How Many People Work in Your Internet Department? 255
WorkinTooHard asks: "Two years ago, I took the job of Internet Marketing Manager for a international company, with a crazy idea that I could convince senior management that the internet wasn't a fad. The only problem was that I didn't expect a (respected) mid-level manager to be the road block. We are in the middle of a major website redesign (the current site has not been updated in over 8 years) and everyone is asking why it takes so long to complete, and almost daily I have to explain that I do not have enough manpower. Of course, I can't prove ROI until the new site is launched (a great Catch22). How many people do you have working in/on your company's Internet/Intranet and Extranet sites and applications? How many full-time web-application developers, content providers, analytics people, UI designers, email marketing people, and so forth?"
"Please note that this includes anyone who works directly in building and maintaining your companies current website, electronic marketing and Internet applications. If you can, include the size of your company, number of employess, the number of active products being sold/supported, and how much outsourcing you do? The company I am currently working for has over 13,000 active products and over 30,000 products which need to be supported. We do no outsourcing, have over 900 employess in North America (over 8000 worldwide) and a total of 2 full time web developers, 1 part time developer/SQL guru and 1 content/data person as well as two people in our MarCom office which periodically write copy."
How about a Demo? (Score:5, Informative)
This is not a DIY project (Score:3, Informative)
Triply so if it's been 8 years.
you're woefully understaffed (Score:2, Informative)
Here are some comparisons from my recent past. Currently, I'm the only tech guy, and I do everything. But that's because I just left my last job to found a start up
Prior to that, I was one of three developers in a department that also had a designer, a writer, and a project manager. That was to service an organization of ~1,100 people. Some departments also had their own techish people who'd do departmental sites and the like. We did no ecommerce at all nor any desktop support, just interweb stuff (we wrote and maintained a fairly sophisticated Struts-based CMS system), and we were stretched way thin -- there was a greater demand for our services than we could reasonably comply with. We were also a non-profit, which meant more people wasn't a realistic option.
Before that I worked for a company with ~30,000 employees worldwide, and while I was there we were just rolling out ecommerce. We had three dedicated developers, a DBA, a network guy, two support people (just for the web site, they didn't do desktops) and two managers. That was some years ago, I believe they've grown since then. This was mostly to maintain a site -- the design and development of it has initially been outsourced. This felt like reasonable manpower, but again, we were doing incremental change on a project that had been built by a larger team. And the pace was, shall we say, bureaucratic.
It's possible to educate managers about what resources are required for a given volume of work, but you'll have to communicate well and be direct when you know you're right. Good luck.
Totally Dependant (Score:3, Informative)
Basically, it comes down to:
1. Understanding of the final product or content
2. How much you interconnect with backend data providers, and if you require filtering.
3. Your team's experience in the language and dev environment
4. The speed at which those languages lend towards the final development.
Note about languages -
My experience is that Java is by FAR a slower dev environment than PHP, Perl or Ruby. The whole compile cycle and the complexities of app servers make for a much more complicated project. The exception to this is JSP - which comes closer to Perl - but entails it's own complexities in getting at databases, etc... Plus, java makes no wins in uptime, speed, or clustering compared to Perl (utilizing mod-perl), or PHP.
Yes, I have been on the large person java team that architected the connections between the three largest online travel providers - don't whine at me.
Re:Push Back (Score:5, Informative)
"I can't tell you when it will be finished until we know exactly what we're building. Help us nail down the specifications, and I'll be able to give you a finish date with the current staff and workload."
"There is a finite amount of manpower available to do this work, and the schedule I gave you is firm, unless we either add people -- and that won't be a one-to-one improvement, it will depend on how much work can be run in parallel -- or reduce scope. Help us do that, and I'll be able to get you a new completion date."
"I know you want these changes in the initial launch, and I want to give you these changes in the initial launch. However, there will be some impact to how long it will take, because a lot of work we've already completed will need to be redone. Help us nail down the new specifications, and I'll be able to give you a finish date with the current staff and workload."
Repeat ad nauseum until the project is finished.
Re:Push Back (Score:5, Informative)
Two cases in point:
1) My company (3 people) was hired to redesign the corporate website (twice) and build the entire employee intranet for a $300m/7,000 employee company. This client had a 60+ person web team in the corporate division alone, yet had to hire out to a tiny team of crack individuals to actually get anything done.
2) I'm working with another client now - smaller but much older - that would much rather have us (as the consulting firm) hire and manage the people we need for the project and pass the cost on to them (plus a markup) than hire internally.
I don't understand the accounting side enough to know what the benefits are there, but from a management perspective, it's very nice to be able to make a single "entity" responsible for the project (as kind of a black box) than to have to think about and deal with an internal "team".
Thank you for listening to
Don't do ROI. (Score:3, Informative)
The trick is not to do ROI - instead you do risk analysis of not undertaking the project.
You mention that the current site is 8 years out of date. In your risk analysis, state that prospective clients that view the website will see that information is outdated and will look elsewhere. This qualifies as a cataclysmic severity since it means no inbound customers (as they are more attracted to some webpage that is moderatly up to date.)
