I was travelling recently, and while waiting at the gate, I had the opportunity to observe one of those Best Buy vending machines. You’ve probably seen the ones I’m talking about in airports. They sell iPods, tablets, headphones, portable games, cell phones, and other electronic accessories. Some items like iPods or tablets are cost hundreds of dollars while more basic accesories cost anywhere from $20 to $100.

Over the course of almost four hours, I saw about 10 to 15 people stop and interact with the touch sreen interface on the kiosk (which is probably a euphemism for “vending machine” that doesn’t have the connotation associated with $0.50 candy bars). Nearly everyone who walked past it took a hard look without breaking stride. I even saw one gentleman purchase an item, although I wasn’t able to spot whether it was a pricier gadget or a cheaper item.

There are probably lots of to be said about how Best Buy is hypothesizing that consumer’s willingness to spend large amounts of money sight-unseen for electronic items over the web has helped them get over the hurdle of purchasing the same items from a web-like touch screen interface for items that they’re able to see behind glass and immediatley enjoy. But, the real thing that I noticed is that people like to browse (see this article from the HBR blog for some discussion of the importance of touching products in the buying experience). Especially when products are presented in a visually interesting way as they are in the slick and modern kiosk. If people like to browse, then I think it’s pretty reasonable to surmise that people also like to shop more generally. In other words, people enjoy the experience of buying things (obvs).

How has the web stacked up in providing a satisfying browsing and buying experience? To answer my own question, I’d say we’re definitley not there yet. The web was and still largely is, a means of accessing information quickly and efficiently. Think about Google, with it’s austere homepage and its promise of passing you along to exactly what you’re looking for in mere seconds. Browsing, however, is by definition not quick or efficient. What we’re talking about here is product discovery. The modern ecommerce experience is built using the same guiding principles of the early world wide web. Modern ecommerce says “tell me what you want and I’ll get it for you.” It doesn’t say “come on in and take a look.”

Think about the experience of buying on Amazon, often considered the gold standrd of ecommerce. Let’s say you want a coffee maker. You arrive on the Amazon homepage and have a couple of ways of getting to your final product. Immediately searching for “coffee machines” is probably what most people do. Once you get a list of coffee makers, one might filter by desired attributes like color, price, size, etc. Then your task becomes one of feature/price comparison (figuring out what features are important to me such that I’m willing to spend some amount). Or, instead of searching, you could have drilled down through Amazon’s category hierarchy – home and garden > kitchen > coffe > machines. You’d arrive at pretty much the same list of products and switch into feature/price comparison. This is all well and good, but what if you didn’t know you wanted a coffe machine. This is really only the bottom half of the purchase funnel. What’s missing is the demand generation stage.

Is Amazon really capable of creating that moment where you think to yourself how old your current coffee maker is, and how a new one would really brighten up your morning? Maybe if it there was a coffe maker ad on the homepage. I’d assert, though, that by and large Amazon and most ecommerce site can’t create that moment. There are too many links on the homepage (by definition, if you landed deeper into the site you’re already past the browsing stage of the purchase funnel). If you’re on the homepage, how do you know which link to choose? Can you casually get high-level product information with little effort? No, you’d have to drill down, or search, or filter and all largely without context.

Maybe it’s not just a failure of Amazon, or any specific site, but a failure of the web more broadly to support a more casual browsing experience. Therein, I believe, lies an opportunity for a next generation of retail experience that sells virtually fulfilled goods like music or apps. By combining the best of both the brick-and-mortar and ecommerce worlds, a new retailer could follow the Apple store model by creating tangible brand experiences that support casual browsing, personal recommendations, try before you buy, and digital fulfillment.

It could look something like setting up a wedding registry at a homegoods store, where customers electronically register accounts and set certain preferences, and then walk around the store to touch and feel the products, all the while adding them to their account via barcode scanning devices.

By creating a physical shopping experience for something virtual, you would model analgous shopping behavior that is familiar to customer (clothes shopping, grocery shopping, old school cd and record shopping). And you would be tapping into proven locla high-traffic shopping areas (important since people spend a large protion of their income within a small distance from their home).

As a quick example, think about a kiosk that sells MP3s – where you could sample, buy, and download directly to your phone – instead of an airport kiosk that sells electronics or batteries.

