How to understand the search engines and grow your website!

learn how to use to grow your site & your business!


Once upon a time in a Starbucks, far, far away, there was a man with a website. His name was Tom...

Tom did love his website. He loved it a lot! He named it 'Wesley'.

Et il voulait toujours ce qu'il y avait de mieux pour Wesley. Il voulait que Wesley soit heureux !


But what is it that will make a website happy you ask? Websites are notoriously hard to please little creatures, what with all that HTML, CSS, JavaScript, AJAX and other hardly mentionable acronyms? Well! It turns out that for once there is a simple answer!

Wesley, comme tous les sites web, ne voulait que deux choses :

  1. Lots of people coming in, looking for what Wesley has to offer (or, as in people wearing grey suits like to say 'Lots of qualified traffic')

  2. Que le plus grand nombre possible de ces visiteurs fassent ce que Wesley (et Tom !) souhaitent qu'ils fassent. S'abonner à un bulletin d'info? Acheter un cyclotron (ou deux)? Contacter Tom pour plus d'informations? En fait, qu'ils fassent ce pour quoi Wesley a été conçu. Beaucoup de visiteurs c'est bien, mais si personne ne fait rien, c'est très ennuyeux pour un site web !

    For those of you in suits, we are talking of course of maximizing Wesley's conversions rates (plural, because who knows? Maybe we want the visitor to do several things!)

Let's take a look and see how these two things might be achieved...


Let's start from the beginning and see what Tom did to drive people to his site. We are simplifying a bit but the main ideas are what matters!

There are two main ways to bring people to a website

  1. Make the search engines fall in love with Wesley, so that they send him lots and lots of visitors.

  2. Acheter du trafic à des sites à forte fréquentation (et portant sur un sujet similaire au vôtre), ou bien acheter du placement par mots clés sur des moteurs de recherche comme Bing, Yahoo ou Google


a. Search Engines looking at me...

Getting the search engines to love your site is no easy task. As a matter of fact a whole industry sprung up, trying to deal with that problem: it is called 'Search Engine Optimization' (SEO). Here is a quick summary of what search engines are doing when they evaluate a site for ranking:

  • Lis le site et essaie d'extraire du contenu le sujet traité afin de voir à quelles questions il pourrait répondre

  • Analyse de la structure du site. Quelles pages sont importantes? Quelles pages le sont moins?

  • Regarde le nom de domaine et décide de son 'autorité'. D'autres domaines ont-ils un lien vers lui ? Ces domaines ont-ils eux même une autorité ?

Once a search engine has all that information about a site it is able to decide if it should include it in the results returned for a given query. Well established sites go on top of course, and newer or smaller sites tend to be found in page 2 or 3 (or more!) of the results.


Whenever someone finds your listing and clicks on it (even if it is on page 8), the search engine sits and watches what happens. Did the visitor stay on the site a long time? Did he look at a lot of pages? Or did he come back right away and clicked on another result?

In the first case, the search engine will assume that the visitor was happy with finding the link, and the site gets a bump in the rankings. From now on its average position will be a bit higher. In the second case, the opposite happens of course, and the site drops. The process is then repeated continuously.

In other words, search engines measure people's engagement with your site, and if they like what they see, you climb in the results!

b. How do I maximize people's engagement?

When a visitor clicks on your site, you want to be ready and make sure that he stays as long as possible, and sees as many pages as possible, and you DEFINITELY do not want him to 'bounce' off your site and go back to the search engine to try another result! vous fournit l'information dont vous avez besoin pour y parvenir :

  • Engagement Visiteurs par Pages d'Entrée. Quand un visiteur entre sur votre site sur une page ou sur une autre, combien de temps reste t-il en moyenne? Combien de pages va t-il voir? Et quel pourcentage rebondit sur votre page d'entrée pour repartir sur l'URL d'origine?

    Visitors Engagement by Entrance Pages

    Find a page where the bounce rate is highlighted in red? You might have a problem with that page: its bounce rate is worse than the average for your site: you need to go take a look and see what might be turning visitors off. Same thing if you get a time-on-site or a number of page views highlighted in red.

