Sunday, August 7, 2011

Moved to blogger.com

As I'm always striving for perfection, I'm testing out blogger.com as a substitute for wordpress. The export/import worked fine. Now to check consistency, layout, differences in UI and what didn't get transferred.

Monday, July 18, 2011

A software engineering approach to marketing real-estate



  1. Create ad

  2. ?

  3. Profit


Evil log of the Software Engineer advertizer. I've put the log on the top. For a first impression, read the Intro first.

24. June: Bought our new house. Started planning of selling the apartment.

6. July: 3 rounds of interviews with potential real estate brokers. One showed a sub-par interest, like any normal Norwegian :) Second was one we had met a lot during our peeks into the market for the last 5 years. The third was a younger guy with a non-salesguy look. A real outsider, and he gave us a damn good presentation and showed real interest. Also would cost 8.000€ more than the others, but had a more expensive package. The decision was a tie between the well-known and the outsider, but the well-known had the advantage of knowing the area well and knowing that it could be a tough sell and remedies for that. Not to self, if you have some own meanings to how selling your apartment should be and want to give feedback to the process, it's perhaps better to choose the salesperson which you can "mold" :)

15. July:  The taxonomist (?) viewed our apartment. His impression was very positive, but he measured our apartment to be 1 square meter smaller than we were expecting, and estimated it to be 5.000€ less than we'd prefer...

17. July: First writedown of the original ideas and elaboration

18. July: Rented a http://www.lagerboks.no storage-box on a trailer, fetch'n delivery of the box + 3 months storage for 400€. Expensive, but damn good solution. Placed the stuff we don't need in boxes and got 8 cubic meters of stuff shoved away and cleaned out almost everything before the photo-shoot.

20. July: Bought 20€ worth of fruit in green, yellow, orange and red colors to place in strategic places. Ikea for towels, bedspread and pictures of ducks. New pillows, fragrant soap with red color and other stuff from around Oslo. Photo shoot went real well. Removing all the clutter was a big hit, you have NO time to arrange stuff during the shoot. Also had the original prospect of the apartment from when we bout it handy. The pictures have added lots of light to them, so strong bright colors is a must, and preferably as few colors as possible per picture frame.

26. July: The finn.no advertisement is out!!! I believe it looks good. Got just under 2.000 impressions on finn.no on the first day which is quite good compared to the market when most are on vacation. Similar apartments have ~1500 impressions after 2-6 days. Did some viral marketing on Facebook, Google Plus and MSN. Will save Twitter till later. Am saving Twitter marketing till later.

27. July: Have created Facebook and Google ads/adwords. Google adwords is for rocketscience advertisers! Damn hard to figure out why nothing is showing! But a lot of knobs and levers to fiddle with. Facebook was real nifty - got 12.000 impressions and 12 klicks in zero time. Needed to pay 4$/click to make it interestin to facebook :) Set down rates now that I can see the thingamajig is working. Am using both CPC and CPM on FB on two different ads. Will change the text and titles to "renew" the ad in the forthcoming days so punters don't get annoyed. Am also afraid to hit the $ ceiling, it looks so horrible that money is pouring out of my account like that. Damn yee marketing leeches! Am recommenting on FB and google+, hoping it attracts more views and publicity. Am going to bed with 2820 impressions on finn.no. Will try to go physical tomorrow.

28. July: Ad campaign running 08:00 3006 shows on Finn.no. Facebook squeezed out the last penny of my 45$/day campaign. At least got me 45 clicks. Google adwords got off to a very slow start and ran away with  5$ for 7 clicks. Can't get the search words quite to work. Feeling that my ad campaign on the net is more for annoying people than anything else :) Have got a rather good feel for the parameters - $/click or view. Tweaking the parameters and seeing the result is like a game engine! No wonder sales people have the incentives model they've got. On my way to work this morning, I wondered what it'll cost to get one of the beggars aligned down the street to hold my ad "Nice 3 room for sale" with the QR code which people can scan with their smartphones. 1€/hour perhaps? :) Don't think they'd dislike a stable hourly income. And although it'd attract negative attention, it'd be quite cool in a sense. Perhaps get some media attention as well.

Had 8 people from 4 groups at a private showing which I held. Fun stuff :) I should market myself as the Honest Real Estate Guy. I bet that it was severely in my disfavor. But they were really interested in me talking. Last ones left after 2 hours. Status of tonight - Accumulated 3699 shows and <100 bookmarks on Finn, which is quite nice. Have annoyed 52,824 punters on facebook over 4.5 times, gotten 76 clicks which has cost me 84$.  Have gotten 4 246 shows on Google Adwords with 30 clicks and cost me 50$. The annoyance/cost favors Facebook until now :)

29. July: Where are people coming from? The paths people take into my finn.no ad may vary widely because of the different channels (FB, Google, QR code cliks, FB personal comments aso) without me knowing. The statistics analysers at finn suck top notch, and FB and google don't give me much info. Recommend to use multiple URL shortener links which gather much statistics from the users. That way, you can differ how they got to your ad.

