On the morning meeting with Steve. As we were going through emails, we had more requests for custom homepages and even some users wanting to go back to the classic invoice homepage vs the newer graphical invoice homepage. We then started talking about all of the different levels of custom...

We went into corp-wide settings, group settings (12 main player groups), page level settings (page control - how does it look and what does it contains and how does it work), and clear down to personal settings.  That got us into a small talk and conversation about an adilas "fracture" account. This is a concept of breaking everything down into mini pieces that could be stringed and/or put together in any combination. This could include pre-sets, pre-built options, special skinning, etc. It could also get into real in-line database extensions, themes, core black box options, etc.

What we are really selling is... We have tons of things pre-built and have a viable solution that has tons of functionality. This is ready and out of the box. We then offer people the chance to build their own application and virtually extend the functionality to what they really want. Basically, Steve was talking on a sales level, and was thinking about having a number of checkboxes... when meeting with a company, you could virtually check the boxes of what we have, we could also allow them to then pick and choose what they want to add on. Basically, we, as adilas, could tweak out our own product and create all kind of custom wire jobs.

We need to apply our own client filters and see who wants to play. We don't want to get over our heads, but we have a great number of guys who can do and handle certain pieces. We have a solution that is usable, right out of the box. We also have a platform that is customizable and moldable (like clay). Steve was also talking about selective selling and really selling our model vs just the system. Not everybody is a fit for what we do and/or offer.

Idea from Steve... Run the system for 6 months and get good at the system. Then come back and tell us what you want to tweak out and/or change.

People want major functionality... however, they also want it to look a certain way (how pretty is it), they also want a level of control (what can they tweak and change).

Function, form, and control - Alan was mentioning the word "Control" - that gets clear down into that fracture type level. We could have preset templates, show/hide settings, toggle on/off settings, permissions and sub permissions, main pieces, sub pieces, and even show/hide of the different modules. Currently, we have everything available... we may actually want to look at module level controls. For example: do you use payroll? do you use stock/units? do you use x or y?

Idea from Alan... What if we did some testing on new users and had them do certain tasks. We then see how intuitive the process are. This would be a refining process and then work from there. As a side note, we have been adding on for years and years. That is our current model. Maybe we need to rework our inter core and make it more module based and then be able to show/hide and/or be able to toggle things (full sections and/or pieces) to an on/off level.

We also talked about packaging and limited packaging of the system... For example: The limited package, the standard package, the deluxe package, and the mother load... package. Not sure what to call things... but it could be really cool. Basically, it all exists, we just show it (smoke and mirrors) in certain ways. Based on the package style and/or level, we could show/hide other pieces and/or features. Another analogy is a race car engine and then we offer the body styles like cars, motorcycles, vans, trucks, semi trucks, or even UFO's (see attached for an older image).

Adilas has a lot to offer... but what if I only need it for this or that? It sometimes feels like taking an army tank out to hunt birds, say something like pheasant hunting or duck hunting. Almost an over kill. What if we could dumb it way down and make it more approachable. Make it into smaller bite size pieces and then let them get deeper if they want. As a side note, like the ice berg analogy, we only show the most basic pieces. Then, if they want more, we allow them to get to it. Maybe either offer certain packages and/or hide everything by default and make them turn it on (at a corp-wide or world level and also at a payee/user level).

Steve - how do people want to buy things? We then need to cater to that level. See the photo gallery for an inverted funnel analogy. Currently, we have people enter with all options (like the top side of the a funnel) and then have them select what they want to get to the smaller level. What if we flipped that model and started out super small and basic... then they could either buy and/or turn of additional modules and/or features as they need them. So, basically, we start super small (like the bottom of the funnel) and work outward (getting bigger like the top of a funnel).

Alan - would like to do a cognitive walk through... What path do they take to do a task? Where did we want them to go and/or do? Where are their eyes on the page and where are they spending the most time? How easy was it for them to follow the path that is wanted? The cognitive walk through would include a small user group, and then the persons being tasked with the tasks would have to explain (verbal and/or written) what and why they are doing what they are doing. Basically, getting into the minds of the users and what they are seeing, focusing on, and/or actually doing.

Wire frames (interactive outlines) and getting to the basic decisions and taking that research back to the coding project. What is the phycology behind the decisions. What are the human levels that we might be missing and/or could really enhance just by display, menus, and structure. In Dustin's words... "putting rails on things".

We also talked about doing different cognitive walk through groups (users not familiar with the system) and how they would do their own different tasks. If they get stuck, why? What were they looking for and what were they expecting? This is more of a cognitive (brain activity) exercise to see what the thought process is and/or could be. One of the challenges for us might be - so many different business verticals and so many different choices... we might need to get a sample and make sure that we aren't hitting just one niche and/or personality type. Lots of different flares and possibilities. We would like to match who would be using the program with what we would be asking them to do. This testing also has some down-sides, such as preconceived notions, using other existing and/or older products, and actually testing people who would be using it.

Alan and I were talking about who we knew and what resources we could tap into to get some basic testing. Just throwing around some ideas. We added in a new community funded project to get in there and do some research and some of the cognitive walk trough's.

After that, Alan and I were reporting on our different projects. We love the options of having black box code, but there are teeth to that as well. We are starting to see the black box section as a huge time sink for the developers if they are making core changes. We have code that is not in master (in the main code repository), it could be changed at any time, and no one knows who has the last cookie. The developers work directly with the clients, and thus, they think that they have the latest code, but a new change could have been cascaded through to the other pages. We definitely need the black box code options and framework, but we need to eliminate as much duplicate code as possible. It is becoming an issue. It is not on fire, but it becoming a thing that needs to be addressed.

Wayne popped in and reported on the database migration project. He is planning on running some new tests tonight. He went backwards and did some major back testing and going back to a test driven design philosophy. He said that he got a great reminder on why we need to stick with a test driven design philosophy (watching out for the little things). The data migration process is super deep... it is basically allowing two different servers to run as they are... then in the background it takes data and tries to re-sync the data to get a perfect match, even though they are different databases with different auto id numbers. That is advanced mixing and blending.

We got into a bigger discussion about standardizing data and using special tables such as flex grid tie-ins and custom dates, numeric, and custom text fields. Some of those tables are great double agents and/or chameleon type tables (they switch as needed). Good stuff and it really seems like Wayne is trying to think beyond the current project.

We are constantly chasing a moving target... part of the fun... :)