Client: “You guys can really pull this off?”
Me: “Depends which team you get.”
No, no one at a big tech firm is supposed to honest like that. Anyway, Hertz sued Accenture for $32M for failing to deliver a website.
Yes, a website.
Sure, there are big, fat databases and an integration layer and mobile/desktop and multiple countries. But still. This isn’t rocket surgery. It’s a website for renting cars. On earth.
Clients tended to like me because I told the truth … like that there are good teams and bad teams everywhere. The good teams are amazing while the bad teams will ftp data all day while debating whether Cobol is better than Fortran. And that debate probably added another $1m to the project cost. At some point the bad team will move into your basement, claim squatter’s rights, and host Magic tournaments. Yeah, they’re that bad.
Anyway, I’ll tell you what happened, but first, from the Hertz v. Accenture lawsuit…
“After Accenture put on an impressive, one-day presentation for the Hertz team that included a demonstration of the transformed Hertz digital experience…”
I’ve been in many of those. All I can say is that sales people are really good at what they do. That’s why they get the fat commissions. Their sole incentive, though, is to make the sale. By the time the dumpster fire breaks out, they’re long gone. Pro-tip: always have your in-house technical/project team at these meetings. Always. (Don’t have an in-house team? Then you’re not ready….)
“Accenture served as the product owner”
Wut? Never, ever outsource product ownership. I’m in favor of public executions for these kinds of mistakes.
“Accenture committed to delivering an updated, redesigned, and re-engineered website and mobile apps that were ready to “go-live” by December 2017.”
Wut? So you expected 100% of the product to go 100% live to 100% of your U.S. customers? I’m pretty sure you just confessed that you’re insane. Everyone claims they’re “agile” — well, that’s not agile. That’s not even waterfall. That’s the whole damn ocean. And Accenture failed to boil it. Well. Duh.
Here’s what you do: take some functionality or feature-set and have it go live in some place that doesn’t really matter, like New Zealand.
No, I’m serious. It’s called New Zealanding, because if it crashes and burns in New Zealand, then no one will even know because New Zealand doesn’t even exist (probably). If New Zealand doesn’t work for you, then you perturbation theory your way into your own New Zealand.
Also, each New Zealand test is a checkpoint. Work gets approved. People get paid. You have 2-week sprints? Ok, you also have 2-month deliverables (or whatever). You have enough New Zealands and eventually you have a real country. Or website. Or whatever you’re building. According to the lawsuit, the project went 17 months before “everything didn’t work.” So, there weren’t any deliverables before month 17?
“Accenture failed to perform proper testing of the software that it developed. Accenture did not perform tests on many components of the system. When Accenture did perform tests, they were seriously inadequate, to the point of being misleading.”
You know who else didn’t perform proper testing? Archduke Franz Ferdinand. And we all know what happened after that.
Seriously though, here’s a little dev secret for all those planning digital innovation projects (or whatever term is used on the PowerPoint): no one does QA.
Well, some people do, but it often isn’t done at all and it often isn’t done well. This leads us back to product ownership.
If someone is building something for you, then you own it, no matter what. Don’t rely on the people you currently have — it’s a multi-million-dollar project … so hire a few people. For example:
Server — hire someone who’s (1) very good on the back-end and (2) has nothing else to do. Their full-time job is to understand/approve/improve the server side of the project. (Big project = hire more server people.)
Database — you have one? Ok, hire someone who’s very good whose sole job is oversight of the db on the new website. Oh, you’re spending millions? Then hire more than one person … you know, in case one gets abducted by aliens.
Data — you might have data, eh? Most teams are missing a data person who’s just watching inputs/outputs and making sure it makes sense. Don’t hire a 21-year-old data-theorist. Hire a for-real (e.g. probably expensive) data person. These people are often missing from teams.
QA — all quality control/testing/assurance etc. is on you, no matter what the Accenture/Wipro/InfoSys guy with perfect hair is telling you. You’ll need a few people and you need to hire them pronto. They can write the tests (test driven dev) or UX/behavior (behavior driven dev) up front so the outsourced team knows the objectives. Or you can just have the QA team QAing as the deliverables come in (which they will every 2–3 months, right?). Regardless, your super-awesome in-house QA team should be involved on day 1 and testing with each New Zealand.