  • Engagement Visiteurs par Largeurs d'Écrans. La largeur d'un écran détermine la façon dont votre site s'affiche. Que se passe t-il sur les écrans très larges? Ou sur les téléphones? Y a t-il des tailles d'écrans qui posent problème a vos visiteurs? vous le dira !

    Visitors Engagement by Screen Widths
  • Engagement Visiteurs par types de Navigateurs et d'OS. Les différents navigateurs et OS affichent chacun votre site de façon légèrement différente. Avez vous un bug sous Firefox pour Windows qui amène les visiteurs utilisant cette plate-forme à quitter votre site trop tôt? Si c'est le cas, vous alertera !

    Visitors Engagement by Browser and OS types

With these three reports, allows you to quickly identify problematic entrance pages, scren sizes and platforms for your site, so that you can address them and increase your visitors engagement. After that, it's time to check the results

Après toute modification, il vous faudra attendre un peu et voir l'impact sur l'engagement des visiteurs sur le site dans son ensemble. Avez-vous amélioré les choses ? Nos rapports journaliers vous le diront :

Visitors Engagement over Time

In conclusion

The reports we just showed you help you understand where and how your site may be under-performing, so that you can fix things and monitor your progress. As your site gets better at retaining visitors, your rankings will increase in the search engines. They will start loving you more and more and sending more and more people your way!

But what about the other way to get traffic? Purchasing visitors? Can help me with this?


a. Don't want to wait? Buy your way to the top!

There is a sure way to get as much traffic as you want, if you have the budget, it is to buy it. Search engines like Yahoo or Google sell placement at the top of their search results. If someone searches for a keyword that you have purchased, there you are popping up in the #1 spot! (and the search engine charges you the agreed-upon fee for each click on that link that brings a visitor to your site). This can get expensive, but if the return on investment is positive (if the traffic you are getting is buying your product), then it may be well worth it.

It then comes down to measuring your results. First of all, since you are paying per click, you do not want to be charged for double-clicks (someone who clicks on your link twice in a short period of then). Then, you want to be able to measure and compare your advertising campaigns performance. can help you do both.

b. No double-clicks for me!

One of the frequent uses of is to check that you are not being overcharged by the company selling you advertisement. They will claim that they sent you x number of visitors over the last week: you need to confirm this number using's Referrer Summary report.

In this example we can see that our advert on Google Adwords for one of the keywords we purchased (keyword 1) has generated 1,126 visits. This number should roughly match the number given to you by Adwords (small differences are OK. Large ones are not)

Referrers Summary

Do not hesitate to ask the people you buy advertising from for supporting data (specifically the timestamp an IP address of each visitor for which you have gotten charged). You can then compare with the same list, as generated in your 'visitors details' report. We filter all double-clicks out (and also get rid of all artificial visits, like robots), so the data that we show you is clean and can be relied upon.

Our Detailed Visitors Report

Detailed Visitors Report

Note à propos de Google. Si vous utilisez Google Adwords pour générer du trafic sur votre site, assurez-vous que vous n'utilisez pas un autre produit Google (Google Analytics) pour mesurer les résultats de ces campagnes. En effet vous créeriez ainsi un conflit d'intérêt. Il vaut mieux utiliser un tiers pour effectuer ces mesures., ou d'autres, mais pas Analytics.

c. Mon argent est-il bien dépensé ?

Now that we know we are paying the correct price for our new traffic, it is time to evaluate the performance of our investment and if we have several advertising/keywords campaigns, to compare them. This way we will know if the money we are spending is actually going to a good use. We will be able to stop investing in poorly performing ads or keywords and reallocate that money to our star performers!

For this we will look at the Referrers Report again. First we look at the engagement of the visitors coming from our different campaigns or keywords.

In this case we will concentrate on the bounce rate: this is the most important metric. Time-on-site and number of page views are important also, but bounce rate takes precedence. If visitors follow a link to you site and leave right away it means that they were not correctly targeted (assuming the landing page does not have bad issues in need of fixing, per the previous section)

So. What does the report below tell us and what can we do to optimize our advertising and allocate our money wisely?

