The Future Belongs to Robots, But …
Every buzzword-laden bid optimization article sloshing around the search engine marketing blogosphere is utterly fixated on a single subject: automated cost-per-click management via one of the myriad SEM software suites available.
This isn't one of those articles; I'm here to blaspheme and discuss bid management outside the auspices of pricey SaaS platforms touting unique proprietary algorithms forged from a heady combination of bleeding-edge analytics and massive amounts of venture capital.
Let me be clear: I'm a huge believer in software-based automated bid management. There are some truly excellent platforms out there, and for those of us fortunate enough to afford the price of entry, they're a godsend, freeing up a huge chunk of our work week and providing a competitive advantage over the unwashed masses still toiling in spreadsheets.
But the problem is just that: only some of these platforms are excellent when it comes to automated bid management (and the latest Forrester Wave report agrees with me on this). Every platform's reps can talk articulately at length about low-volume keyword clusters, but turn their bid software loose on your accounts and it will be, well, a cluster of an entirely different kind.
Then again, that's only a problem if you're at one of the midmajor to global size outfits where you're even entitled to the regular privilege of free lunch with a side of algorithmic jargon. Some of us are still the scrappy underdog, or least held to overall marketing program ROI goals that make percentage points of fees on media spend untenable.
At my company, we're fortunate enough to have access to bid management software that we frequently leverage and constantly test. When we evaluate potential new platforms in any media channel, we demand evidence of the efficacy of those platforms’ bid management tools. Sometimes, however, there are just some unique cases where for any one of a million reasons, our automated weapon of choice just doesn't work.
That's why I created a simple, Excel-based tool: Bids Are Love! Before I get into the pulse-pounding discussion of spreadsheet mechanics, some brief rationale on why I think this tool can improve on a lot of pre-existing processes.
Put … down … the … buckets!
A common method we've probably all encountered is separating our keywords into different "tiers" or "buckets." The problem here is fairly obvious and easily illustrated by the following example examining a handful of keywords with similar performance, and how a bucketing system would deal with them. Let's say that one of your buckets is keywords with a return on advertising spend (ROAS) between 70 percent and 80 percent, and you have the following three keywords:
- keyword A: 80.1 percent ROAS
- keyword B: 79.9 percent ROAS
- keyword C: 70.1 percent ROAS.
Keywords A and B are virtually identical, but will receive a different bid. Keyword C is much less similar in performance to Keyword B than Keyword A is, but will receive the same bid. The problem with the dividing lines is that the keywords closest to them are dealt with in a very arbitrary manner.
Here's where proponents of this antiquated idea will either say that they just make lots of buckets or, better yet, intelligently taking stock of where there are clusters of keywords with similar performance, define their buckets around those sets. Both of these are just half measures working towards what you really need: a completely gradient method that handles every keyword individually.