Of course, your project team also needs a project manager and a masseuse and a dev or two and an ML person if you’re into that. Here’s the crucial part — miss this and you‘ve fumbled the ball — give this team full authority over the project — authority to terminate, change, sacrifice virgins, etc. Don’t let authority rest with some manager higher up who can be sweet-talked with jargon on PowerPoints.
“Front End Technology (Angular2) has been a challenge for us to deliver.”
Angular 2? Yeah, you got the c-team. You know the stable version of Angular 2 is built on less-than-stable tech, right? I mean … no one caught this?
“Much of the testing that was performed was “happy path” testing”
Well, you outsourced QA to these clowns and they drove around in a QA clown car … what did you expect?
“Accenture’s developers wrote the code for the customer-facing ecommerce website in a way that created serious security vulnerabilities and performance problems…”
“serious security vulnerabilities” — probably not true. Just lawyers lawyering.
“performance problems” — probably but you could say that with any v1.0. You could say that about Twitter today.
“As Accenture’s project leaders acknowledged, Accenture ‘spent a good deal of time fighting through integration of RAPID’ into Hertz’s environment.”
There’s a lot of evidence that Hertz got a c-team from Accenture. Angular 2 is another tip-off. Expensing a pallet of duct tape and 40 tubes of crazy glue is another.
Ok the last bit didn’t happen but you get the point. This was not an a-team or even a b-team.
So here’s what happened, probably:
1. Hertz has a dumpster fire of a backend and an integration layer. Oh, it all works, but it’s all duct-tape and bubblegum. I’ve seen it many times. Hertz’s IT/digital transformation people either don’t know and won’t say (really, what are they going to say — that their backend is a disaster that will cost millions just to get in order? Millions just to understand?). You start with mediocre code, add patches and random features over a decade, and then you call up Accenture and ask them to change everything. Again, I’ve seen it many times. It’s a ticking timebomb of duct tape.
2. Accenture sales people have no clue what Hertz is sitting on. Sure, they do a technical review, but either (1) it wasn’t deep enough to uncover the awaiting disaster or (2) the tech people aren’t going to be honest with a Vice President of PowerPoint and ruin his $800,000 bonus. So Accenture prices the job using average assumptions — this is why they keep asking for more money later. They’re probably going to lose money on this. I’ve had situations where we were 6 months into a 5-year project and offered the client $50M just to walk away … that’s how much duct tape we discovered.
3. Hertz hands Accenture the entire disaster with little understanding or internal capability. Accenture assigns a c-team. Hilarity ensues. (Note: it’s possible Accenture assigned a c-team because they thought “oh it’s just a website,” but in my experience, most people in upper-management don’t really know the varying quality of their teams. So, really, it’s a lottery. Further, sales usually hands these projects off to a project group that assigns a project manager who may or may not get to pick his/her team. So whether the client gets a good team or not is often lotteries all the way down.)
Essentially, this is the lawsuit:
I mostly blame Hertz for outsourcing what shouldn’t have been outsourced: responsibility.
Also, Hertz, here’s what you should have done: publicly announce the project/specs and that you’ll be in Austin for a week. You’ll get a dozen meetings that week with teams who will and can do a much better job for less than $32M. In other words, there are many small and mid-sized shops that are great … I have no idea why you’d employ Accenture for something like this.
Tragi-comic update. After this post went up, I received a recruiting email from Accenture. The genuinely funny part is the header warning from Protonmail that I maybe see twice per year ….
 The suit: https://regmedia.co.uk/2019/04/23/hertz-accenture-website.pdf. FYI I’m guessing what happened from experience. I don’t have Hertz moles. Or Accenture moles.
About Nathan Allen
Founder of Xio Research (A.I.), Applied Magic (A.I.), and Andover (data). A.I. strategy and development leader at IBM. Academic training is in intellectual history; his most recent book, Weapon of Choice, examines the creation of American identity and modern Western power. Don’t get too excited, Weapon of Choice isn’t about wars but rather more about the seeming ex nihilo development of individual agency … which doesn’t really seem sexy until you consider that individual agency covers everything from voting rights to the cash in your wallet to the reason mass communication even makes sense…. Lectures on historical aspects of media, privacy/law, and power structures (mostly). Previous book: Arsonist.