White Label SEO
seo in dubai Mark Meyerson – Search Engine Land News On Search Engines, Search Engine Optimization (SEO) & Search Engine Marketing (SEM) Fri, 03 May 2019 12:21:22 +0000 en-US hourly 1 Static reports are dead: Here’s why you need to move to Google Data Studio /static-reports-are-dead-heres-why-you-need-to-move-to-google-data-studio-316389 Fri, 03 May 2019 12:21:22 +0000 /?p=316389 With real-time reporting, interactivity and shareability, dynamic dashboards are a time-saver for marketers and keep clients informed.

The post Static reports are dead: Here’s why you need to move to Google Data Studio appeared first on Search Engine Land.

So many agencies and marketers are still using old school static reporting. Why? With enhanced shareability and real-time analytics, the time has come to move to dynamic dashboards.

There are several dashboard options on the market, but for this article, I’m going to focus on Google Data Studio (GDS). Other options are usually expensive, difficult to customize and less flexible in their data connections. GDS is free, tends to perform well overall and the reason I benchmark with it.

Key advantages for dynamic dashboards

Real-time reporting: Unlike static monthly reports generated at the end of each month, GDS provides real-time reporting. Some data sources are refreshed every 15 minutes, others are longer, but the lag is not more than 12 hours for most sources. Report dates can be adjusted on the fly, allowing you to view the previous month or any time period you like.

Interactivity: Aside from dynamic timeframes, you can also setup the report to filter by dimensions. With the click of a button, you can adjust all graphs and tables to only show traffic filtered by devices or traffic sources. The viewer can interact with the dashboard and drill down into more specifics. In this clickable example, Google shows how interactive charts work. In the below dashboard screenshot you can see a filter for “age group.”

Templates: Possibly the most powerful feature of GDS is the ability to connect your data sources with existing templates. As an agency, this means you can use one template and duplicate it for multiple clients. While static tools often come with templates, since GDS is free and open to the market, there is a wide third-party user base creating and sharing their templates. Google Data Studio features templates from the community that anyone can use. And full disclosure, I have also developed templates specific to digital marketers.

Pricing: Being a Google product, GDS is free to use while many of the static reporting tools are costly. For any Google product integration with GDS, you don’t need a third party. But if you want to integrate other non-Google data sources, it will often come at a cost.

Third-party connectors: While third-party connectors cost money, they do provide added usability, allowing you to connect GDS to a wide variety of platforms. GDS has over 100 connectors available at the time of writing.

Shareability: A static report is usually emailed as a PDF or excel file. Beyond generating a printable file, GDS can be shared via a link and with password protection if you want it. It can also be embedded on a website. Here’s an example of an embedded dashboard on my agency site.

Transparency for clients: When a client can check in whenever he or she wants, there is more transparency between the agency and the client. It also gives the client a greater sense of control that they know where and how their budgets are spent.

Time savings for marketers: The ability to template your work and share between multiple clients is very efficient. Using an already-made template (there are lots of free and low-cost templates available) saves you from developing your own new reports. The sharing options also save time on sending, tracking and following up emails every month.


Some agencies run static reporting in Excel which might have certain flexibilities in data calculations. But these can usually be performed in GDS. The main issue is its limitations compared to other Business Intelligence tools. For instance, data preparation and blended data need improvement in GDS. And it needs a better content management system to make content handling smoother. However, that’s a topic for another article. For digital marketing reporting purposes, and especially those that want to get up and running quickly and easily, GDS is still the best tool in the market.

Getting started

When choosing a template, these are the issues I’ve identified that need careful consideration:

  • Data connection and integration
  • Presenting data with the relevant visualization (charts and tables)
  • A background theme to tell your story and provide the right flow

There may already be templates created that match your needs. But even if not, adjustments to existing templates are usually easy and can be launched quickly. Build on the work that has already been done by others to get started with dynamic templates in GDS.

The post Static reports are dead: Here’s why you need to move to Google Data Studio appeared first on Search Engine Land.

A case study for delivering performance in a mature Google Ads account /a-case-study-for-delivering-performance-in-a-mature-google-ads-account-315176 Wed, 10 Apr 2019 12:00:59 +0000 /?p=315176 Over the course of a year, 80 experiments tested a wide array of features (like responsive ads, target CPA bidding and others) for a large legal client.

The post A case study for delivering performance in a mature Google Ads account appeared first on Search Engine Land.

Running a Google Ads campaign can be a case of you’re damned if you do and your damned if you don’t. If the campaign turns out to be a flop your obviously in trouble, but if it turns out successful, you can run into other challenges. Campaigns that are successful over a long period can present challenges in continuing to deliver performance and add value.

