Technology
Services
Resources
This infographic provides a walkthrough of how to take your good results and drive towards better PPC performance using Big Data, Machine Learning, and Human Experience. The goal: extract the best possible results by using forward-thinking strategies and technology available today.
In the modern data environment of the internet, customer journeys are more intricate, making a “Conversion” in Google Ads worth less - and your own ability to track revenue so much more meaningful. Your data tells a better story, you just need to capture it, unify it, and process it.
Given technological advances in scalability and infrastructure, the application of machine learning is the no-brainer approach to process all that data. And show you what is really “converting” lower down the funnel.
But machines, automation, and computers can’t do it all alone - not without human involvement! A relationship is necessary. Learn more about the relationship between your data, your automation options, and your role in the process in our new eBook: The Path to Great PPC: Merging Big Data, Machine Learning, and Human Experience.
![Learn how to Optimize PPC with Big Data and Machine Learning [An Infographic]](https://centro.net/wp-content/uploads/2019/08/Path-to-Great-PPC.png)
To download this infographic, click here. Alternatively, if you’d like to request a demo with us and discuss all the ways we could drive improved SEM performance through big data and machine learning for you, get in touch here.
This infographic provides an overview of running through a root cause analysis on your SEM program, plus touches on two other types of audits you’ll want to keep in mind when assessing and solving PPC performance troubles.
While auditing poor performance isn’t anyone’s dream job, it’s simply a requirement for SEM managers–and can really make a big difference if caught early and addressed correctly. Regular audits are a healthy part of program maintenance.
Sometimes an issue is truly from a “root cause”: business changes, landing page issues, competition in the market, or culturally irrelevant keywords. However, often you’ll find more holistic drops that can be better explained by dimensions or segments that may just need to be adjusted with bid modifiers. Beyond that, sometimes your entire optimization strategy may be built on incomplete data or metrics, or just missing a key step in the customer journey that could drive value.
Addressing PPC issues isn’t fun, but this infographic can help you look into a few elements. However, you can learn even more about all of these topics in our new eBook: Advanced PPC Auditing Guide: Determining Root Causes, Bleeding Dimensions, and Optimization Strategy.
To download this infographic on solving PPC performance issues, click here. The full eBook is also available, digging into all aspects of the infographic. Alternatively, if you’d like to talk to our team about how to improve performance on some dimensions, or get more out of the metrics you’re optimizing towards–get in touch here.
It goes by many names, but you're paying money for it and it’s driving a lot of value to your business. Pay-per-click (PPC), search engine marketing (SEM), paid search, search advertising, Google Ads, etc… it’s the marketing channel that has the biggest opportunity to truly optimize, and do so based on hard statistics. Why? Because the big data it generates enables very accurate statistical analysis and computation.
Paid search optimization. It’s the practice of making the most effective use of advertising budget towards your SEM goals. More specifically, it’s identifying trends and opportunities based on past performance data in your search engine marketing campaigns, and applying those learnings to drive better results. Whether by eliminating waste and cutting costs where they aren’t leading to revenue or conversions, or (of course) investing more on the keywords, segments, and topics that have better results.
But you’re not here for that high-level overview, and neither are we. PPC bid optimization is a rigorous and technical field, and the goal of this article is to discuss the approach to automation and optimization of SEM campaigns with big data.
You and your business are likely using Google Ads (et al) in a very sophisticated way. Large SEM programs not only pay a lot of dollars to search engines and gather a lot of dollars from ad-conversions, but they are generating massive amounts of data as a byproduct.
For example, every search query that leads to a click and subsequent conversions has numerous elements along that entire journey that are being tracked inherently by Google… and then by Google Analytics, tag managers, “3rd party” pixels and analytics tools, and conversion tracking tools… and then deeper funnel data points about latent conversions, offline sales, and lifetime value that may live in a CRM or data warehouse. Plus you can collect browsing behavior and seemingly unrelated data on the sidelines that ultimately may correlate to some future value.
The point here is that when your organization uses PPC as a primary channel to generate revenue, you are (or should be) collecting a very, very large–and very powerful–dataset.
Alright, so what?
I'm glad you asked!
Big Data may be a buzzword that gets thrown around a lot lately, but let’s take a look at it in two manners:
1. The definition from Wikipedia: "Big data" is a field that treats ways to analyze, systematically extract information from, or otherwise deal with data sets that are too large or complex to be dealt with by traditional data-processing application software.
2. I like to simplify it into a simple concept: When nearly everything is tracked, you do get a lot more noise, but you also end up with a much more complete picture of associations and correlations. With big data, correlations trump causation.
Big Data removes the necessity of understanding cause and effect (the "Why?”) and turns the problem into a game of correlations. You see that when “A” happens, then “B” happens very frequently. To make that information actionable, we needn't make a claim about cause nor reason. There is a correlation. And that correlation has power.