No good idea is original. After I wrote this post and while I was still editing, TechCrunch ran a story aon a brick-and-mortar app store. In the story, Openspace seems to follow my line of thinking as they’re tackling the discovery and browsing problems inherent in today’s small screen app stores.

I do a lot of work in Excel. Call me a nerd or whatever, but I enjoy it.

Everytime I’m faced with an Excel task (and most tasks actually), I veiw it is a brain-teaser. I ask myself, “How can I tackle this problem such that I have the least chance of making an error in the shortest amount of time possible within the given output/analysis constraints.” Thanks to some great Excel teachers at NERA, I’ve built a great toolbox of methods to use.

One of my main responsibilities at work is financial and operational planning which comes into focus around the end of the calendar year. This is when we put our heads together to try to figure out what we think next year will look like. The process requires many iterations, as strategies, storylines, assumptions, expenses, and tactics are fleshed out. Once, we have a good baseline forecast we run it through various scenarios to check out the “what ifs.” Easily jumping between scenarios, changing scenario assumptions, and easily comparing outputs is a critical step in the process and makes for an interesting Excel challenge.

The solution I use revolves around the Data Validation feature which allows you to create a dropdown menu in a cell. This post assumes you have an intermediate level of Excel knowledge.

#tldr

Problem: Use Excel to view the total sales from three predetermined 12 month sales scenarios

Solution:

1. Create an in-cell drop down menu using Data->Data Validation to select the name of a scenario
2. Use the text value selected by the dropdown as the lookup value in a match() function to return that scneario’s row number in a source table
3. Pass that row number to an hlookup() function that uses the month as its lookup value to pull the sales number for a given month and scenario

Download example Excel file

The Problem

Let’s say you have three simple sales scenarios, called Plan A, Plan B, and Plan C. Let’s also say that these scenarios each represent the next 12 months of monthly sales under different assupmtions and you already have a model that uses sales figures as inputs to create some outputs (say variable costs, revenue, or cash). Your task is to see how the model’s outputs change with each different sales scenrio.

I’ve set up an simple example book that has a sheet called “Sales scenarios” with a table of three rows representing our scnearios and columns representing each month of the year.

Selecting the Inputs

As I mentioned above, the solution here revolves around an in-cell dropdown menu created using Excel’s Data Validation feature. Let’s create a new sheet in our model and call it “Inputs and summary”. We’ll use this sheet to select a scenario and quickly view a summary of outputs from the model.

In another sheet, called “Sales scenarios”, I created a simple table with the names of the three scenarios. The columns are months, and the rows are the different scenarios. Each cell in the table has a sales number corresponding to a certain scenario and month.

Back on the inputs sheet, let’s create a list of scenario names. The scenario names in this list must be exactly the same as the ones used in the Sales scenarios sheet to lable the table rows.

To create the dropdown, select an empty cell and go to Data->Data Validation->Data Validation on the Excel ribbon. In the resulting dialog box, under “Allow:”, select “List.” For the “Source:”, select the range that contains the scenario names on the “Inputs and summary” sheet. Make sure the checkbox on the right that says In-Cell  dropdown is checked and hit OK. The cell should now have a dropdown menu showing the three scneario names when selected. We’ll use the value of the dropdown in the next section to grab the correct sales figures.

Grabbing the Scenario Data

Grabbing the data will involve using the text in the dropdown menu we created above to dynamically lookup the values from our “Sales scenarios” sheet. Depending on the complexity of the situation, this might involve an index() function, but the way our example is set up, a simple hlookup() function will do.

Since both our “Output” and “Sales scenarios” sheets are set up with months as the labeling the columns and the months labels are all in the same Excel row that our table is in a horizontal format. This requires the hlookup() as opposed to the vlookup(). The vlookup() function is used when the index values are arranged in the same Excel column.

Since the dates are common to both tables, we’ll use that as the lookup value. The table array will be the range containing our “Sales scenarios” table. The only tricky part is figuring out the row_index_num, since the row_index_num will be different depending on which scenario is selected in the dropdown menu. To accomodate that, we’ll nest in a match() function. The match function searches a list (arranged as either a single row or column of values) and tells you which position in the list it finds the given input. In our example, we want to know which row in the sales scenarios table our selected scenario resides. If we think of the row labels as a list of values arranged in a column, we can use the match() function to tell us which row our desired sales figures are since the position in the list corresponds to rows in the table.