In my experience, new Google Ads accounts have quick wins and low hanging fruit. Over time, these become less apparent and there is more need to innovate. One of the ways that we’ve been able to deliver performance over time on already successful accounts is through Campaign Drafts and Experiments.

In this case study, we were running a campaign for a large legal client over five years. Results were phenomenal over the period and we’d seen extraordinary growth. The Google Ads campaign was in a mature state where we were happy with performance and CPA levels, but we were challenged to continue to deliver lead growth. In this competitive industry, it was important to constantly test features and push new boundaries. For a year we ran 80 experiments to test a wide array of features. We’ll take you through some of these tests, the results we received and what we learned.

Campaign drafts and experiments

Before we do, a quick summary of campaign drafts and experiments is in order. We’ll call these “experiments” for short. This feature in Google Ads has helped us crack the issue and continue to deliver performance in a mature account. The basic process for using the tool is:

  • Clone an existing campaign as a new draft
  • Make desired changes within that draft to test some hypotheses
  • Run this draft alongside the original campaign for some time
  • Split the traffic between draft and campaign (usually 50/50) as an A/B test
  • Report on results in real time throughout the test and provide updates when results are statistically significant
  • Apply draft results to original campaign or reject draft campaign with a click of a button

Google has a detailed guide for setting this up which is the best resource to use as a guide

The tool has given us the freedom to rethink how we run an account and involve our clients. We can now sit down with a client and come up with a set of hypotheses that we want to test. These hypotheses are designed to align with upcoming client goals and also push performance limits. Clients are involved in a decision-making process, which is completely transparent. They were able to see the process from the formulation of the question/hypotheses through to results.

Experiments also provide a safe environment for implementing and testing new features. When account performance has been strong, we are often hesitant to rock the boat. But we still need to try out new features. Take for example the recent introduction of machine learning features and tools in Google Ads such as automated bidding strategies and responsive ads. Handing over the keys to ML algorithms can be daunting. While ML might provide incremental performance improvements, there is risk these algorithms may not perform, and then account performance will suffer. Experiments allow you to minimize these risks through testing.

What we tested

In consultation with our client, we ran an array of experiments. These were tested on a rolling basis throughout the year. As a sample, some of the key hypotheses we tested were:

  • Automated bidding (maximize conversions) will provide more conversions than manual bidding.
  • Automated bidding (Target CPA) will provide a better conversion volume performance than we currently achieve with manual bidding.
  • A more granular campaign structure based on SKAG’s will increase the quality score for the campaign
  • Responsive display ads will provide better CTR then static banners
  • Responsive search ads will provide better CTR then expanded text ads
  • A new landing page with less clutter will prove better conversion rates
  • A new landing page with a different hero image will provide better conversion rates
  • Ad copy with a question rather than statement in the first heading will provide better CTR
  • Running ads at a lower position will provide a better conversion rate
  • Bidding 20% higher on desktop devices only will improve conversion rate

Note that the hypotheses are specific. We are testing for only one outcome and using a specific metric to evaluate.

Aside from campaign experiments, we also ran several “ad variation experiments.” These are slightly different from campaign experiments, as they can be cross-campaign. This goes beyond the scope of this article, but we strongly recommend running these as well.

Below are the results of four experiments we ran:

1. Ad copy change example

Hypotheses: Ad copy with a question rather than statement in first heading will provide better CTR.

What we changed: Adjusted ad copy for heading one for all ads in the draft to be question-based.


Decision: This experiment ran for 18 days. The CTR increased by 1%. Results were not significant so we decided not to apply.

What we learned: There was no performance increase in having question-related ads rather than statement ads in a general sense. These needed to be adjusted on a case by case basis, based on the search query and ad.

2. Responsive search ads

Hypotheses: Responsive search ads will provide a better CTR than static search ads.

What we changed: Introduced responsive search ads into the campaign draft.


Decision: This experiment ran for 47 days. The CTR increased by 1%. Results were not significant. We still decided to apply the results, since the responsive ads did not harm performance and they were a new feature allowing us to rotate more ad copy.

What we learned: Despite not having a performance increase, we see that searchers engage well with this new ad type. We were able to minimize risk through the experiment. We continued to monitor these ad types after implementation and performance have been strong.

3. Landing page changes

Hypotheses: Adjusting the hero image on the landing page from male to female will increase conversion rates.

What we changed: Adjusted the hero image.


Decision: This experiment ran for 30 days. The conversion rate increased from 7% to 14.88%. We applied the experiment and only ran the new landing page moving forward.

What we learned: The increase in CVR was dramatic and showed that a small change, like changing the gender of the image can have a dramatic effect. We also learned that it’s likely that users respond better to female imagery in general.

4. Target CPA bidding

Hypotheses: Automated bidding (Target CPA) will provide a better conversion volume performance than we currently achieve with manual bidding.

