1. This website uses cookies to improve service and provide a tailored user experience. By using this site, you agree to this use. See our Cookie Policy.
    Dismiss Notice

Mediabuy Beginner’s Mistakes - A Danger Foreseen is Half-avoided

Discussion in 'Media Buying' started by mobidea team, Jul 30, 2015.

  1. mobidea team

    mobidea team Junior Member

    May 19, 2015
    Likes Received:
    Home Page:
    We are all doing mistakes and it helps us learn lessons, develop, grow personally and professionally. Of course we would all prefer not to be caught by the same trap as other people have been, therefore it is better to learn their lessons. Here is the list of common mistakes of beginner mediabuyers for you to try to avoid and reach success faster:

    - Start slowly ? You cannot embrace the boundless, so don?t try to be everywhere and cover every segment at the beginning. Just choose one or two countries with right targeting and explore it deeply, try to become expert on it, understand what influences your profit in this country and when you will be comfortable with managing it ? then you can move on with other countries. The main goal here is to get used to the platform, its traffic and the optimisation process.

    - Choose easy countries - Opt for stable countries with quite big amount of traffic and small payouts, like this you can gather statistics faster and don?t lose a lot of money. Moreover, you can understand behaviour of traffic easily, so don?t risk where you don?t need to risk!

    - Launch several spots ? Don?t stick with only one ad-format. One spot can work well for one country but fail for another. If you launch only one spot and you are not successful, that doesn?t mean that you cannot get money out of that country, but mayshow that spot for test was chosen wrongly. That is why I would suggest you to start with 2-3 spots to test a country for understanding and see if it has potential.

    - Learn to wait - Don?t get crazy with optimization! When you launch the campaign it will most likely not be profitable at the beginning, you will be losing money, your heart could not stand it and you will have an urge to change lots of things quickly. But please don?t do it! Wait! It is a very important phase: you are gathering stats. So, don?t panic and don?t make changes within first 3-5 days (of course, it depends on the traffic amount you have). You need to have relevant stats to be able to evaluate behaviour of the traffic. Otherwise, you can make changes not considering conversions in delay or based on low-performance hours or even days. Have patience to watch you ?losing money?. But still don?t exaggerate ? if you see that all your money ?was eaten? by ?bad? website or not-convertible banner ? cut it.

    - Don?t optimize too often - Give some days after optimization for your traffic to work and your performance adapts to changes. Remember when you did changes and what exactly influenced on your performance. What you can do is: write what you changed (bid, capping, banner, etc.) in which day and a couple of days after, check the impact of it (the volume increased a lot, the performance is better, and so on). With this, you will have a clear picture of what can be impacted by what. It is important for you to be organized and choose correct period and tools for further optimization.

    - Don?t change everything possible at the same time - When you make some strategic changes (like changing of bid/cap/banner/targeting) you need to understand what exactly influenced on your performance, that is why make one change and wait a bit to check its impact. If it is negative ? you can always go back, but if it is positive you can keep optimizing. And when you optimize your campaign, you cannot consider the parameters individually as you need to know that one change can impact all the other ones. For example, if in one campaign you see that all your websites are negative, the problem cannot be here. Perhaps you are bidding too high, you are purchasing traffic on an OS that is damaging your campaign. Correcting one of these, will make your websites profitable.

    - Don?t use Daily Max Budget at the beginning ? I understand that you want to control your costs and you don?t want to enter in high spends but for the stake of worth data keep the campaign running for the whole day. Otherwise you will take the risk to avoid the best hours or to run only in three or four hours and like this it?s impossible (at least, not correct) to say that the campaign, offer or targeting is good.

    - The golden mean of banners - If you are working with banners ? choose 3-5 of them for one campaign. If you use less, you cannot compare performance. If you use more ? you have to spend more time and money to get relevant stats for the performance analysis of each banner.

    - Split campaigns based on targeting with different performance - Let?s imagine that you are working on an ad-network which allows you to split traffic by carrier. For example, the carrier X has an eCPM of 15? and the carrier Y has 7? and both seem reasonable to work with. If the volume is good enough, the good thing to do would be splitting campaigns by carriers, because competition for these two carriers will be different. For the carrier with 15? eCPM, advertisers will tend to bid higher than the one of 7?. So, basically if you launch only one campaign for both carriers ? for the carrier X you will receive less quality traffic and your performance could be lower than it might be with a slightly higher bid. On the other hand, for the carrier Y, you could receive good quality traffic, but pay for it so much that you could not afford and be profitable. The strategy for both carriers should be different ? for X one bid must be increased, but for Y decreased. And the only one way to do it is to split the campaign.

    - Consider the average payout of a segment ? When optimizing, it?s very important that you always have in mind the average payout of the offers you are promoting: for example, when you optimize, if you have a website that is spending 7? without any conversion, your next step will be different according to the payout. If you are on country where the payouts are big (10? for example), you should let this website run since one conversion will put it positive. If you are on a country with low payouts (0.50?), this is definitely a website to blacklist. I used the example of websites but it?s applicable to almost all the campaign parameters (device, OS, banner, etc)

    I hope these tips will help you to be more organized and save your some money and time. I wish you to learn on the mistakes of others and take only right steps on your amazing mediabuy way. Good luck to all!
    • Thanks Thanks x 2