It’s important to note, the way I’ve set up the tables. The sales table is 4 rows by 12 columns. It’s 4 rows because we have to count the date index row. To have the match() function return the proper row to the hlookup() function, we have to make sure that the table in the match function has the same number of rows as the table used in the hlookup function. In our example, both the sales_scenarios and the scenario_names ranges have 4 rows.

The lookup value for the match() function here is scenario_inputs which a reference to the value selected in the dropdown menu on the Input sheet. When the value in the dropdown changes, the results of the match() function change as well. This will pull a different row from the scenarios table into our output table.

My scenario_names range contains just the first column of the sales scenarios table. This column has the names of the scenarios that match the values in the dropdown input cell.

Outputs

For convience, I like to put a copule key outputs next to the input selector. That way, if someone doesn’t have two monitors or doesn’t want to open two windows of the same book, they can view the inputs and the outputs together in one place without having to bounce between sheets.

In my example, the output of the model is simply the total sales for 12 months. In the example workbook, you can see that by changing the dropdown menu, the total sales output cell on the same sheet will display the appropriate number for that scenario.

Conclusion

Feel free to leave questions and suggestions in the comments. What other tricky Excel situations have people run into?

I love checking out creative office spaces. Office Snapshots is a pretty cool blog showing the work spaces of startups, creative agencies, and the like that I recently came accross. This Ain’t No Disco is another good blog with the same theme.

Office Snapshots had a nice comprehensive gallery of shots from Weiden + Kennedy. I used to be pretty obsessed with working there.

Meet me in the basket.

Time to start a new blog.

I’ve had a good run on my personal site, The Consortium of Greater Minds. And it’s not over; posts here will be crossed-posted there. But, this is an excersise in personal branding.

After reading a post by Mark Suster on his blog on getting out there, hitting the streets, using some shoe-leather, meeting some people, etc. I decided I needed to do just that. Initially, I created the Consortium of Greater Minds to keep up with my close group of globally dispersed friends outside of the noise of Facebook. It was a place to pontificate on music, the street culture scene, and just update my friends once and awhile on what I was thinking. In contrast, My 50 Coffees will be a place where I can write about more prfoessional interests. I envision it as somewhat of a portfolio that I can pass along with my resume or share it with new friends and colleagues to add personal and professional depth to my web presence.

And, I’ll also keep track of my progress towards having 50 coffee/lunch/dinner/drinks meetings. It’ll keep me both motivated and honest in the journey to broadening my network in the relatively small scene that is LA Tech.

I invite you to check out my LinkedIn profile. If you think I can help or want to chat, drop me a line and we can have coffee.

http://youtu.be/9dZW1C3neao

The LP, produced by Russel Simons, Vincent F Bell, Kurtis Blow, and a than you credit for Rick Rubin, only has a couple of electro R&B tracks. The rest of the songs are in a late 60s sweet soul style, which is kind of odd next to the early gangsta aesthetic of In the Rain.

In addition to a couple of soul LPs, I picked up a bunch of 80s rap 12″s. Some record stores under price the early rap stuff, especially out here on the west coast. Dealers out on the east coast all run in the same collector/record show circuit and can figure out what’s collectable. Picked up a couple of Freddy Fresh 2 star joints (see Rap Records), including this one from the LA Dream Team.

Espresso + ground beans for the week, roasted yesterday.

Trying to track down the broseph.

Without further adieu, los links.

old critisized SFJ post: http://www.newyorker.com/arts/critics/musical/2007/10/22/071022crmu_music_frerejones

the problem with rap critics: http://wearerespectablenegroes.blogspot.com/2009/10/problem-with-these-kids-rap-critics.html

das racist response to sfj post: http://flavorwire.com/45316/das-racist-to-sasha-frere-jones-stop-trying-to-kill-rap

og sfj post about the death of rap: http://www.newyorker.com/arts/critics/musical/2009/10/26/091026crmu_music_frerejones

more sfj criticism: http://gordongartrelle.blogspot.com/2009/10/long-ass-email-about-sasha-frere-jones.html

nyt background on das racist: http://www.nytimes.com/2009/07/24/arts/music/24das.html?_r=1

coverage on white rappers: http://www.illdoctrine.com/2008/07/how_to_tell_people_they_sound.html

response to das racist response: http://www.washingtoncitypaper.com/blogs/artsdesk/general/2009/10/23/das-racist-goes-after-sasha-frere-jones-for-being-white-n-educated/

Untitled Rough Mix