Nowhere in marketing does this type of correlation have more easily accessible power and immediate actionable application than large paid search programs. SEM is the “rocket science” of marketing. Bold claim? Yes. Accurate? Probably.
Your data, much of which is simply a natural byproduct of running your program and having tracking set up in a healthy way, has enormous potential to optimize and automate results in your SEM campaigns. Of course when you add in the other tools, tracking and analytics providers, and your offline data (CRM, data warehouse, historical results, etc), then you've got yourself a gold mine of data.
But you need to activate it. You need to turn that potential and make it kinetic. (Thanks, Physics.)
How do you take that fuel, that tinder, that coal, and turn it into a humming machine driving optimal results and performance? It’s similar to how a car does it: you need an engine.
Manually parsing, sorting, and analyzing data to make decisions is out of the question. You don't have to be an Excel pro to realize that hundreds of thousands of clicks on your copious keywords and all the accompanying data from publisher attributes, conversions and revenue data, and analytics data won’t be easy to draw insights from in a spreadsheet.
Maybe you’ve been there too, with 20 tabs in Excel, trying to stitch, pivot, and analyze. Hitting “Save” is precarious by itself.. you might waste 15 minutes waiting! Besides the speed and infrastructure constraints of spreadsheets, statistical models need to be run on all that data, and preferably on a very frequent basis to make sure that the insights drawn from all that data can be applied to your program via bidding decisions.
The point is, you need something more robust if you’re going to use that data for a meaningful approach at optimizing your paid search campaigns.
At this point, it’s noteworthy to mention that an SEM Agency could be the next step. Give the agency access, and ask for results. Here’s the issue: unless that Agency is employing some additional technical solution, the optimization efforts are still manual and simply can’t hit optimal performance.
Big data, as noted by the Wikipedia definition, requires different methods and tools because you’re dealing with “data sets that are too large or complex to be dealt with by traditional data-processing”.
So, what are your options?
All jokes aside, hopefully you’re here reading to learn about how to truly optimize PPC programs with big data, automating this revenue-driving channel for your business. (So, let’s count out the cyborg option.)
Building a script is taking the responsibility into your hands to figure out that relationship between all your data and the results you desire. It’s a big undertaking, but can be very fulfilling.
Frankly, building any script for any job feels good, and makes you proud. In fact, if you’ve been involved in building anything (from a coding or programming or even process-driven perspective), let’s take a minute to realize you’re driving innovation in the small and immediate ways that help to make humans more efficient; the virtuous cycle of ever evolving and improving! (Pause and smile!)
Okay, back to action. Scripts are definitely a good first step if you’ve mastered the programming methodology, bidding methodology, and the syntax and logic. Most often you’ll have to hire the resources to build and maintain that solution.
The benefits? You have total control. Your engineering resource will know all the inner workings, the inputs and outputs, and logic for making bidding decisions based on the data. At a minimum, you’ll need someone with a very strong data analytics background, or better yet a data science background. Needless to say, it’s a lot of work. It’s the classic control versus effort challenge.
There’s another option we noted: purchase a bidding optimization tool.
PPC bidding software automates and optimizes your SEM program through a few steps, starting by integrating all of that big data: online and offline data sources. It models the relationship between the data on keyword performance, predictive data from publishers, audience and attribute segments, and revenue and conversion outcomes. It uses those machine learning-based models and projects the impacts of bids versus revenue. The big data comes into play heavily when estimating the keyword value from a Revenue-per-Click standpoint. Throughout the bid calculation process, PPC optimization tools will crunch your big data sets into accurate, actionable bid decisions.
Taking a step further than just the keyword-level bids, the front-runner PPC tools are able to use a slightly different set of your big data to calculate and automate bid adjustments for attributes like device, audience, or geography. The correlations seen between audience segments + the keywords they clicked and converted on in your data sets provide this.
I’ll paint the picture with a bit more color. Your revenue funnel can be sliced and diced into incredibly specific niches based on your big data. The performance outcome of each slice informs the bid that should be set. Layer in all the keyword data with your contextual data, and audience segments, time-of-day, day-of-week, geography, device, seasonality, etc etc etc: the venn diagram overlap of each of those various attributes will each have a different “worth” to your business. Processing that data with a data science based approach in an SEM optimization tool will figure out which of those segments have value (or not) and adjust bids based on the expected return of each.
Further, tools like these can provide other functionality that is built off of your data–such as forecasting ROAS or ROI. The more data there is to ingest, the better.
There are some tools and vendors designed for smaller programs without as much data, while some designed for bigger programs with the types of data and scale we’re discussing here. These must have the infrastructural underpinnings that enable clean integration, fast processing, and actionable outputs.
