How long to implement AEM from day one of contract signing?

[adamb0mb]

Hi, I'm a consultant, and have implemented other content management systems for clients. This one has a hard-coded site, is a furniture company (60,000 assets and online shopping cart). Let's say they sign up for AEM, and have internal resources (IT, UX./UI, designers, info architects, database admin, etc.). Let's also say that they have a good foundation for sophisticated project management. But they don't have any experience with content management systems.

Would a migration to AEM take 3 to 4 months? 5 to 7 months? 9 to 12 months?

Thank you for any advice, thoughts, or even your slams for my asking what you think is a dumb question.

My research shows that Adobe refers to migration being "seamless", however, Gartner warns (in the Magic Quadrant study):

"Adobe customers often underestimate the effort and complexity of an Adobe Experience Manager implementation, especially in the initial phases. Adobe's acquired offerings have introduced various architectures and repositories under the cover of its Clouds, and customers sometimes find that product complexity is increasing and integration is difficult — far from being "seamless," as is advertised. Unifying the architecture is, of necessity, a work in progress."
                                                                           Magic Quadrant for Web Content Management, July 2017

[VeenaVikraman · Accepted solution]

Hi Adam

          For last 5 n half or more years , I am working with AEM. I have been a developer for many projects starting from scratch projects to many migration and maintenance support ones. I totally agree with the above statement. Always , the expectation with the tool is not set properly to the clients when estimating or pitching for the same. Let me tell you frankly, AEM is not a plug and play migration. The content structure and the way the content is rendered does differ from many content management systems. I don't have the authenticity to talk about other systems since AEM is the only one which I have ever put my hands on.

  - to answer this question , I would recommend a proper estimation

     1) Estimate the effort for the UI development - are you going to migrate as it is or are you going to do a revamp of the website ?

     2) When the UI is being designed , make sure to consider the re-usability factor of the components you are designing? (In most of the cases I have seen , without proper planning , all end up creating a lot of templates and a hell lot of components for something which would have been more simpler )

     The best thing to do is to have a proper requirements study for a month and come up with realistic time line. I can add up a lot of stuffs here from my exp , but at this point those may not make sense to you. Since you have a lot of experience with CMS systems, i would recommend to consult with a very good AEM architect and come up with the time .

Thanks

Veena

[adamb0mb]

Veena,

Thank you for such a thoughtful reply. I love your comment about doing a requirements study, and finding a very good AEM architect.

You asked if we are thinking about "migrating" or "revamping" the existing website. I'll share my thoughts, and I'd welcome your feedback. I'm just learning about this new client, and your input is invaluable.

Today, the existing site is hundreds of pages (not thousands), the sector is furniture (so some seasonality, but not a rapidly changing inventory), with around 50,000 assets (pix), with an online checkout process and chat. Not a very complicated site. For this reason, I'm thinking that a "start-from-scratch" approach is best, where we build the AEM site offline, test until perfect, and then take down the old site... launch the AEM site (open to your thoughts!).

I'm about to find out how proficient the IT team is, but I'm guessing that the AEM implementation requires a different skill set than what's on the team (because today's site is conventional html/java/containers). I'm thinking that to get to an AEM instance, it might make sense to use Adobe's Manged Services Enterprise, a cloud-hosted solution. What I can't seem to find out (from Adobe) is if we can launch on the Adobe-hosted site, and later migrate the whole site to the furniture company's servers. Again... any thoughts?

Thank you,

Adam

[VeenaVikraman]

- I am not sure what you meant by Adobe hosted site ? (like fully owned and operated by Adobe ?). Basically as far as I know when you purchase the product (AEM here), Adobe has the pricing according to the number of servers you are getting. In our case we have DEV , QA , STAGE and PROD, each with 1 author and 2 publish. 2 publish is the minimum recommendation for not so heavily traffic website. I don't believe adobe will take the cost and effort to host your applications in some server other than the one your company has purchased.  Again I don't believe myself to be fully qualified to talk about the infrastructure and stuffs.

     I would really like Jörg Hoh​ smacdonald2008​ Feike Visser​ orotas​ or anyone they think can guide you proper to have a better understanding.

  - Regarding the skill set, have a team of developers who are good in Java and have a good knowledge of AEM. (Even if they are new , have some people who are enthusiastic enough to learn it fast. ). Little knowledge on the technology is always harmful, I have seen a lot of projects which was developed so badly that it is a hell task for the team which maintains it. To satisfy client and meet the delivery time (which unfortunately you only commit to the client) , you end up creating the project in such a way that , it will become a helluva task for the client and the team to maintain it going forward.

[Jörg_Hoh]

Hi,

I cannot give hard data, even not an estimation for this, because of the lack of data and details. From my point of view any migration to AEM is determined by these factors:

* The will and the ability of the business to adapt to the new system. Each system comes with a different philioshophy in terms of user interface, interaction model, data structures, workflows etc. Trying to mimick the old processes in the new system is a source of constant pain and lot of adaptions, which are very costly.

* How familiar are the developers and the architects with AEM? When they are all Spring experts and need to work now an AEM, it's getting hard. They likely try to re-use Spring bestpractices and apply them to AEM. Might work well, but I would assume, that the hierarchical content model will cause a lot of headaches then. And replace Spring by any other technology stack. Get a trained team including architects to support and train you and accept their way of doing things.

* The migration process itself. If you need to take over substantial amount of content from an older system, you typically want to filter old stuff and to map it to the hierachy in AEM. Often you need to re-think your content architecture and the way you store data. And then create a process to automate it.

I know this is quite abstract, but it should give you some insight into the things which I think you will face. I would not recommend to undertake such a migration on your own if you don't have experience with AEM. It's easy to spend time on a certain path and learn months and years later, that your approach does not work at all, and you just hoped that it will work. Get experienced people on board which advise how to design your application and content structures.

Jörg