The Real Estate industry in India's metros is fast-growing and sees a majority of marketing budgets allocated to digital. Search as a channel for marketing real estate is very competitive - especially in Pune, Mumbai and Bengaluru. This meant, though the leads that come via Search are very relevant; the CPAs aren't that competitive. In such a highly competitive market; with bidding for the same generic keywords; getting maximum impression share, better CTR and maintaining average position at lower CPC was quite a challenge. A manual approach on preparing a keyword set, opening up the entire funnel of keywords wasn't going to work. We also devised strategies to bid for long tail keywords, tried multiple ad scheduling techniques; but manually managing the ad-to-keyword relevancy, keeping track of all ad-groups didn't seem feasible.
That's when we decided to build an automation platform on top of Adwords Search, a platform that understands real estate as an industry, knows competition, has a rule engine to generate over 20,000 keywords that are relevant to your project. A platform that could generate appropriate text ads to increase relevancy and improve CTR, could learn directly from performing keywords and optimize the campaign 24x7, 365 days a year.
With these buckets and a good understanding of how keywords were manually created, what people searched (learnt via actual queries and webmasters); we created a rule engine that could generate keywords under these buckets with any given real estate product. With a system now generating possible permutations & combinations, also modifying them into long tail keywords; we could not just reduce the time to set up the keywords but also got a very comprehensive list to begin with.
We then decided to make our media planning for search more algorithmic & data driven. We went back and analyzed our CPAs for the above keyword buckets & individual keywords. We were successful (using historical data) in devising a strategy connecting the available impressions, probable CTRs, our conversion rates on websites, quality of leads (from Sell.Do Sales CRM). We could assign a priority to these buckets and now knew which buckets could drive us leads, at what approximate CPA, what quality & at what velocity (leads per day). Here's what we found as a priority with the CPA increasing as we move right & quality dropping.
Brand Keywords > Primary Micro Market > Secondary Micro Market > Generic Keywords > Guerilla campaignOur algorithm ensured we stuck to only the campaigns that delivered enough velocity at the lowest CPAs possible, while also considering quality of leads.
Kolte-Patil Developers Limited is one of India's leading real estate companies & is a publicly listed company. It's headquartered in Pune, has a legacy of 25 years and are dedicated to creating spaces that are perfectly harmonious with nature, aesthetics, practicality and value. Kolte-Patil is one of Pune's largest developers and has completed over 1 Crore sq. ft. of landmark developments in the city and also in Bengaluru. It is also cementing its presence in Mumbai with new and plush redevelopment projects.
Life Republic is a sustainable township by Kolte Patil Developers that’s built on hundreds of acres of undulating greens. It is designed to achieve one single objective: a meaningful way of life for the thinking mind.
As an example to showcase how we used the advertising automation platform, connected with the website & sales feedback; we would like to consider a search campaign for Life Republic - A project by Kolte Patil.
The objective of this campaigns was to generate maximum number of qualified leads at a lower CPA & increased Brand awareness. Though Adwords display and other platforms were used for the campaign, we will focus on how search and an automation platform was extensively used to bring down the CPAs.
The advertising automation platform churned out the initial media plan, the keywords split across above mentioned buckets and also generated the relevant text ads. Long tail keywords which later proved to be major contributor to leads, were also generated. With multiple ads for every ad-group, it also allowed us to run A/B tests around different ad communication, monitor the quality score and the CTR’s. Based on the recommendation engine we could optimize the ad copy and keep only the ones that were performing.
The advertising platform used Google PageSpeed insights to crawl and recommend changes on our landing page. Ensuring a minimum score before a campaign can be pushed live, ensured every change to the website adhered to the minimum threshold norm.
The platforms preemptive rules would add new keywords to the list automatically. N-gram analysis was utilized to come up with a dynamic set of negative terms to limit the wastage of ad spends.
It takes the automation platform about half a day to gauge the market and competing bids. The bidding was set such that we could achieve the required impression shares, ensure its for the most cost effective keywords, and would still guarantee us the per day velocity of quality leads.
Advertising platform suggested recommendations in the conditions based on the value of the clicks from specific keywords and campaigns.
We could easily set up event tracking on our landing page (with a pre-populated JS snippet from the platform) that was linked back to the advertising platform. Based on what content was consumed, time spent on individual pages; the recommendation engine gave us more suggestions for ad copy. The platform is also connected to Google Analytics and other user behaviour tracking software (heat maps etc.); which helped us understand the performance of our landing page.
1. Click through rates increment can be attributed to the ad relevancy and search query relevancy, keyword generation algorithm and N - Gram analysis increased the search exact match IS by 27% which helped us in making keywords extremely relevant to search queries and thus bringing down the CPC’s and in turn reducing the overall cost of achieving the leads.
2. Impression share increment can be attributed to the optimisation done by the algorithm to reach the required impression share.
3. Golden mean is the term given over here to the point beyond which the CPC’s for a click would rise steeply. Golden mean is the optimal point in the impression share at which the campaign is at its most efficient state and CPC.
4. It needs to be noted here that impression share beyond the golden mean would contribute negatively towards the campaign leading to higher CPC’s and consecutively the CPA’s. It even allowed us to decrease our search lost due to rank thereby increasing the leads.
Subscribe to Amura's Blog
Join fellow marketers and get Amura's latest marketing articles straight to your inbox.
Enter your email address below: