Suggestion for nice earning website

14ned

This morning I was switching over my electricity provider as I do every twelve months because for some daft reason Irish electricity companies actively hate and punish loyalty. With over a quarter of my annual bill saved it's worth doing.

Yet everything I did this morning is totally scriptable: (i) find the lowest provider on three price comparison websites (ii) enter my details, MPN and bank account (iii) set reminder in Google calendar to do this again in twelve months from now when the new 12 month contract expires.

This got me thinking: why isn't there a website which automatically checks every twelve months for the best package for me and automatically changes me over to it, no intervention from me involved?

That website could charge me 10% of what it saved me. For me this year that would be about €20. Run that up to say 5,000 people, each paying €20 a year and that's a nice €100k earner per year.

There is a fair bit of work in implementing such a website, probably three or four months of full time work. There is also several weeks per year maintaining it with new tag parsers as the price comparison sites change their output. You'd need to set yourself up as a Direct Debit capable company so you can charge your 10%, and the test suite and one way encrypting people's bank details would all be a fair bit of work.

On the other hand, potentially €100k upwards per year revenues for only a few weeks per year of maintenance work. If well executed by someone competent, could be a nice little earner for them, at least until the market shifts and electricity providers stop being so stupid.

Would anyone be interested in implementing it? For the record, I want no involvement in the implementation, I would just like to not have to waste a pre-work morning every twelve months.

Niall

Buford T Justice

Not a bad idea. One of the biggest pains I'd imagine is the providers changing their web pages.

Unless they provide a api?

14ned

Buford T Justice wrote: »

Not a bad idea. One of the biggest pains I'd imagine is the providers changing their web pages.

Unless they provide a api?

If you do the provider change booking through the price comparison site they'll earn their commission, so I'd imagine they'd be more than happy to make an API available to you.

If not, then good old fashioned screen scraping is where those three weeks of maintenance per year are likely to go into.

N

colm_c

It's a nice idea alright, although you'd have your work cut out managing fraud, non-payers, switchers, billing addresses etc.

Most likely will turn into a full-time job, which is totally dependant on other businesses (never a sound idea IMO).

Probably be easier/cheaper just to setup an energy company and resell it at the cheapest rate with no frills!

Talisman

14ned wrote: »

If not, then good old fashioned screen scraping is where those three weeks of maintenance per year are likely to go into.

Or you could use the API of the greatest scraper of them all - Google provide an API that allows you to perform target search queries and get the results as JSON.

I used it on a project for a client and shaved €12k off of their expected spend. The cost of using the API was less than $30 and the script took a couple of hours to perfect.

A service like that described in this thread probably wouldn't incur any costs for use of the API because it would require so few requests.

Buford T Justice

Talisman wrote: »

Or you could use the API of the greatest scraper of them all - Google provide an API that allows you to perform target search queries and get the results as JSON.

I used it on a project for a client and shaved €12k off of their expected spend. The cost of using the API was less than $30 and the script took a couple of hours to perfect.

A service like that described in this thread probably wouldn't incur any costs for use of the API because it would require so few requests.

Do they have a specific scraping api?

zig

Would be a handy service, but it strikes me there would be less work/investment in simply doing this manually for clients rather than automating it.

1. No full time work required to build it initially
2. No dependence on scrapers or APIs
3. More focus could go on getting people signed up and spreading the word rather than building the application.

Then come the end of the year if you have 5000 and all their details, you put the head down for a few weeks (or hire some student) and go through all the accounts one by one.

It would make you some sort of broker for brokers!

14ned

colm_c wrote: »

It's a nice idea alright, although you'd have your work cut out managing fraud, non-payers, switchers, billing addresses etc.

As you'd have direct debit authority on the account it's easy to have a script put a hold on your 10% fee. If that fails, send an email to the holder saying that the bank account debit has failed and you'll retry in a week. Retry then for say 26 weeks after which delete the account.

Most likely will turn into a full-time job, which is totally dependant on other businesses (never a sound idea IMO).

Anyone in business is totally dependent on other businesses. But I see what you mean in the sense that say bonkers could simply duplicate your service and make you obsolete.

You could simply aim to sell out your business to one of the price comparison sites? Could still be a nice six figure earner.

Niall

BrokenArrows

Its a good concept.
Not sure of the legality of scraping these sites.

Also how will you automate and guarantee a cheaper rate?

(Below figures are random numbers just to show a point)

Lets say the current customer is on Plan A with the below unit prices
Gas: 1.5p
Elec: 2.5p

Plan B quotes:
Gas: 2.5p
Elec: 1.5p

If they use majority gas to heat their house then Plan A will be cheaper, but if they use Elec then Plan B.

Without knowing what their usage numbers are for the whole year you cannot tell which one will be cheaper for them. And also their usage may change from year to year.

14ned

BrokenArrows wrote: »

Its a good concept.
Not sure of the legality of scraping these sites.

Also how will you automate and guarantee a cheaper rate?

(Below figures are random numbers just to show a point)

Lets say the current customer is on Plan A with the below unit prices
Gas: 1.5p
Elec: 2.5p

Plan B quotes:
Gas: 2.5p
Elec: 1.5p

If they use majority gas to heat their house then Plan A will be cheaper, but if they use Elec then Plan B.

Without knowing what their usage numbers are for the whole year you cannot tell which one will be cheaper for them. And also their usage may change from year to year.

It's a pretty straightforward constrained optimisation problem. The more advanced solvers let you build in some probabilistic latitude such that you can hit 95% of maximum savings 95% of the time unless the customer does something deviating from historical trend like not live in their house for most of the year suddenly.

Niall

beauf

Seems to me that many of these providers deliberately have different bundling and discounts to make direct comparisons, very difficult. The problem will be maintaining the transformation rules between disparate plan/billing structures, in a way that isn't labour intensive.

Other than being an automatic renewal what would the difference be to bonkers.ie or switcher.ie?

At what point would you stop being a broker and start being an energy supplier?

Looking at the above sites they both reference being regulated by the commission for energy regulation.