Ad management platforms that “sit on top of” Google Ads (Adwords), Microsoft Advertising (Bing Ads), and Yahoo Ad Manager have a few flavors. Some focus on campaign management and easing up workflow. Some help to cater to multiple channels beyond just paid search. And others are in-line with our current discussion: focused on big data ingestion and activation for peak performance. All of them should do a variation of what we’re talking about here: using performance data to make bidding decisions. However, there are multiple depths to traverse.
Let’s introduce another word: predictive. Predictive bidding optimization tools approach the problem from a slightly different angle. While fundamentally the same, there are nuances that stand out:
With big data and optimizing PPC performance, the underlying problem is one of scale. The amount of data we’re talking about isn’t easy to process. There are two solutions that the best SEM bidding optimization software will employ: infrastructure and lower space machine learning.
From an infrastructure perspective, distributed cloud servers with in-memory processing and anomaly detection help to manage the load and manage the scale with speed. Reporting on millions of rows can take a couple of minutes instead of the better part of an hour. These infrastructural elements also prevent problematic data from interfering with bids through bias correction and anomaly detection.
Machine learning strikes again to manage the scale. Specific algorithms can allow the interactions between millions of interactions to be simplified in a different “lower space” to remove the load issue, and reduce the problem into one that’s more practical. This class of algorithms is similar to those used by companies like Netflix, who is also dealing a scale problem: millions of users interacting with thousands of shows and movies.
Big data is difficult to deal with and make work in your favor without the right tools. However, with the right methodology, technology, and incentives (ROI!), all that data your collecting about your customer journey is the perfect weapon for optimizing PPC performance.
Bid optimization is very serious and technical business. It gets deeper and deeper the more you look at it. It’s not easy to manage with only our human brains (until we become cyborgs), but the modern suite of automated bidding software for paid search is doing a great job at extracting peak performance from programs with lots of data.
When you’re losing potential revenue from campaigns that aren’t performing at expectations, it’s painful. Some of us work just to live our lives on nights and weekends. However, others are looking to really get ambitious about how to add more value at work, and turn our profession and career into just that: something we are professionals at!
This infographic breaks down portfolio bidding as a PPC optimization strategy. This is the modern approach to SEM optimization that looks at shared data across a group to make decisions that benefit the whole portfolio, rather than any individual keyword within.
Google Ads documentation provides a high-level view, but may not answer the questions you have about the value of this type of bidding methodology.
Take your paid search campaigns to the next level with this optimization strategy. Learn more in our other blog posts on portfolio vs keyword bidding, or in our new eBook: SEM Optimization Techniques: Are You Overpaying Google?
To download this infographic, click here. Alternatively, if you'd like to request a demo with us and discuss all the ways we could drive improved SEM performance for you, get in touch here.
Let’s cut right to the chase: there are various PPC bidding optimization techniques and practices floating around, but portfolio bidding is probably the right one to use, provided you have a large program.
I sold it to you that quickly? Terrific.
But just in case, we’ll go on...
Portfolios are groups of assets, and in search engine marketing, these assets are the keywords and “publisher objects” that you can apply bids and bid strategies to. The power of a portfolio strategy is twofold.
One, it allows similar keywords to share goals and data, and be pointed in the same general direction. Two, it executes in a way that optimizes toward (and hopefully achieves) your goal in aggregate across the group. Results are more controllable because decisions are made based on shared information within the portfolio.
But what really is portfolio bidding? And how do you start using it? We’re about to find out.
If you ask Google Support, you may read a definition for portfolio bid strategy that sounds like this: “An automated, goal-driven bid strategy that groups together multiple campaigns, ad groups, and keywords. Portfolio bid strategies automatically set bids to help you reach your performance goals.” Thanks Google!
Okay, so it’s this thing that automatically sets bids on groups of publisher objects, based on your goals…
But what is it really? How does it work? And why? Is this thing Google is talking about the same thing you’d expect for the core methodology for PPC bid optimization?
Let’s dig in a bit.
Modern Portfolio Theory was an economics concept originally introduced in 1952 and describes the technique for limiting risk and optimizing outcomes by sacrificing individual components to benefit the larger whole. It was first discussed in terms of stock market investment strategy (diversifying investments so some will win and some will lose, but the whole investment portfolio comes out on top). Presently, it helps to describe this SEM bidding optimization technique.
The solution involves bidding on a group (or portfolio) of keywords towards a target goal, while also maintaining an efficiency metric. With this methodology, the goals of the group outweigh any specific keyword-level goal: some keywords will perform worse to maintain the efficiency metric of the group, while other keywords will be bid up to drive the target goal. As a whole, the group is optimized based on the context of the entire portfolio.