What we changed: We set a target CPA in the draft, at the same CPA we were already achieving in the campaign with manual bidding. The hypotheses would test whether we can achieve more conversions with target CPA bidding.


Decision: This experiment ran for 34 days. The experiment achieved 53 conversions, the original campaign achieved 70 conversions and had a lower CPA as well. Therefore we decided not to implement target CPA bidding for this campaign

What we learned: Automated bidding strategies are not ideal just yet. We should add as well that the target CPA has worked better in other campaign tests we ran. We’ve spoken to Google and their recommendation was to run the target CPA draft for longer. We agree, but this is not always practical for clients with limited budgets to burn through.

Experiment considerations

As a final note, there are two key issues that are not widely discussed or reviewed in Google help articles. They are very important to consider when setting up experiments.

Having a goal in mind when generating this hypothesis is critical. You should write down in a notebook what the goal is and what metric you want to test. Defining the metric is also critical at the outset since it easy to stray from it. For example, if you are testing a new ad type, then your hypothesis should probably be written in terms of CTR and not CPA. Your results might show a better CPA, but this should not sway your decision since your hypothesis is framed in terms of CTR and you should not apply the experiment!

Another way you can get a false positive result is due to time design issue. This occurs when the experimenter increases or decreases the run time of the experiment to achieve a significant or desired result. This occurs unknowingly, the experimenter doesn’t realize that they are creating a false positive. Think about it like this: if we increase the run time for another week we might get a significant result, if we increase another week further we may get a non-significant result, so altering the period to suit our needs is not a fair test. Even in well-designed university experiments, this bias occurs.

It is important to set a time frame before the experiment starts running and stick to this. To counter this, I include the end dates within the experiment title so I know when it has to end. As a rule of thumb experiments should run for at least one month. You can also try using an ab test sample size calculator If your testing for changes in conversion rate.

The post A case study for delivering performance in a mature Google Ads account appeared first on Search Engine Land.

Regression analysis to improve Google Ads performance /regression-analysis-to-improve-google-ads-performance-313898 Wed, 13 Mar 2019 17:29:08 +0000 /?p=313898 Using advanced techniques to make better predictions can help you stand out. Here's a step-by-step guide to learning how to do a regression analysis.

The post Regression analysis to improve Google Ads performance appeared first on Search Engine Land.

Advanced digital marketing requires us to go beyond what everyone else is doing and approach from new angles. One of the ways to stand out in your SEM analysis and performance is through advanced techniques like regression analysis. Regression is actually a form of basic machine learning (ML) and a relatively simple mathematical application. This type of analysis can help you make better predictions from your data, beyond educated guessing.

Regression might sound scary, but it’s not that advanced in the world of mathematics. For anyone who’s passed year 10 maths, you have probably already worked with regression formula previously. We’re going to look at using regression in your Google Ads to predict the conversion volume you can achieve by adjusting campaign spends. Building the model and applying it is far easier than you would think!

What is regression?

A regression model is an algorithm that tries to fit itself to the presented data best. In essence, it is a line of best fit. It can be linear, as a straight line through the data, or non-linear, like an exponential curve, which curves upwards. By fitting a curve to the data, you can then make predictions to explain the relationship between one dependent variable and one or more independent variables.

The plot below shows a simple linear regression between an independent variable “cost” (daily spend on Google Ads) on the x-axis and a dependent variable “conversions” (daily conversion volume on google ads) on the y-axis. We have fit a linear regression line (blue). We can now say that at $3k on the axis, that point on the regression line would match up to 35 conversions. So, based on the regression model fitted to the data, if we spend $3k, we are predicted to receive 35 conversions.

Headstart on feature selection

I’ve been running many of these regression models and I’ll share what I’ve found to be true, which will give you a headstart in where to start looking

Multiple regression is where some independent variables are used (rather than just one, as in the example above), to predict one dependent variable. With Google Ads, I’ve found that there is always one independent variable that is the strongest predictor of conversions. You could probably have guessed which one it is already.

When running ML model’s on daily labeled training data to predict whether certain features would lead to a conversion, we continually found that all other things being equal, campaign spend is the strongest predictor of conversion volume.

The following table shows the “Root Mean Squared Error” (RMSE) for different ML models.

RMSE is a measure of error, it shows how far off the fitted model is from the training data.  The lower the error the better – it means the model is more accurately fitted to the data. (2) All features include: Day of week, keyword, CTR, CPC, Device, final URL (landing page), ad position & Cost.  

We ran five different machine learning algorithms: Decision Tree, K Nearest Neighbours, Linear Regression, Random Forest and Support Vector Regression. In most cases, removing “cost” as a feature in the data set, increased the error value by more than removing any other feature. This means that the model became less accurate at predicting the correct outcome.

