial bid. So it's one more step in all your major metrics you're putting together initially, to ad revenue per click. But then it makes it easy to look at your ratios, revenue per click and ROAS, to see what your initial bids really should be. So if you want to set your initial bids based upon projected conversion rates and the ability to hopefully at least break even when you start, and hopefully you make money. That would be ideal as you underestimate. You make money day one advertising. It happens all the time, but so to be able to figure that out, you need to know what your estimated conversion rates are and your desired return on ads spent to set those bids. Now with this type of calculation often you're under bidding a little bit cause you're conservative on your overall conversion rates. However, you really lose money using a system like this. Other times you just want to start with traffic, you just want to get traffic, and then see the metrics afterwards. And then figure out what your bids need to be to they really make money and where you need to task or change landing pages around.

2.17 Adwords Keyword Tool

So another way of setting initial bids is based upon exposure and not just return ad spent. So if we go back to the AdWords keyword tool, we can see our keywords, we can see the search volume, we can see the approximate CPCs, right within the AdWords keyword tool. So now we can take this information, we can download it right to Excel file. Now Excel file, we're going to organize our keywords by Ad Groups. So we have Ad Group one which is Auto Glass Repair. It contains one keyword. We have another Ad Group, Auto Body Repair. Then we have another one, Auto Repair Shop. It contains keywords like "auto repair shops," "auto repair shop" singular, "auto repairs." Then we have another one auto repair service and then finally air conditioning auto repair. So we've got our various ad groups, we have our search volume, and we have our approximate CPC's for every single keyword. So then all we have to do is determine, based upon the search volume and CPC, what should our ad group level bid be? And this is based upon estimated CPCs from Google, which should give us some nice page one exposure from when we first start. So, if you have only one per ad group of course your CPC is the approximate CPC. If you have multiples then you can just take the local monthly search volume times the approximate CPC for all the keywords in an add group, and then divide that number by the aggregate monthly search volumes for the add group, and that'll give you a weighted CPC. And so now we have a simple weighted CPC, we can use this number for our initial add groups. If you want to, you could even use the approximate CPCs by keyword level when you first start, either way, and then this should give us some initial exposure, we may make money, we may lose money, we don't really know with this method. We're going for overall exposure with this method. So this should give us some nice exposure when we're starting to run our accounts to get data to make decisions.

2.18 AdWords Traffic Estimator

Now another way you can do the same thing of setting your initial bids by exposure is using the traffic estimator. The traffic estimator, you can put in your keywords and you'd only want to do this by ad group. So if you have five ad groups, you want to do this five different times. So pace in your keywords for ad group one. Find the top of the curve based upon your budget. So input your budget and then see where the curve is. The top of the curve. That's going to be your best CPC on a pure exposure base is, based upon your budget. So now if you have five ad groups and a $100 budget really, each ad group only has a $20 budget, assuming they have roughly same number of clicks and presence possible. So then you would really split out your daily budget when you use this tool at the ad group level of information. Find where the curve is for that ad group, then do it for each of your ad groups. And Google do a lot of the math for you but you'll, it takes a little bit longer because you've gotta put in each ad group individually then just doing weighted numbers in Excel.

2.19 Recap

So to recap our initial bids section. First you really want to think through, are you going to do your starting bids based upon a return on add spend. So having a projection of what you're actually going to make, or exposure, getting lots and lots of metrics, because that will determine what method you use to set your initial bids. So then, if you're doing it based upon return on ad spending, you need to estimate your profit per conversion. So, what do you actually make when you get a conversion. And what's your cost of hard goods. You need to find a good enough average conversion rate to start with. So again, you're not going to be perfect with your conversion rates when you start unless you have a lot of data from other traffic sources previously. Find something good enough that you're comfortable with for your initial formulas. Then determine your starting bids based upon these simple formulas. So we have a simple formula, we're trying to ad spend, a simple formula for estimating traffic by exposure. So use these formulas to determine what your starting bid should be. When you first start, feel free to bid at the ad group level. Doing keyword estimations for every keyword, knowing the information isn't even 100% correct is a lot of extra work. So it's okay to do ad group level bidding when you first start. But when you accumulate enough data bid at the keyword level. And this is a tough one, because if you have small accounts, aren't spending a lot of money, you might always bid at the ad group level because no keyword gets enough data, that's okay. If you have high traffic ad groups and every keyword has lots and lots of information, then you're going to bid at the keyword level. So there's not a right or wrong. Best practice is keyword level to bid. But if you don't have enough traffic, you can never make good decisions. It's okay to bid by individual ad group. So then with bidding, what you really want to determine is what is your return on ad spend. What are you comfortable with? How much do you want to make back? For every time you get a dollar, because the higher your return an ad spend, the lower your bid's going to be. Which means the lower the placement on a search page and the less clicks. So there's a balance to be had with return on ad spend being closer to 100%, getting more traffic, versus return on ad spend being very high and getting less traffic. There's a balance in the middle there. But when you look at here's what we want to make. Here's what we actually make right now, here's our conversion rates. Here's what we should do to some testing to get better. Then setting bids is not that complicated once you have the data by just following these simple formulas.

  • Disclaimer
  • PMP, PMI, PMBOK, CAPM, PgMP, PfMP, ACP, PBA, RMP, SP, and OPM3 are registered marks of the Project Management Institute, Inc.

Request more information

For individuals
For business
Phone Number*
Your Message (Optional)
We are looking into your query.
Our consultants will get in touch with you soon.

A Simplilearn representative will get back to you in one business day.

First Name*
Last Name*
Phone Number*
Job Title*