The relationship between spend and return is non-linear, and that is true across keywords. Portfolio methodology models this relationship granularly at the keyword level, but adjusts for the findings collectively–across the portfolio.
So by way of example, let’s say you want your keyword group to maintain 220% ROAS while maximizing revenue. The portfolio bidding algorithms would use data from clicks and conversions to bid up the more valuable keywords that drive revenue while simultaneously bidding down other keywords to stay within the ROAS limit. The shared goals and data allows the bidding algorithms to execute towards a multi-faceted goal accurately by sharing the context of all the keywords in the group with each other.
To put it simply, one does not simply “do” portfolio bidding without utilizing an optimization tool.
As a very simple “two-keyword example”, an exercise can be run to look at the volume/efficiency curve and select the values (based on some data exports) where you’ll bid so that a target goal for efficiency is maintained while otherwise maximizing for the revenue metric… however, it’s simply too much to do manually for any real program.
Obviously, Google allows portfolio bid strategies to be applied to groups of keywords you define by using some of the goals available in Google Ads: Target CPA, Target ROAS, and Enhanced CPC, plus some others. How do you apply this to your programs? If you’re using Smart Bidding, you can effectively just follow the on-screen instructions.
But Google’s method is limited in the same way any other bidding method is limited with Google: it’s subject to the same requirements for volume and conversions; it doesn’t utilize any customer journey data outside of what Google tracks; and it’s built as a one-size-fits-all solution, to help small businesses, mid-sized spenders, and big programs all try to accomplish their individual business goals.
Any third-party PPC bid optimization tool (well, any modern one worth using) will utilize portfolio bidding in one way or another. Remember, it is a conceptual approach to solving a problem. However, there is certainly a great deal of variability between different solutions: between the best practices and most robust models, versus standard approaches.
To get started, you’ll want to ensure your conversion data is being tracked and is cleanly integrated into your optimization tool. Data integration is key to making smart decisions.
Next, you’ll determine which keywords should be grouped together in the same portfolios, or ideally utilize a tool that helps to automate this. Whether it’s a handful of keywords, ad groups, or campaigns, you select the best group of objects that share sufficiently similar attributes and goals to put under one portfolio.
Select which goals and strategy you want applied to each group and define the target metrics for each: goals for maximization, and goals for maintaining efficiency across the portfolio. However, be aware that setting goals too extreme up front can cause issues with volume or spend, and lead to disappointment.
Here’s what we see: even with the technology available to automate and optimize with something as complex as portfolio bidding, it still requires human decision-making and touchpoints to thoughtfully structure the program, define goals, and strategize.
Portfolio bidding is a technique that is here to stay. Why? First, because it is conceptual - not some specific tool. Second, it is a concept that aims at what we digital marketers are after: ensuring the aggregate performance across a group of keywords is hitting the target.
However, we must not forget that this technique, like many facets of life, fits in with that old adage from Drake’s song Preach: “Doing is one thing. Doing it right is a whole different story.”
Many PPC automation tools will use a portfolio bidding mindset to optimize SEM performance, but more is involved for great outcomes and peak performance: machine learning involved at every step; infrastructure built for accuracy and speed; data ingestion of customer journey data and other relevant context; and so on.
In paid search programs, the levers you can pull to improve results come in many forms: landing page optimization, ad copy testing, program structure efficiency, automated workflow, and reporting functionality. The most powerful tool, though, is unquestionably keyword bids.
SEM bidding optimization is the most significant way to achieve peak performance in paid search. A number of techniques and tools are available to help advertisers take on this challenge, but there are strengths and weaknesses among these. Over time, innovative transformations have sprouted up, but there is an apparent divergence between what benefits smaller versus larger search programs.
In this article (and more thoroughly in our new eBook) we explore some of the transitions, strengths, and weaknesses of different PPC bid optimization techniques - all to help performance marketers like you find success in improving paid search results.
![SEM Optimization Techniques: Are You Overpaying Google? [eBook]](https://centro.net/wp-content/uploads/2019/07/Blog-BiddingGuide-CTA-03.jpg)
Today’s landscape is varied, and many of the tools are provided by our beloved search engines - predominantly Google and Bing.
Manual bidding is a powerful technique that gives the advertiser control, but at the expense of effort. Like some of the other native tools, it can work well for smaller, local programs, yet any sizable program (in terms of budget or keywords) will quickly hit a problem of scale where true optimization, frankly, can no longer happen.
Google’s Smart Bidding options are the clear stand out for cheaply and easily applying real optimization tools to your PPC bids, but even these have limitations. Restrictions arise when it comes to the amount of your business data that can be used for optimization (outside of the Google tracking ecosystem), or when it comes to long-tail keywords without enough traffic or conversions to get any optimization.
