Behind The Scenes of Yahoo’s Panama Platform
Yahoo! has been taking a PR pounding lately, both in the press and now famously in the internal Peanut Butter memo. Unforgiving, Wall Street has predictably added insult to injury.
At RKG, we have a very different opinion of Yahoo!. We think the current criticism is undeserved, and we want to take a moment to point out some of the many things Yahoo! is doing right.
The new Yahoo! search-marketing platform, code-named Panama, is fantastic from our perspective. Yahoo! has pulled off a Herculean feat, from an engineering and management point of view, and the’ve done it right. RKG is an early adopter, and started working with Panama at the first opportunity (I can’t say exactly when due to NDA issues, but it was very early).
I wrote a great deal of the code to integrate RKG’s systems with Panama. I’ve coded against every other major platform out there, and most of the small ones too. Panama’s API is unsurpassed, and is even better than some of its major rivals. It’s just built right. As an engineer, there is nothing I want that it doesn’t provide. We have already migrated some clients to it, and we’re watching it like a hawk. We have not had a single hitch — not a single one.
That’s not all, though. In addition to planning and delivering a huge system that appears flawless at this point, Yahoo! very nearly hit their delivery date. Oh, I know there was a small “schedule slip,” and I know Wall Street punished Yahoo! for it, but from our point of view, the delay was a concession to advertisers panicked about heading into the holiday shopping season on an untested software platform (if you’re a retailer, you know what I mean).
Plus, it was a tiny delay. I forget now — was it a month, six weeks, a few months? Give them a break. Windows Vista is how late now, and with what fraction of the originally announced functionality? (Hint: very late, small fraction). How late was Windows 2000, and Windows XP, and how much promised functionality got axed each time? In comparison with most software engineering projects, Yahoo!’s schedule slip is non-existent, period.
This leads into my next point. Yahoo! delivered more than they promised with Panama. I know because when we wanted things that weren’t in the specs, we asked, and they delivered. Once we asked for a major change to the reporting functionality. They could have said “no, the schedule is more important.” Instead, they said “it’s more important to deliver the system you need, than the one we want you to have.”
It wasn’t just that they made the change, either — it’s how fast they did it. We asked for this over email, and the next thing I knew our phones were ringing — and there were a bunch of key engineers and managers conferenced in on the other end, on the chance that we might have time to talk to them. We did. I could practically hear their pencils scribbling notes. They asked for use cases and business justification. We explained how the change would be a huge win for us, enabling us to do more work with far fewer API calls. Not every agency is smart enough to optimize API usage the way we do, and most wouldn’t understand or take advantage of the feature we were requesting, but Yahoo! knew it was important to us. Yahoo’s people are smart, responsive, friendly, and respectful. That’s not something we take for granted.
I’m a huge fan of the whole experience with Panama, from the invitation to be an early adopter, through coding and rollout.
Don’t write Yahoo! off yet. Look behind the hype and the short term, and see how they’re positioned. They’re a strong company with great people and great technology, and we think they have a great future.
– An RKG Engineer