http://www.cmo.com/features/articles/2016/3/9/data-decisions-dsp-vs-dmp.html

As marketers assess their requirements for marketing technology, the question facing many looking for platforms rather than tools and systems will be whether to invest in a data management platform (DMP).

Certainly, “DMP” seems to be the new industry buzzword playing out in the field. According to an Econsultancy study, 68% of marketers believe DMPs are “key to the future of programmatic marketing,” so it is understandable that many would believe that employing one is an absolute essential, and for some companies this will definitely be true. It might not be for everyone though.

A Brief Intro DMP/DSP
While once the overlap between what a DMP and what a demand-side platform (DSP) offered was minimal, today most DSPs also offer some analytics and data management tools. Traditionally, the DMP gathers and sorts data, while the DSP was used to activate insights on, for example, digital media buys. Analyst firm Gartner says that DMPs and DSPs are “codependent.” Before embarking on the process of engaging a DMP, it could be worth investigating if your demand side platform (DSP) partner could step up to the plate and into the DMP’s shoes.

When Is A DMP A Necessity?
There is no one-size-fits-all answer to this question. Whether a DMP is a necessity or not varies from company to company. The key to finding an answer to this question often lies in the volume and type of data that your organisation holds. Broadly speaking, an organisation that holds large volumes of CRM data is likely to require a standalone DMP. On the other hand, companies that mainly rely on behavioural data for marketing campaigns will often find that their ideal setup is to work with an advanced DSP that offers data management and analytics tools at no extra cost.

Looking at two quintessentially British brands, the difference becomes clearer. British Airways gathers huge amounts of CRM data, which can provide insights into their customers past purchase behaviour and influence future purchase decisions. A DMP becomes fundamental in processing this data, in order to segment customer audiences for targeted advertising.

But when you consider a brand such as Cadbury that uses digital channels mainly for branding campaigns, the case for a standalone DMP could be less of an imperative—assuming there will be little to no CRM-type data. In this case, a DSP will typically be able to accommodate the use of first-party behavioural data should it provide material value to the campaign activity.

Broadly speaking, the use of data warehousing and the management capabilities of a DMP are more likely to be appropriate if using CRM and customer purchase history data, which adds significant value to your marketing campaigns.

Making The Choice
In the process of choosing a DMP or DSP with advanced data management and analytics capabilities, it is also important to consider some other factors. Marketers are often led to believe that it’s a good solution to work with several different DSPs (one for mobile, one for video, etc.) and then feed all campaign insights back to their chosen DMP. This can easily lead to data loss, as not all systems talk to each other, meaning the cross-device customer journey could become fragmented and, for example, frequency capping would be inaccurate.

Using a single demand side platform could be much easier and more efficient—impression-level user data is fed back into the system in real time as it doesn’t need to talk to another system. And let’s face it, using one system is more cost effective than using two—but only if it’s right for you.

The questions we have posed provide marketers with a good jumping-off point when considering 2016’s data needs. Assuming you have chosen your DSP partner first, it’s worth making sure that, if your data management needs change in the future and you decide at some point that you do, in fact, need a DMP, all of the information is easily transferred to a standalone DMP.

Marketers need to empower themselves to make the right decision for them and their brand, and to pursue industry buzzwords only when it’s appropriate going forward.

Data Decisions: DSP vs. DMP的更多相关文章

  1. DSP, SSP, DMP

    先了解下广告的产业链有哪些人群: 广告主advertisers 显然是指想为自己的品牌或者产品做广告的人,例如宝马.Intel.蒙牛-- 媒体publisers       则是提供广告位置的载体,例 ...

  2. dsp与dmp的cookie mapping

    dsp   ad.com   在  meijiu.com上部署广告. 假设dmp叫cm.api.taobao.com 建立gid映射表 (1) ad.com在meiju.com的页面上部署,指向dmp ...

  3. 中国程序化购买广告解析:RTB/DSP/Ad Exchange/SSP/DMP,思维导图

    中国程序化购买广告解析:RTB/DSP/Ad Exchange/SSP/DMP 概念 程序化购买( Programmatic Buying):通过数字化.自动化.系统化的方式改造广告主.代理公司.媒体 ...