Scripts on Google and Bing, or other “out of the box” tools, provide some functionality to optimize, but many of these are made for problems of simple logic, programs without many keywords, local businesses, or businesses that collect meaningful lower funnel data. Third-party tools are outperforming native tools in many instances due to differences like these between simple, small programs, and large, complex build-outs.
Methodologies have also transitioned over time. The old folder model was a great way to optimize PPC bids in the past, but it has since become outdated. A folder would optimize for a given goal in a simple and direct way: higher-performing keywords would be bid up, and lower-performing keywords in the folder would be bid down. This, unfortunately, resulted in some keywords becoming lost after being bid down so regularly - not the ideal outcome.
Aggregated bidding leads to inefficient bidding as it does not reveal or eliminate hidden waste at more particulate levels. In general, aggregating simplifies the task at hand (which is great in some ways!). However, when we’re talking about SEM programs that are core revenue producers in your business, this mindset promotes just-above-average results.
With the current data environment we live in, PPC marketing creates massive troves of click-level data that follows the customer through their journey to multiple purchases, sometimes offline sales. This data is the key and the fuel to unlock and boost SEM.
Granularity is of massive importance in the modern age of bidding optimization. Click-level information leads to click-level correlations, which result in the ability to optimize at very granular and specific levels for different audiences and keyword queries.
The tried-and-true SEM optimization technique known as portfolio bidding is actually still one of the best possible ways to make PPC bid decisions for programs that have revenue or conversion goals, but still need to stay within a cost efficiency limit. Most bid optimization tools utilize this technique, but there are some key differences that will separate the good performers from the greats. The three key aspects are:
SEM teams looking to optimize PPC bidding face a number of challenges and blockers. There are some obvious ones, such as insufficient budget or already averaging position one, but we discuss some of the blockers that are hidden at first glance, including:
Our new eBook, SEM Optimization Techniques: Are You Overpaying Google? looks at the landscape of bidding optimization techniques (from tools to methodologies), some blockers teams may face, and the variability between average versus great third party PPC optimization tools.
Why is portfolio bidding for PPC the best methodology for many large advertisers? Why - for others - is there an even better way than portfolio bidding? Is infrastructure that important for fully optimized PPC bidding? How does more granularity and more data create better results? Read the eBook to learn these answers and much more.
![SEM Optimization Techniques: Are You Overpaying Google? [eBook]](https://centro.net/wp-content/uploads/2019/04/Blog-BiddingGuide-CTA.png)
Manual PPC bidding versus automated PPC bidding. It’s not really an age-old question, but there are certainly plenty of people on both sides of the discussion who have passionate views. But… let’s call it an important dialogue, one that can have a meaningful impact on business results. (It certainly shouldn’t damage any relationships!)
While tactically there are breakdowns of each automated bid strategy available for advertisers - from Google and plenty of others - with pros and cons, we aim here to join the conversation from a different angle. Let’s take a look at the needs and the philosophical goals of search engine marketing and PPC bidding.
With Manual bidding, what are the core byproducts we should extract for discussion?
The first is control. One of the most powerful ideas behind manual bidding is that control falls into the hands of the advertiser, and not into the hands of a search engine ad-publisher or a third party tool. The advertiser has control and visibility into the decision making, the data, the performance, and the outcomes.
Why is control important? There are two primary reasons:
The second byproduct of manual bidding: effort. The other primary idea inextricably coupled with manual bidding is that effort is required, and often a lot of it. Time investment is needed. Continuous analysis, review, and mental energy must be applied on the same set of tasks. By definition, manual bidding takes time and effort!
Why is the effort discussion-worthy? We’ll dig in:
What are we really after here? We want to merge the best that “control” can offer with the best that “effort” can offer.
Obviously our underlying goal is to set and accomplish business outcomes: to have program results be acceptable, or better yet, excellent. We believe control is powerful. We should have the right amount of control. However, we are also limited in the amount of effort we can meaningfully exert.
How do you accomplish this balance?
Well, it starts with reflection (or analysis - whichever word you align with more). Spending a bit of time evaluating your desired level of control (for you personally, and for the programs you manage) and your current versus desired level of effort and time investment.
If you’re reading this article, you’re probably the type of person who is interested in learning and improving the way you operate, and thus likely to take a meaningful shot at reflecting on your present methods (which is a great thing!).
This reflection and analysis should help you to identify options for automating sections of your program: First, where you can take advantage of the elements of control (tactically or strategically). And second, where you can invest effort into places it best serves you, your goals, and your business goals.
No article or blog post can tell you that exact mix of manual versus automated for your specific program, but it can act as a catalyst to begin that reflection process!
Where does automated bidding fit into this type of conversation?
Control has psychological elements, but ultimately the aim is to understand and control performance. You feel and know that you are getting the best performance possible because you are seeing the inputs, “turning the screws” yourself, and seeing the outcomes. Performance and results are ultimately the goal of having all this control.
