Hi
For a new venture I've bought 10m optin email addresses but they are split into a load of septate files. I need to somehow add them all together and sort by postcode so I can geo target my campaigns. I've spent ages looking but can't seem to find a way to combine these into some sort of CRM database. Does anyone have any ideas or recommendations??
what format are the files in?
Are the fields consistent?
If the data is consistent, you can pretty much use one of the email services as Mailchimp, Aweber, Ubivox etc? They all offer importing of large lists - however at 10M you might be in for a rather huge bill..
Hi
They are CSV and consistent. There are a few files which are a bit different but most are the same.
I did look at mailchimp etc but its too expensive. Been looking at SugarCRM to see if i can upload them there and then sort as I need a CRM to feed leads into to get qualified, signed up, visited and contracts sent to a provider.. New to this stuff though.
BTW planning on sending a few emails with sendblaster to begin and then get a pro mailing software set up on a dedicated server with quite a few IP addresses.
You might be better off sorting them before hand as that is a very big list.
SQL Server express (free edition) is worth a look if you have time to play around - http://www.microsoft.com/en-us/download/details.aspx?id=1695 You will be able to import the csv files to a table, query that table as required and then export it into any format you need
I am sure there are lots of other free options out there too though.
SugarCRM is not what I was hoping... but might be ok to manage leads through
Thanks saint, I will take a look at that. I'm thinking a simple DB is the answer now but there are also a lot of spam traps i need to clean out as well.
oook then. got them imported into access :) now just trying to remember how to run reports from my school days..
If you've got Excel 2010 then this should do the job :) - http://www.microsoft.com/en-us/bi/powerpivot.aspx
/me is very interested in how this pans out...
Downloading that now Chunkford, thanks. I ended up unselecting "all" in the counties column and only selecting the desired county, then I just ctrl+c ctrl+v into excel and saved as csv :D Now imported the 1st 20k records into Sendblaster to test reactions.
Ah another thing, I use CMD to combine all the 1st batch of csv into one.
Thanks again Saint, without your post I wouldn't have tried Access again and would still be trying to figure this out.
What part are you most interested in JasonD??
> What part are you most interested in JasonD??
Everything, but most of all where you have pain or things didn't go well.
Identify the pain, build the painkiller (and yes, there may be something in the works... )
I will keep you updated. Without the pain it normally means for me I've gotten lazy or I'm not trying tbh. There's 12-24 months of very high liquidity in a market I'm already somewhat familiar with so I'm hoping I develop some serious stretchmarks.
I've actually done this before, taking flat files and putting them into a site. I did it a few years back and it was a b*tch. I think I used a script called xeoport to dump them all into a php database though. Then I had to make my own content management system to deal with them.
Great way to learn PHP and regex.
Heather, I think you meant to post in another thread ?
First off, why on earth did you buy 10 million records?
We are using VoloMP for managing our mail and a few ESP's for delivery based on whether it is bulk mailing or premium.
It's cheap. I bought 3m a few months ago and had very good results. The data does need cleaned as there are a lot of dead or weird numbers in it.
This data is useless for normal email marketing but for a previous offer similar to what we are now offering, we had fantastic results.
It'd be interesting to know how you're cleaning them, if you think it warrants a new thread.
Re BoL's point on cleaning.
Are you aware of the inbox ratio difference when the list is clean versus dirty?
If the SenderScore is between 40 - 49% then you will inbox at around 6.7%
whereas if the SenderScore is between 90 - 100% then you will inbox at around 99.4%
There may be a project coming soon that makes sure your ratio is in the 90s + and also gets a much higher overall convrerstion ratio
So higher conversions on a much higher inbox ratio is a win all round ;)
Cleaning in 2 ways. There are a load of weird tld so I removed them. Must be a few hundred thousand.. Then I use http://www.glocksoft.com/email-verifier/ just to see if the email address is active... I can't do too many at a time because my ip gets banned, and it doesn't give me any gmail stats so I just send and pray to gmail. It basically just takes out the hotmails/yahoo and a few others who return "mail undeliverable" but without sending an email.
Is there a better way??
I havent tested inbox ratio, my main concern is getting into IP or banned domain terrortories because 10-15%+ emails are to dead addresses. I know hitting inbox is key to massive conversions but to be honest, right now the aim is to find a few clean streets to send a survey team.
Although I have 10m I'm only sending 5-20k ish per time for now until we have more infrastructure in place.
How to do it:)
Get 10K clean ones and put on dedicated ip then send from nice domain and site with interesting email
Take rubbish ones and send from lots of shared ip's to a different domain
Remove all bounces
Now have semi clean list
Run semi clean list on different shared ip's
Remove all spam reports etc
Now you have an almost clean list
Drop 1K lost into the clean dedicated pot
Keep going as the clean list gets better you can add more to new send
Takes time, money and friendly email companies
Doug
> 10 million
> load of weird tld
this isn't the cleanest or highest quality list you've bought is it?
Thanks doug.
Jason, some of the data is from phone leads. When that's the case they put phonenumber@leadgencompany.com. Clean there should be around 6-7m if I apply the current rate to the whole list. But like I've said I've had very good success previously offering a similar product with limited appeal.