  4. DMP

    1.dmp-data mabagement platform数据管理平台 数据赋能,营销智变 2.定义 把分散的第一,第三方异构.多源数据进行整合,然后纳入统一技术平台中,并对这些数据进行标准化和细分 ...

  5. 【exp/imp】将US7ASCII字符集的dmp文件导入到ZHS16GBK字符集的数据库中

    [exp/imp]将US7ASCII字符集的dmp文件导入到ZHS16GBK字符集的数据库中 1.1  BLOG文档结构图 1.2  前言部分 1.2.1  导读和注意事项 各位技术爱好者,看完本文后 ...

  6. 转-【exp/imp】将US7ASCII字符集的dmp文件导入到ZHS16GBK字符集的数据库中

    原帖地址:http://blog.csdn.net/lihuarongaini/article/details/71512116 1.2  前言部分 1.2.1  导读和注意事项 各位技术爱好者,看完 ...

  7. OMAP4之DSP核(Tesla)软件开发学习(四)ARM核与DSP核通讯示例

    首先,安卓系统完全启动4AJ.2.1. 其次,查看OMAP4的Tesla相关信息,检查Tesla是否使能.(有如下显示,则OK) cat /d/emoteproc/omap-rproc.0/versi ...

  8. 使用Graph工具观察FFT波形

    2014年8月1日,经过很长时间的上网查阅资料,走了很多弯路,终于可以成功使用Graph工具仿真波形了,虽然这个的确很简单,但是经过自己摸索出来的,兴奋之情难于言表. 明天就是七夕了,刚和女朋友分手的 ...

  9. windows与linux之间文件的传输方式总结(转)

    当然,windows与linux之间文件的传输的两种方式有很多,这里就仅仅列出工作中遇到的,作为笔记: 方法一:安装SSH Secure Shell Client客户端 安装即可登录直接拖拉到linu ...

随机推荐

  1. Hive日期格式转换用法

    如果想把 20180123 转换成 2018-01-23,可以使用: select from_unixtime(unix_timestamp('${p_date}','yyyymmdd'),'yyyy ...

  2. Elasticsearch 的 Update更新

    Update更新操作允许ES获得某个指定的文档,可以通过脚本等操作对该文档进行更新.可以把它看成是先删除再索引的原子操作,只是省略了返回的过程,这样即节省了来回传输的网络流量,也避免了中间时间造成的文 ...

  3. mpush 服务端配置 for windows 服务自动运行

    mpush 服务端配置 以下安装部分是参照官方的步骤, 一.安装jdk1.8并配置环境变量 示例:  http://www.cnblogs.com/endv/p/6439860.html 二.Wind ...

  4. mac os x+paralles使用source insight

    将Mac OS X下的目录共享到Paralles后,source insight创建工程.但是当再次打开时却打开失败.提示:there was an error opening project 网上对 ...

  5. Centos 7 搭建OpenStack 私有云——(1)基础环境配置

    1.简介: OpenStack是一个由NASA(美国国家航空航天局)和Rackspace合作研发并发起的,以Apache许可证授权的自由软件和开放源代码项目. OpenStack是一个开源的云计算管理 ...

  6. 倍福TwinCAT(贝福Beckhoff)基础教程2.1 TwinCAT常见类型简介

    常见数据类型可以参考EXCEL表格的内容,在准备值中输入非法数据,然后回车会有错误提示(例如BYTE数据输入256)   右击可以切换二进制或十进制显示,wData5可以被挨个采集每个位上的数据,也可 ...

  7. JAVA Eclipse如何设置编程环境字体

    窗口-首选项-常规-外观-颜色和字体,文本字体  

  8. Java自动内存管理机制

    1.运行时数据区域划分 2.程序计数器 作用:可以看做是当前线程所执行的字节码的行号指示器. 解释:字节码指示器就是通过改变程序计数器的值来指定下一条需要执行的指令.分支,循环等 基础功能就是依赖程序 ...

  9. group by having和connect by

    --使用group by 子句对数据进行分组:对group by 子句形成的组运行聚集函数计算每一组的值:最后用having 子句去掉不符合条件的组.--having 子句中的每一个元素也必须出现在s ...

  10. 去掉input框后边的叉号

    ::-ms-clear, ::-ms-reveal { display: none; }在样式里加上这句话即可