The reality in many, many cases is that performance will be better when automated processes and calculations take care of the bidding. There are times when automated use cases may not apply as much (when thinking about impression share, or page position, or match type benefits, etc), but where there is data on conversions or revenue, the right strategic goal, and enough volume to merit it, automated bidding will optimize PPC campaigns better than manual human efforts ever could.
The second point, effort, is completely flipped on its head when using automated bidding. The day-in and day-out effort and time investment to manually analyzing performance and making bid adjustments is removed. Yes, you’ll still be reviewing your performance often, but in the world of at-scale paid search, that’s a given.
You create time savings by removing the pain of manually managing countless ad groups and keywords and products, and can use that time on more strategic activities to grow your business. Time and effort will be required, and required in abundance. However, those hours will be worth more. Assuming one doesn’t thoughtlessly apply a single automated bidding strategy to an entire paid search account, the campaigns and portfolios that have specific bid strategies will be operating at a much higher clip, creating time for people to focus on the other parts of the program that need the human touch.
Automation in SEM goes far beyond just doing a process automatically, it gets deep and complex quickly with optimization techniques. This is where artificial intelligence, machine learning, and other engineering and statistics innovations are attacking the process of improving performance in paid search. Automating some campaign management and workflow tasks are helpful to save time - but from a dollars-in, dollars-out perspective, big data applied to bid management has the most potential.
There are obvious places where a human’s manual thoughtfulness is better suited than any sort of machine or automation. Some of those are:
In the past, other items would have been on this list (like where device and audience bid adjustments should be made), but those specifically are now being automated fully by automated providers (from data ingestion, modeling, calculation, and execution). “Machines are taking our jobs” is real; but in the best of ways. You (the human, I presume) get the job of thinking, planning, and deciding, then the robot gathers the data, performs the calculations, and executes hundreds of thousands of actions for you.
Machines (a fun way to name any automated software or system) are obviously better at the niche, nitty-gritty analysis, calculation, operating at scale, and executing at ridiculous speeds. Bidding is one of the best places for automation to take over; it’s a problem that’s based in statistics and numbers that correlate directly from higher funnel (position, impressions, and clicks) to lower funnel (conversions, LTV). The machine’s platform will have a U.I. to make big or small tweaks and test the performance, providing control to the PPC bidding optimization effort.
Then, there’s the data...
Data is the other element that takes the benefits of automation and exponentially amplifies it. With an increase (almost every week, it feels) in the amount of data we can capture about the customer journey, there is so much more to feed the machine! That statistics problem can be worked with a much more robust data set, over a longer and more nuanced customer journey. The requirements from a human to piece that data together to make meaningful decisions for optimizing performance at scale just isn’t feasible.
Automation applied to huge data sets is a beautiful thing.
But how do we maintain control; the type of control described early on in this article? There are three points to make clear at this stage:
Clearly the stance offered here is that automation can add significant value. Nothing works without humans captaining the vessel, but we do believe that the cooperation of humans and machines is the best way to achieve (and exceed) goals in paid search. Machine learning PPC algorithms and automated bid management have tremendous potential, but shouldn’t be unleashed without preparation and thoughtfulness.
Now is as good a time as any to reflect on where it makes sense to invest your time, effort, and hours. In smaller programs where the overhead isn’t enormous and there simply aren’t many huge strategic decisions to make, the decision to proceed manually is likely a great option.
Alternatively, working with a program with over five hundred thousand keywords and a monthly budget around that $500,000 range as well may have much more to gain from automating bidding, and putting that manual, human effort in other areas that allow the machine to perform better at automating, and ultimately allow the business to prosper.
Try to maintain the optimal control and get the best return on effort, but don’t fall into a false sense that full control by a person is better than joint control with an automated solution.
How do you determine a search keyword’s value? Are you making data-driven bid optimization decisions based on keyword data? Should you be focused on keyword revenue-per-click in your PPC campaigns?
Search keywords. They’re valuable. Some are really, really valuable! Others have little to no value. Others you simply don’t know the value of for one reason or another.
What we do know is this: keywords are the lifeblood of paid search. Building a keyword list and then collecting data about keywords is paramount in search engine marketing, especially if you’re looking to optimize PPC campaigns.
If a keyword is generating a lot of traffic and clicks, it’s seen as important - but it may or may not be valuable to your business. You need to understand if it actually drives revenue and conversions.
If a keyword is a “monetizer” - or a high-revenue driving keyword - you’ll want to evaluate how much impact the keyword has on your business. You would look at it and say: “If we pause this keyword, how much will revenue decrease? By how much? If we decrease the bid, will we lose a lot on conversions - or if we increase bids, will we still only generate the same number of conversions?” There are many questions to ask to determine how valuable a keyword can be to your paid search program.