30. July: Cleaned up several plastic bags of garbage in the vicinity. Your neighbors are always pigs. Especially their children!!! hehe... But a couple of hours work really payed off. Made the approach alot more visually pleasing. Ended up not going totally Gorilla with the QR code advertising. My access card at work is shut off in the weekends, so no printer.

31. July: Show of the apartment - had a lot of visitors. 22 wrote their names on the list - more than twice that of a regular good showing. 5 from private shows I held. 4500 views on finn.no, ~50% better than a normal good viewing. Was extremely lucky with the timing - only 5 other apartments had showings the same day in the surrounding area. There still were some people out and looking even though summer vacation officially wasn't over yet.

1. August: The Sales Round is where I hope using a seasoned real estate broker will pay off, who's given potential buyers confidence during the show and handles the selling round well.

Introduction


I am in the process of selling my beloved 3 room apartment in central Oslo. I have chosen a real estate broker which I believe will do a great job and knows the area well. The marketing package is a straight-forward ad in the only credible real-estate marketplace in Norway, Finn.no, which assumes everyone uses it. I am on a low budget, but wan't as many as possible to know what my apartment has to offer. I can do better!

Problem elaboration


Finn.no is market leader and everyone uses it. Very few use other sources and the brokers know it and don't have incentives to think differently. But that's not quite true. Everyone nowadays start with a Google search, even to find the portal they already knew the address to. And it looks like getting a Google Adwords on search results like "3 room apartment oslo" can't cost too much. Paying for ads feels like cheating in a game of marbles by being the bully of the block who doesn't care for his lunchmoney. So why not apply software ingenuity to manipulate Google search hits by fair and square cheating. You won't find any apartments listed in Google results for Finn.no or the brokers, because they "hide" their pages from Google by requesting that their containing pages aren't to be indexed. What has happened is that searching for apartments bring you into a vast void of non-relevant search results. Lets manipulate:

  1. Create a dummy Apartment Landing Page which in time will contain some of the same info from the upcoming Finn.no ad when it's been written. http://leilighet.kompo.st

  2. Write some more text, add outgoing links to Finn.no and so on.

  3. Make friends and followers notice the landing page and reference it from their channels (Social Networks, Facebook, Google+++)

  4. Check out if Google finds landing page

  5. (...)

  6. Profit!


What about IRL Gorilla marketing in the streets!? My next step is to wallpaper these around in Oslo on random places and venues:

"Great 3 Room Apartment in Oslo"



Oslo will never know what hit'em!

Some statistics


What statistics does Facebook give Advertiserists? 689700 In norway use FB between 18-35 18.6% have a college education 48% are not single 13260 Swedish speaking in norway, 60% live in Oslo

Auto Draft

Monday, November 2, 2009

Open Sourcing the Business Model

IT and Tech have started playing with open cards, exposing their handiwork and knowhow. What will happen if Business open up their inner secrets? Will the company still stay competitive? Will openness attract customers and open new prospects?

Sunday, October 25, 2009

Thursday, June 18, 2009

Simple REST server with Grails

I have recently worked on two small projects that need a simple data store for more intelligent clients. Choosing REST was a no-brainer after some discussion. Grails' HTML CRUD interface is a great template to start off with, but it lacks dealing with XML/JSON.

Sunday, March 22, 2009

Facilitating web page error reporting

Are there currently any products for facilitating reporting errors in web pages?



Problem description


A end user is browsing through your site, and stumbles upon a page which for him has an error or bug. It might be a mother-of-all-stacktraces shining at him, a number which is wrong or single space missing. Often, the end user is the only one who can see the problem, and it is difficult and timeconsuming to report through email or on the phone.
Preferrably, the user can send a screen shot. But it is a difficult process to explain, several manual steps involved, and the state of the application cannot be captured by a screen shot.

Possible solution


A "report bug" link which can easily be placed on every page, that when pressed, creates a report containing
* The offending web page as html
* A dump of all relevant state for the page
* A field where the user can type in what's wrong
* A crude marking device to easily tell where the error is located
The report can then be emailed and fed to an issue reporting system.

Relevant technologies


A javascript library or reusable SiteMesh decorator or similar. Preferrably as little intrusive as possible.

Does this already exist?


Please tell me if it does! :)