Just remember one change you have to make in your risk analysis: s/risk/certainty/g.
If the manager insists on ROI, head to the advertising department and ask them for their figures. As you know, a website is merely another way of advertising, and no advertising means no business - in fact, advertsing may give you advise on working around your roadblocks as necessary (or otherwise work on your behalf.)
Re:ROI? That's impossible... (Score:3, Informative)
Do I have some customers? Did they find me becasue of my sign? there is your base for an ROI.
Of course, no one did it because it was pretty obvious you needed a sign.
For a web site, yiour ROI is generated traffic, as well as removing costs from other places of business. For example, can a support web site save you money in man power? telco costs? Does it have a marketing value? Does it have a sale value?
That like saying yu ca't get some ROI number friom a tv commercial, which you can.
Re:Push Back (Score:5, Informative)
They're called "recurring costs", and hiring outside consultants don't generate them while hiring employees do. Let's say you have to accomplish some project, and it's going to require roughly 1600 man hours to complete (three programmers working for three months, eight hour days). You don't have three programmers with enough time to dedicate to this project. So, you have two choices:
a) hire three guys
b) hire an outside firm to do the work
Finding and hiring three qualified programmers is going to take longer than finding one qualified firm, but even if it wasn't it's still cheaper to go with route b in this instance. Three programmers are going to cost you about $250,000 a year, give or take $50,000 depending on your market. So even if you spend $100 per hour on the outside firm, you're coming out ahead at only $160,000. Not to mention, once this project is done, if you go with route a, you're now stuck with three more employees, for whom you have to find something to do or else they're just going to get disgruntled and spend all day posting on Slashdot.
Re:ROI? That's impossible... (Score:2, Informative)
I think your understanding of ROI is a little narrow....Yes, in many cases it boils down to who bought your product in the long run but it's not always that simple in the short term and reporting aspects. What you track to prove ROI depends on the company and for a website it's definately not always "customers bought an average of 500 more [PRODUCT] on our site per month since we launched the new site design."
For example, on a new car dealership site you usually don't expect to sell the car online. You use it to get qualified leads. Get the prospective customers information and decide if they're worth the manhours it would take to close the sale in person. Schedule test drives and leave it up to the sales guy at the dealership to close the sale. He'll spend less time talking to people who aren't really serious and (if everything works properly) end up with more commissions and make more money for the dealership. The number of people referred from the site is something you CAN track as well as the percentage of them ended up being successful sales. Compare that to the number of people who wander in off the street and what percentage of them are successful sales and you have your proof that the site was worth the $$ spent.
Re:lone wolf at my state agency (Score:5, Informative)
I call bullshit. It's incredibly difficult to fire state employees, and given that there's documented evidence of a similar organization taking 8 times the resources to complete the same project, there's no way you can get fired for this from a large corporation, let alone the state.
If your manager has literally threatened you with termination over this, stop working on the project and go directly to your HR department, do not pass GO. Tell them about your stress (it helps if you have a doctor's note or, even better, a note from Epstein's muttah, stating that you're under immense stress and borderline to a breakdown) and make sure they know you've been threatened with termination for not doing what four people couldn't. You'll be surprised at how fast they move to make sure you're taken care of.
Why Was He Given A "Manager" Role? (Score:3, Informative)
Your Sim is miserable in his apartment. He has a crappy TV that doesn't make him very happy. You are confident that a big plasma TV would make him much happier and thus he'd work harder, get promoted, and earn far more than the cost of the TV back.
Sensible person's approach: Earn what you can. Upgrade the TV for a slightly better one. Earn faster with the better TV. Upgrade once you can. Repeat until you have the plasma TV.
Idiot's approach: Sell the old TV. Afterall, it sucks. Go to buy a plasma. Discover you don't have the money to buy the plasma and now you can't even afford to buy back the crappy TV. Watch your Sim get depressed, skip work, and get demoted, earning less.
Yes, a wholly new website would be spiffy and, quite possibly, keen too. But there's not enough money for a wholly new website that proves the shining vision of the net as somewhere to invest.
A sensible manager - or a ten year old kid who's played The Sims - will thus not try and gut the whole damn thing. They'll take a week, a month, whatever period they figure they can handle and do the best job they can on one small part. They'll then use that success to argue for better resources to work on a bigger part next. In time they'll have proved repeatedly that money spent on them is a good investment, be doing large chunks at a time, with a well resourced team - and all those small parts will add up anyway.
If I walked up to my CEO and said, "I need 8 server side engineers, 4 html guys, 3 artists, 2 content writers, 2 sys admins, 3 db guys and two secretaries, can I have a $2.4m budget please?" I would damn well expect to get laughed out of his office.
On the other hand, if I took myself and the one other guy I had and proved a $100,000/year return on investment from our first $25,000 of work, I'd expect a much easier time of justifying an extra head count to hire on maybe an artist. The next project spends say $40,000 and makes $150,000 extra a year and I can likely get a dba. Repeat enough times, consistently giving the company more than it pays and I'll get my 24 person team. Ask for all 24 of them all at once and I should absolutely expect to get laughed out of the office and then fired for being an utterly clueless manager.