Which Comes First: the Crew or the CMS?
Well, Frankie Lee and Judas Priest, they were the best of friends.
So when Frankie Lee needed money one day, Judas quickly pulled out a roll of tens
And placed them on a footstool just above the plotted plain,
Sayin’, “Take your pick, Frankie Boy, my loss will be your gain.”
- THE BALLAD OF FRANKIE LEE AND JUDAS PRIEST
Corruption in Tech Paradise?
Janus Boye recently posted a thought provoking article on his blog (Is corruption an issue?) which highlights various activities in the online industry which he considers extremely dodgy. He gives some examples of some goings-on which sound pretty shifty. After listing these issues, he goes on to say:
I’ve seen many examples of contracts being signed with a vendor that was not actually the best fit for the project. As my old mentor always used to say: “The best product never wins”. Perhaps he was referring to the fact that many buyers are corrupt.
This got me thinking. He is, of course, correct although the “influencing” happens at many different levels. At one level, the corruption and kickback can get rather big. About two years ago, the US Justice Department sued Accenture, Sun and HP for fraud relating to exactly this. Many technology companies were also looked at (including a few vendors I deal with), and IBM and PWC have each coughed up more than $2 million. I’ve got no idea if the case has ended, or still going on. EMC are being investigated at the moment for similar things, although the events in this case are 10 years old.
But I digress. Before I get to the dilemma, here’s a bit of background. I work for a “full service” digital agency. I’ll write a post on what the hell this means at some point in the future, but in a nutshell the client hires us to do everything. This includes strategy, branding, research, media, creative design, user experience work and a whole lot more. Importantly, it can also include vendor selections, site build, rollout, hosting and support. There are many good reasons to want to have a single supplier perform all of these tasks for you. There are also many reasons why it might be a terrible idea. Maybe another blog post on this later too. I strongly believe that in most cases the positives outweigh the negatives – or I wouldn’t be working where I do – and for the purposes of my argument I’ll trust you to humour me.
A Hypothetical Project
Picture the scene. Client X has an idea for a large public facing web site, and engages a full service agency. At the risk of horribly over-simplifying, someone needs to do the following:
- Step 1: Firm up the requirements to an appropriate level, help with a business plan and plan the project
- Step 2: Perform some user research and testing to ensure the idea is a valid one
- Step 3: Design the user experience (UX), and do the creative work
- Step 4: Pick the tools (for example WCMS, Search and Analytics) that will satisfy the requirement, design and UX
- Step 5: Build and launch the site, which includes integrating the selected third party products
- Step 6: Continually support and improve the site post launch
Now this seems like a sensible order of events to me, with the vendor selection exercises being performed as late in the project as is sensibly possible. The more information we have at point of vendor selection the better. On all large projects, these exercises are formal and involve the customer’s Procurement department, who exist partly to combat the very corruption mentioned by Janus.
So here is the dilemma. Even assuming no vendor kickbacks, no bribery and pure hearts everywhere, how can the agency/systems integrator that is going to do the build possibly be impartial?
Regardless of size, all implementers will be more skilled with certain products. As it is highly unusual that only one of the candidate products in a vendor selection exercise is fit for purpose, the deciding factor will often be which can be implemented in the most low-risk manner. Which boils down to selecting a product that your implementer is confident enough to guarantee delivery on. If using a formal scoring system, and Product A which I know well and have implemented many times scores 86/100, while Product B, which I’ve never heard of, scores 90/100, it will be better for everyone if we pick Product A. A different integrator would correctly select Product B if they have the appropriate skills. Experience is everything in the CMS implementation game.
So does that mean it isn’t ethical to select the implementer before selecting the tool as the “best product” may not win? If that’s the case, many projects are going to suffer horribly. And full service agencies like mine wouldn’t be able to offer the full service with a clear conscience. We’d have three options (referring to the simplified steps earlier):
- Perform Steps 1-4. Do the upfront planning, research, requirements and design. We’d help the client select the objectively best tools for the job (which I believe we can do), and walk away. The client would need to find an expert in the tool(s) we recommend for the build, who’d we would need to work closely with. Even if it was a tool we knew very well, we still couldn’t build as is it may look like the tool was selected for the wrong reasons.
- Perform only Steps 5 and 6. Build the site only once someone else has defined the solution and selected the products. We do a fair bit of this, but this isn’t full service so the client still might end up with all the issues associated with The Agency Finger Pointing Game.
- Perform all Steps except Step 4, and pray to all that is holy that the non-corrupt, impartial vendor selection exercise decides on a tool we can actually use.
In the first two cases, we aren’t performing our “Full Service”. The third case simply wouldn’t work. So if you can’t select the implementation team before selecting your tools, how do we take heed of the advice of one of Janus’ Web Content Management Inconvenient Truths:
It’s the crew and not the tool – forget about finding the best CMS, but do work hard to find the best implementation crew
A Little Crisis of Confidence
Wait a sec. What does this actually mean? I’ve always taken it to mean you pick an Agency/Systems Integrator (the Crew) first and then let them pick the tool (the CMS) for you? And tell Procurement to look the other way? Is this ethical? Maybe it doesn’t mean that at all. Do you pick the Crew and the CMS as a team, using another consultant with no ulterior motives to help you? If this is the approach you choose, you need to select the CMS very early in the process. It certainly isn’t a workable model for an agency like mine. Or do you pick the CMS first (maybe just flip a coin?) and put the real effort into the selection of the crew around that CMS.
Hmmm. Confusing. I’m starting to wonder if we (agencies and systems integrators) should offer formal vendor selection exercises at all. And which does comes first: the Crew or the CMS? Answers on a postcard. Help. Somebody. Please.