Good afternoon. I'm in Toronto today and in this episode of conversations with Kip, I wanted to talk about a metric engine. first think of a search engine as a comparison to a metric engine. In a search engine, you can find a lot of narrative textual data, you can often find the answers that you want, after searching many different times, perhaps changing your search terms, but you can find a lot of useful information. Again, though, it tends to be textual narrative data. When was the last time you entered a number that wasn't a year or an address into a search engine.

We don't do that very often. Our search engines present us with quantitative data. But that quantitative has been pre calculated, it's been pre determined. You can find in a search engine you can find statistical information from Sports from economics from financial All that stuff's available. But it's all been pre calculated. Searching for individual transactions in the search engine is not something that would be very helpful as search engine does not aggregate in any way.

It doesn't do calculations that are required in order to give us a position to show us a balance to calculate some metric that we need. That's where in my opinion, we need something new, something closer to the flexibility and strength of a search engine. But something which gives us the metrics that we require for measuring for understanding our world in a measured way. A metric engine, on the other hand, would give us the flexibility of a search engine but would be able to deal with these issues we've been talking about on quantitative data, what would be the impact of having a metric engine? well as our systems are set up today, with these many different data supply chains, the metrics we find in search engine Results are different. That's the reconciliation problem we've been talking about.

If we can find the answer at all, in many instances, the answer is not produced by today's metric engine such as they are, they leave gaps between the answers that they provide. Their systems were set up to provide particular types of balances in one structured way, and can't provide other balances because the detail is long gone. Their ability to calculate roll forward the balances to produce another set of numbers is impossible. So we have many different search engines, in a sense, that gives us a set of answers and but there's nowhere near the flexibility that you find around narrative or textual data. A search engine that would do this sort of process would have to go through these principles we're talking about, it would have to allow us to select the business events that we're interested in. Whether those be intra Florida flyers that have Frequent Flyer numbers from a competing airline, whether they would be the types of car repairs we've made against a particular model of car that we've owned for a number of years, we'd have to be able to select those transactions first to say what metric do we want to create?

Then the metric engine would also need to be allow us to determine how to aggregate that data, how to create the bounce, what attributes off of those business events would be interesting for us think of the world before the search engine, and how difficult it was to find information. How far reaching a search engine has been impacting our world ability to find information, to process information and use information effectively, textual narrative information revolution might happen in a similar way. If we were to actually build a metric engine, it might change our ability to understand our world to measure our world to understand the impact Have decisions to measure the results of activities in business and otherwise, that might be the impact of having a metric engine.