Evaluating a keyword’s value can be a challenging task. Some methods are more simple; some get deeper but may not be actionable; some are totally automated and drive action automatically! What we’re really after is the revenue-per-click metric that truly informs a bidding strategy.
Your keywords drive impressions, they drive ad-clicks, they drive conversions… and hopefully repeat purchases! They’re obviously worth quite a bit, and qualitatively and quantitatively are one of the most reviewed areas within SEM ad campaigns.
The dollar value of a keyword is, in short, the amount of money you can expect to generate from each click from a given keyword. A keyword may have only generated one click, or it may have generated hundreds of thousands of clicks; zero conversions, or dozens per day!
Conversions lead to dollar-driving outcomes. Dollars are our interest here. How much revenue can we earn from each click on each keyword?
Some of those dollar-values are actual and calculable (online or offline purchases for goods; lead form fills that lead to purchases), while others may need to be understood in aggregate based on metrics within your company (subscriptions or lead forms that don’t have a purchase, but create other dollar value).
Google Ads makes it easy to report on keyword performance and rank by impressions, clicks, costs, conversions, and even the aggregate revenue for certain conversions that can be tracked and directly tied back to the Google Ads platform. However, estimating the value of a keyword for a more specific use case (such as using that value for PPC bid optimization) requires a different view of keyword value, which we’ll dig into shortly.
To analyze a more nuanced value of each keyword on your business, you may need to do some serious spreadsheet wrangling. Why? To understand the impact of historical bid changes on conversions over time! (Those time-dependent reports are always hard, aren’t they?)
A method we’ve witnessed involves exporting large change history reports and large historical keyword performance reports, and then stitching them in Excel to look at the trend analysis. To do this, you would map the dates of bid changes on specific keywords to the keywords historical performance on given days on or after that bid change. Then you can get a historical understanding of the impact of that keyword’s bid on revenue, and see how changes impact your business.
Now, is that actionable? Depending on your team’s level of sophistication, it may or may not be - but with sufficient analysis you should have an understanding of how increases or decreases in bids create more efficient CPA or ROAS numbers on your keywords.
The keyword’s value can be understood in many ways, but ultimately you want a keyword valuation model that gives you something actionable to work with. Even more ideally: something that can drive action without you needing to apply all those changes.
In a large program with many keywords, the data export we’ve described may be fairly arduous. Without a program or tool to calculate these keyword values for you automatically, you would need an internal process for calculating the revenue per click on each keyword, which is a significant investment.
Keyword value estimation requires data. Having your conversion data (with revenue numbers) tied back to clicks and keywords is a prerequisite to being able to meaningfully calculate an “estimated keyword value”. However, with more and more data tied back to those keywords, on multiple dimensions, and over time, the true value estimation gets more and more accurate!
When you map all the data about costs and clicks and conversions back to each keyword, you can build a graph of revenue values–what we call “revenue-per-click” (RPC). The RPC graph provides a visual representation of the clicks vs revenue for that keyword. Automated bidding automation tools calculate RPC graphs for every keyword regularly to review data and discern if the value has changed enough to merit a bid change. The RPC calculation at scale enables a mechanism for taking that estimated keyword value and immediately putting it to work in a bidding calculation process.
Revenue-per-click is calculated in a meaningful way only because of large data sets and machine learning models. Without a machine learning powered process, these calculations - and the actions they drive - would lack serious optimization potential.
What is significant about a very granular and specific keyword value? So glad you asked!
When all the conversion data (whether immediate click-to-sale, or LTV from latent and repeat purchases) is used to model a highly accurate keyword value, then acting in a “data-driven” way naturally follows. The data provides a story of how much this particular keyword is worth across other variables, so that the information can be plugged into cost, volume, and bid landscape data to model the optimal bid calculation for any conceivable segment.
For instance, when person A from North Carolina searches for your exact match keyword in the morning on a Tuesday at work using a desktop computer and makes an immediate purchase… the bidding decision made for other similar search queries are valued in a similar light. On the other hand, person B from Los Angeles searching for your keyword in the late evening on a Saturday using a mobile device, and then purchases the next week, has a completely different value. The similar attributes should be accounted for in bidding decisions based on the dollar value calculated from those correlations in the data points.
Estimating the value of a keyword is an important and fundamental step in the process of calculating optimal PPC bids on keywords.
At the start of the process for any bidding optimization solution, the dollar value of each keyword must be determined. There are two notable pieces at play here: the “machine” that builds the model, and the data you feed that machine.
In the highest caliber optimization tools, the calculations for this revenue-per-click model are generated using various types of machine learning algorithms. Clearly processing at this magnitude is an issue with any standard approach (a multitude of keywords x a multitude of data points), so the infrastructure and scalability of the tool is important.