We can also analyze the feature importance used by the random forest (the best model). It’s clear that cost is the key feature the algorithm is using to determine its results:

This shouldn’t come as too much of a surprise – the more you spend, the more likely you will receive sales. Using cost as a predictor for sales is a great place to start your regression analysis.

Building a regression from scratch with Google Ads data

Here we’ll show you how to build a regression model with “daily cost” as the independent variable and “daily conversions” as the dependent variable. We’re going to do this in 5 easy steps.

Note: This will only work with a Google Ads account that has conversion data in it.

Step 1 – Create report:

Within Google Ads, navigate to Reports >> Predefined Reports >> Time >> Day

Step 2 – Prepare report and download:

Once in the report (screenshot below), select the “columns” button (red box), then remove all columns except “Cost” and “Conversions.” Then select a date going back one year from today (blue box). lastly, download the report as an “excel .csv” file (green box).

Step 3 – Generate scatter graph in Excel:

Open the excel file and select columns that contain only the “cost” and “conversions” data. In the example below, cells C3:D17. Then in the menu bar select “Insert’ >> ‘scatter graph.”

Step 4 – Generate regression line on scatter graph:

We’ve now got a beautiful scatter graph portraying “cost” and “conversions.” Generate a regression line by right-clicking on any of the data points and selecting “add trendline.”

Step 5 – Choose best regression line using r-squared:

In the menu on the right-hand side, you are now able to select different regression options (red box). Select the checkbox “Display R-squared value on chart” (pink box). In a general sense, the higher the r-squared, the better the fit of the line. As you cycle through different regression lines, you can view which has the highest r-squared value. You can also decide visually which appears to fit best. Next, add the regression formula for the fit you have chosen (green box). We will use this formula to make predictions.

Making extended predictions using the regression equation

The regression line that we have just created is extremely useful. Even from a visual perspective you are now able to visualize what your expected daily conversions will be at any point of daily cost.

Although this can be done visually, using the regression formula is more accurate and you can also extend the predictions off the graph. In the example below that I have plotted (with a larger account), the regression equation is given as y = 28.782*ln(x) – 190.36.

In the equation y represents conversions, and x represents “cost.” To predict y for any given x, we replace x with a real number. Let’s assume a cost of $5,000. We say y = 28.782*ln(5,000) – 190.36. Using a calculator, it comes out to 54 conversions per day.

Now the real power here comes when we extend this calculation beyond the graph to where spend has not been before. The data points on the graph show the highest spend ever performed per day was under $7,000. If we replace x with 10k, (a predicted spend of $10,000 per day), I can get an estimate using the formula, of 74.7 conversions per day.

Bonus: Finding Optimal points or diminishing returns with CPA

Graphing the “cost” and “conversions” together is extremely powerful for being able to predict conversions at different spends. But in reality, often we’re more interested in minimizing CPA or predicting conversions at a specific CPA. We can similarly graph CPA against conversions to better understand this.

From the CPA chart on the right we identify a minimal point where CPA is lowest on the cost dimension, this is the bottom of the ‘U’ shape. This point also corresponds on the left graph (cost vs. conversions) with the green line.

Using this methodology we can now identify the lowest CPA potential, at what cost this occurs and then also predict how many conversions we would receive at that point. The same can be done for any point on the CPA line.


It’s critical to mention that regression uses historical data only. All of the costs and conversion data is based on what has happened in the past. Therefore if you expect your performance to improve and conversions to increase in the future, this will not be taken into account in these models. To adjust for this, taking more recent data only, such as six months back or three months back could be a better option. Similarly, you can remove or include “days,” during sales periods that may or may not be relevant, in order not to skew the data.

Case studies and application

Using this methodology, we have been able to achieve three key outcomes with clients:

  1. We have helped existing clients estimate what will happen if they increase their monthly spends by $10,000. This is a very common client question and this method is better than educated guesses since it is modeled with data.
  2. We have been able to show existing clients where the optimal CPA lies and how much potential exists in the account. For a major client of ours in the competitive legal space, this has allowed them to decrease CPA’s by over 20 percent and keep conversion volume steady.
  3. than has made new account audits faster and more accurate for us. Without knowing too much about a new client, we have plugged in  historical “cost” and “conversion” data into a regression model to visualise whether they are spending the optimal amount they should be and discover the potential down the road.

Further exploration

Consider that many businesses are interested in revenue and ROI, rather than conversions and CPA. The same techniques can be used to predict revenue as well as options to maximize ROI (we look for maximal points rather than minimal). I’m currently building a PPC optimization tool to automate this graphing and prediction process.

The post Regression analysis to improve Google Ads performance appeared first on Search Engine Land.