Referrers Engagement

We have several campaigns going (everything marked as 'organic' is natural search engine traffic that we are getting for free)

  1. Google Adwords (nous avons acheté trois mots clés et payons pour chaque clic)

    • Mot clé 1 : génère beaucoup de trafic, mais le taux de rebond de 65% n'est pas bon. Nous devrions arrêter de dépenser de l'argent sur ce mot clé.
    • Mot clé 2 : génère peu de trafic et le taux de rebond est tout aussi mauvais (63%). A arrêter également.
    • Mot clé 3 : fonctionne bien mieux que les deux premiers, mais le trafic est modéré. Nous devrions prendre une partie des sommes dépensées sur les mots clés 1 et 2 et les allouer au mot clé 3.
  2. Pubs Yahoo (nous avons acheté deux motc clés et payons pour chaque clic)

    • Mot clé 1: ne fonctionne pas très bien. Trafic modéré avec un taux de rebond trop élevé à 57,2%. Il faut arrêter cette campagne
    • Mot clé 2 : c'est notre meilleure campagne de publicité sur moteurs de recherche. Elle bat toutes les autres de loin avec un taux de rebond de seulement 17,6%. Nous devrions utiliser une partie des sommes budgétées pour Adwords et les consacrer à ce mot clé.
  3. Campagnes email (nous avons acheté deux listes d'emails ciblées et avons envoyé à chacune le même message)

    • Liste 1 : notre première campagne email fonctionne très bien. La liste que nous avons achetée doit-être d'excellente qualité car le taux de rebond est bas et les visiteurs passent beaucoup de temps sur le site. C'est notre meilleur investissement pour l'instant !
    • Liste 2 : ce message a moins de succès. Nous avons autant de visiteurs mais un pourcentage plus important ne reste pas sur le site. Cela nous montre que la liste est moins bien ciblée.

In short

As you can see, the Referrers report tells us a lot and allows us to take immediate action. Specifically we are going to cut back on our overall Adwords expense by getting rid of keywords 1 and 2. The money we save will go in part to keyword 3 and in part to Yahoo's campaign for keyword 2. We will also continue buying email lists from our first provider and send more marketing messages.

Now that we know how to optimize our campaigns, we can ask the next question...

d. Est-ce que je GAGNE de l'argent ?

We are now at a point where we have optimized our site and our campaigns. We are doing the best we can... But does that mean our investment is paying off? For this we need to look at our conversion rates.

A conversion is counted when a visitor takes a desired action on your web site. For example, imagine that the site sells widgets at €45 each. You could setup to count a conversion of type 1 when a visitor views the widget presentation page, and a conversion of type 2 when he reaches the purchase confirmation page (and has paid us €45)

For illustration purposes, we have isolated one line from the referrer report with the conversions shown. This is our Adwords campaign for keyword #3 that we have reworked, improved and funded. We are paying 1.75 per click for this campaign. We now have a bounce rate of 10.6%, which is very good. Let's look at the report:

Referrers and Conversions

What we can see is that of the 1,014 visitors sent by the campaign, almost 30% viewed our product page, which is very good, and 4.8% ended-up making a purchase at €45 each. That is a revenue of 1,014 * 4.8% * €45 = €2,190.24.

Since we paid €1.75 per click and had 1,014 visits, our total cost is €1.75 * 1,014 = €1,774.50. We therefore made a profit of €415.74

Since you are using, you can conduct that analysis for every campaign you run. The data will be shown, along with conversions, in the Referrers report. Note that sometimes, assigning a monetary value to a conversion is not that easy. What is the worth to you of someone signing-up for your newsletter? Is it worth €5? €100? It all depends on your particular activity. Sometimes you have to do the best you can and estimate the value of a given conversion.

In conclusion

By alerting you if you are being unduly charged for double-clicks, by allowing you to compare your campaigns to each other, and by showing you whether of not your campaigns are profitable, allows you to purchase traffic the smart way, and makes your web site as efficient as it can be so that it can produce the results you want.

Assurez-vous de profiter de tout ce qu'il a à offrir !

Addendum : lisez ce document pour apprendre à configurer vos campagnes emails ou pay-per-clic pour une détection optimale par


To be continued...