The data fed-in is likely even more of a secret weapon than the algorithms and modeling. The more sources the better, particularly as they relate to stages of the customer journey and revenue. (Imagine not only online purchases and lead forms, but also delayed transactions and CRM information being utilized).
PPC bidding optimization has come a long way over the years. The bid calculation process for at-scale programs has many moving parts and can be intimidating for data scientists, let alone the day-to-day users, managers, and stakeholders in paid search programs.
Bidding calculation in the modern era has many flavors, but we’re interested in the best and most optimized version; the type of process that is designed to drive peak performance. Our new Guide, Machine Learning Powered PPC Optimization, walks through both the prerequisites of a thorough bidding process and the bid calculation stages that modern SEM optimization tools are utilizing to unlock the most from large PPC programs. Its focus is on machine learning PPC.
The infrastructural foundation to the best possible bid calculation starts with the data architecture to capture brand interactions as a series of events happening in real-time. The system must be flexible enough to ingest all of the data sources that track, measure, or influence the customer journey.
Once that infrastructure is in place, that data can be utilized for execution; for action; for the modeling and calculation that turns a data set into dollars.
Our new guide to modern bidding, using QuanticMind’s calculation stages as a model and exemplar, explains the concepts and process of fully optimizing PPC bid management. The optimal bidding platform leverages the latest advances in Data Science, including machine learning algorithms, Bayesian modeling, predictive performance methodology, and natural language processing, to optimize SEM performance toward specific business goals.
There are six high-level steps.
One: Understand and Estimate Keyword Value
Stage one looks at the modeling that estimates the dollar value of each individual keyword. This process involves ingesting revenue data from any conceivable source and applying that back to the keywords in such a way as to estimate a dollar-value for each. This involves powerful machine learning models generating revenue-per-click graphs, and results in a dollar value that can be used in later steps to calculate optimal PPC bids.
When keywords lack sufficient data to make a meaningful model of the potential value, deep learning text recognition models are used to map semantically similar data-rich keywords to data-poor keywords. As a result, even low-click or low-conversion keywords still get the most accurate possible value assigned.
Two: Understand Click and Cost Elasticity
Stage two aims to understand click and cost responses to CPC changes, with the goal of generating a map of costs and the expected volume. This is where Bid Landscape Data from Google is highly useful, and applied at scale to an optimization practice.
Three: Calculate a CPC that Promotes Your Goals
Stage three combines the estimated dollar value determined from stage one’s artificial intelligence-powered PPC calculations and the cost ecosystem analyzed in stage two. The decision engine applies the advertisers targets, bidding strategies, and goals and then runs through and selects the best bids to maximize performance, given the data, calculations, and goals. Often, a Portfolio approach is used to bid against a target while maintaining an efficiency metric. This is the modern approach to PPC bid optimization that most bid management tools utilize - if they’re designed for medium to large SEM programs. QuanticMind differs in some ways from legacy tools, discussed further in the Guide.
Four: Calculate Bid Adjustments
Stage four repeats nearly the same process completed in the first three steps, but on a different set of data and with a different purpose: calculating and automatically applying bid adjustments. QuanticMind’s model shines at this point, using machine learning to optimize bid adjustments at scale. Device Bid Modifiers, Geo Location Bid Modifiers, and Audience Bid Modifiers can all be automatically calculated and applied, based on their relative successes in the SEM program. The data science algorithms used here are another advantage when attempting to calculate optimized bids at scale.
Five: Anomaly Detection
Stage five moves into the often understated - but highly important - anomaly detection. This is one of several areas where the infrastructure discussed at the top can “flex” its strength. When designed for effective capturing, cleaning, and piping of data from any source, the system provides better data for better execution. However, the opposite has negative effects: when data is missing or seems different than forecasts would suggest is reasonable, the performance can take a hit. Fully optimized bidding platforms prevent these problems by using multiple anomaly detection and issue-prevention steps, ensuring bids aren’t pushed based on bad data.
Six: Bid Push
Stage six is the execution! Push the bids and bid modifiers through the publisher and go live. Data collection is ongoing and fed back into the system. Other uses for more variable aspects of a program, like inventory management or a “maximum capacity of leads” per day or location, can be applied and fed to make decisions even quicker. Ultimately the process is repeated to create a virtuous cycle of optimized PPC bidding.
This guide will walk you through the modern solution to optimized bidding automation. It is a powerful tool to learn the leading process in paid search bid calculation and optimization. It helps paint the picture for how machine learning PPC is actually applied, how well-integrated data feeds the model, and how bid management decision engines step through the process of calculating and pushing bids. It helps answer the question: how can I optimize PPC bids?