Good data before #bigdata – how to fix a process when the data is broken?

Sterling per square feet per annum or Euros per square meter per month? Not too complex a question if you have a few properties outside UK in mainland Europe and you try to com-pile rental levels to perform simple comparison between your properties in different markets.

Add to that consolidating PnL’s and forecasts from various underlying vehicles from seem-ingly disparate source systems and service providers. And the answer is… Yes, all manually in Excel!

All the rent rolls are the same – or are they?

I have been reviewing a lot of rent rolls recently – from Norway, Sweden, Germany, Nether-lands and UK. What so striking is that basically they all contain the same information – a list of lease agreements, units and tenants (including vacant ones) – but are structured differ-ently every time I encounter a new spreadsheet.

This is compelling especially for cross-border portfolios, where asset managers need to con-solidate basic figures from various sources – with different currencies and underlying busi-ness conventions, like ”Is the rent expressed as monthly or annual?”, ”Are these square me-ters or feet?”, ”Is this net or gross lease?”, etc. This is why Excel has been and still is #1 tool out there in the market. It’s all a manual process.

Adding to that is poor data quality, where local service providers have been desperately try-ing to parse information together to fulfill client’s Excel template.

This leads to a lot of frus-tration and wasted time. I wonder where valuations of bigger portfolios are really based upon… maybe all the manual flaws and discrepancies evaporate into statistical noise?

Also a no brainer is the simple ruling for quality check. The other day, I spoke with an asset manager of a big CRE investment corporation who had been investing a pile of cash to cre-ate a system out of scratch for basic data validation.

Once run a test against this validator he said that he identified 700 errors in his own rent roll data set! And he openly confessed that he thought that his own data was intact. Quite an eye opener.

Can you fix a process if the data is broken?

The answer is obvious.

So should we get our act together and have first good data before rushing into big data?

Since it is such a common problem in the industry, we have decided to do something about it. I am a true believer of data consolidation tools.

I also encompass Standard APIs to make the data seamlessly flow between systems and apps – not to mention the substantial saving of time and manual effort!

Hannu Rantanen is CEO of Assetti.

Add a Comment

Your email address will not be published. Required fields are marked *